ML20065Q357
| ML20065Q357 | |
| Person / Time | |
|---|---|
| Site: | 05000000, Clinch River, Crane |
| Issue date: | 08/26/1981 |
| From: | GENERAL ACCOUNTING OFFICE |
| To: | |
| Shared Package | |
| ML20065Q355 | List: |
| References | |
| EMD-81-106, NUDOCS 8210270020 | |
| Download: ML20065Q357 (122) | |
Text
_
3Y ~~F E COV? 70__E7 GEN'E7A'_
Report ToThe Congress O- ~~- E UN ~~EJ S~K:ES GreaterCommitmentNeededTo SolveContinuing Problems At Three MileIsland The Nation's first major accident at a commer-cial nuclear powered electricity generating station occurred at Three Mile Island over 2 years ago, yet the resolution of the resultant problems is still subject to regulatory and fi-nancial uncertainty. Consequently, little prog-ress has been made to clean up the damaged facility or alleviate the extreme financialstress placed upon its owners.
The remedies required to resolve the continu-ing problems at Three Mile Island will require unprecedented coordination and commitment by Federal and State regulatory bodies, the electric utility industry, the financial com-munity, and the owners of the damaged facil-ity.
To safeguard against similar problems in the future, the Nuclear Regulatory Commission should develop accident recovery guidelines and ensure that increased property insurance coverage is available for nuclear facilities.
6#k g
g e
t>
n EM[M1106 AUGUST 26,1981 l
8210270020 821022 PDR ADOCK 03000537 G
(
Request for copies of GAO reports should be sent to:
U.S. General Accounting Office Document Handling and information Services Facility P.O. Box 6015 Gaithersburg, Md. 20760 Telephone (202) 2754241 The first five copies of individual reports are free Jf charge. Additional copies of bound audit reports are $3.25 each.
Additional copies of unbound report (i.e., letter reports) and most other publications are $1.00 each.
There will b& a 25% discount on all orders for l
100 or more copies mailed to a single address.
Sales orders must be prepaid on a cash, check, or money order basis. Check should be made out to the " Superintendent of Documents".
l
~ '
% h & M ax-"
- imras:st.;;;Msww3K x.-
1 t
COMPTROLLE3 CENERAL OF THE UNITED cTATE3
!/
wAssinovon o c. aesas g.]/
L B-199244 To the President of the Senate and the Speaker of the House of Representatives This report examines several key issues involving the financial status of the General Puolic Utilities Corporation, the need for and source of funding to clean up the damaged nuclear reactor at Three Mile Island, and the prospects for continued reliable electric service to Pennsylvania and New It also examines bankruptcy as a solution Jersey consumers. financial problems, and the need for to the utilities' increased property damage insurance coverage on nuclear (1) reactors and (2) an improved regulatory environment for nuclear accident recovery efforts.
We believe there is a role for the Federal Government in the accident recovery ef fort and have recommended Congressional support for a We have also Federal research and development program.
recommended that the Nuclear Regulatory Commission follow the expansion of property insurance coverage for nuclear units, by the private sector and develop guidelines to expedite any f uture accident recovery ef forts.
We undertook the review at the request of Represen-tatives Allen Ertel, James J. Howard, and Morris K. Udall, and Senators Bill Bradley, Gary Hart, John Heinz, Jennings Randolph, and Alan K. Simpson.
Several other Members have interest in this effort.
Because of this, 1
also expressed the requesting Members agreed that the report should oe addressed to the Congress as a whole.
Copies of this report are also being sent to the Director, the Secretary of Energy; and Office of Management and Budget; the Chairman, Nuclear Regulatory Commission.
i I
i Acting Compt ller General of the United States i
1 1
s
COMPTROLLER GENERAL'S GREATER COMMITMENT NEEDED TO REPORT TO THE CONGRESS SOLVE CONTINUING PROBLEMS AT THREE MILE ISLAND E1GE{T The nuclear accident at Three Mile Island (TMI) on March 28, 1979, placed a major electric utility system--the General Public Utilities Corporation (GPU)--on the brink of insolvency while faced with a multi-year, S600-million unfunded cleanup operation that must be completed under uncertain regulatory constraints.
More than 2 years after the accident, a number of important questions remain unanswered:
--Can the utility companies comprising the GPU System continue to provide reliable power to their 1.5 million customers?
--Can the utilities remain financially viable?
--What are the prospects for cleaning up the radioactive TMI-2 reactor building, and how much will it cost?
--Where will the cleanup money come from?
--What can be done to protect the financial and operational integrity of other utility companies that might suf fer similar major accidents?
At the request of eight Members of Congress, the General Accounting Office (GAO) reviewed the current and prospective status of GPU and concluded that:
--Replacement power for the TMI units is available, but future System reliability is questionable unless funds are made available to increase construction and maintenance above present restricted levels.
--The financial condition of GPU continues to deteriorate, and unless sufficient rate relief is granted to restore its financial credibility, its future as a provider of electric power is in doubt.
i EMD-81-106 Ienland AUGUST 26,1981
)
F l-
--Cleanup of TMI-2 is technologically feasible but the uncertainties surrounding the source of the estimated $600 million needed for the task and the regulatory environment in which it must be done have yet to be resolved.
--Tne expeditious cleanup of TMI-2 and the benefits that can be derived are significant enough to warrant the financial participation of several parties rather than putting the entire burden on any one entity.
--State officials in Pennsylvania and New Jersey should take the leadership role in assembling the financial assistance needed for the cleanup.
--On-site property insurance coverage needs to be increased to levels that the Nuclear Regulatory Commission (NRC) determines to be adequate if other utilities are to avoid the financial and operational stress suffered by GPU in the event of another ina3or accident.
g
--Better defined regulatory guidelines for nuclear accident recovery efforts are needed to minimize the delays and added costs that have occurred at TMI-2.
THE ACCIDENT HAS AFFECTED POWER SUPPLIES AND ALTERED SYSTEM PLANNING The TMI-2 accident and the unavailability of the undamaged TMI-l generating unit necessitated an unusually heavy reliance on purchased power to economically meet the GPU System's energy requirements.
These purchases amounted to over 12 billion kilowatt hours in 1980, nearly three times the amount purchased in 1978.
The current excess generating capacity in neighboring utility systems has enabled GPU to meet its energy require-ments to date but these short-term purchases do not enhance the System's longer-term reliability.
(See p. 8.)
The accident, and its affect on the System's financial capabilities, has adversely affected GPU's plans for providing power over the next two decades.
Pre-accident plans called for the addition of a number of new generating units I
ii w tnA m w wmr = = m m ann
l beginning in 1983 and continuing until 1992.
The loss of earnings and restricted access to capital markets resulting from the accident, coupled with a reduction in consumer demand,resulted in G dates or cancelling the projects eptirely.
Unless the rate of growth in consumer demand is less than expected, GPU will have to continue its reliance on outside power purchases to meet future energy demands icnger than expected with a potential decrease in reliability of service.
(See p. 7.)
A detailed analysis of GPU's future reliability and cost of energy was performed by the Department of Energy (DOE) staff as part of this study.
Using a base case scenario that projected GPU's current generating capacity additions and load the analysis growth forecasts through 1994, estimated the relative magnitude o incremental revenue requirements and total power purchases under varying operating conditions.
The analysis showed that the average annual re-venue requirement attributable to not restarting the undamaged TMI-l unit was nearly $421 million.
If both TMI units are not returned to service and a large firm-purchase is not completed, the average annual cost to the GPU System could If this increase by $1.1 billion per year.
occurs, GPU's dependence on neighboring utilities l
for power supplies would increase to the point where reliable service could not be assured.
If GPU were able to clean up and restart the damaged TMI-2 unit in 1986--two years earlier than currently planned--the System's annual average revenue requirements would be reduced by about $30 million.
(See pp. 9 to 11.)
influential factors affecting One of the most future power needs is the rate of change in If the rate of the consumer demand for power.
load growth were reduced from the 2.6 percentforecast the annual average revenue requirement percent,could be reduced by nearly $470 million.
(See p. 11.)
GPU'S FINANCIAL CONDITION CONTINUES TO DETERIORATE GPU's financial recovery continues to be adversely affected by the limited rate relief iii Tor smut
~
f allowed by State regulators and the millions o dollars in unrecovered costs which are being TMI expended for non-cleanup activities on the Tne long-term operation of the System l
to could be adversaly affected if GPU is unab e units.
for regain its access to capital market funding refinancing its debt and making capital improve-While GPU is aole to ootain some short-help the term loans, this arrangement can onlyThe present loan ments.
company to a limited extent.
1, 1981.
agreement is due to expire on October (See pp. 21, 22, 33 thru 35.)
hi The GPU companies are being allowed by t e r ers regulatory commissions to recover from custom about S605 million for purenased power costs.
1981 as Tnis includes current costs tnrough June defer-well as some costs that were previouslyThese measure the GPU's financial problems, however, becausecomm red.
revenues from costs by reducing the companies' during base rates by a total of S326 millionIn addition, uninsured costs borne by GPU stockholders because GPU has the same period.
t been allowed to pass them on to the ratepayers.
This has placed an additional constraint on rces.
the companies' already limited cash resou (See pp. 25 thru 28.)
i tance GPU expected to obtain some financial ass sfro As a now considered to be highly unlikely.
tinue consequence, the companies will nave to con d funds their dependence on internally-generate i cing and short-term borrowings for capital f nanIf the total capit illion needs of the companies of more than $2.7 b requirements.
serious over the next 5 years are not met, d ability questions arise regarding the continue i
power of the System to provide adequate electr ci lly sound.
to its customers and to remain financ a (See p. 34.)
FUNDING FOR TMI-2 CLEANOP MUST BE RESOLVED TO INSURE j~
GPU'S FINANCIAL VIABILITX bout GPU had spent a As of December 31, 1980, S180 million in its accident recovery effort, i
ting yet much remains to be done in decontam na iv
q,
---. m a e mm. m -
r i
l the containment building.
Four major accident recovery cost estimates, including funds pre-viously expended, were made in 1980.
The estimates ranged from $652 million to $1.3 billion--net Of $300 million in insurance pro-ceeds.
(See pp. 42 and 44.)
A cleanup cost estimate prepared in April 1981, projected that the remaining cleanup process will cost about $600 million--net of remaining insurance proceeds.
A proposed cost-sharing plan by the Governor of Pennsylvania on July 9, 1981, estimated the cleanup costs at $760 million
--including operation and maintenance expense and $90 million of insurance proceeds remaining as of January 1982.
Improvements in the regu-latory environment and the cleanup methodology could reduce these estimates, but they appear to be reasonable for current planning purposes given the present regulatory and financing uncertainties.
(See pp. 44 thru 46 and 71.)
The cost estimates for cleanup will be about the same regardless of a decision to restore or decommission TMI-2.
A final decision cannot be made, however, until the damaged fuel is removed and a closer examination of the nuclear reactor components is completed. (See pp. 47.)
GPU has budgeted about $60 million for TMI-2 expense in 1981 with about $40 million covered If this expenditure level by insurance.insurance proceeds will run out in continues, late 1983 with much of the cleanup work undone.
about To complete the cleanup as scheduled,
$100 to $150 million a year will be needed.
it is According to some investment analysts, extremely doubtful that GPU will be able to borrow the needed funds for other capital requirements, such as bond retirements, as long as the company and its stockholders continue to be solely responsible for TMI-2 cleanup costs.
The threat of bankruptcy appears to have passed, favorable but GPU's inability to renegotiate a short-term borrowing agreement in October 1981, could or refinance its maturing bonds in 1983, still trigger such an event.
Although there are too many uncertainties to specifically state what would be best for GPU, it appears that costs to GPU's consumers--and those of other v
I m sheet
g.
I utility companies as well--will be higher if GPU goes into bankruptcy.
One study has estimated that the added costs for new debt and equity could increase by $400 million annually because of the increased risks perceived by investors.
It also is not clear that bankruptcy would resolve GPU's financial problems or accelerate the cleanup of TMI-2.
(See pp. 34 and 49 thru 54.)
OPTIONS FOR FINANCING TMI-2 CLEANUP A number of options have been proposed to provide support for the TMI-2 cleanup.
While each may have some degree of acceptance, a combination of some of the more doable options would probably be the most equitable.
GAO selected six options as representative of the kinds of support that are being proposed:
--New ownership of the TMI units.
(See pp. 56 and 57.)
--A nuclear fuel enrichment surcharge.
(See p. 58)
--A mandated insurance assessment for nuclear reactors.
(See pp. 59 and 60.)
1
--Increased consumer rates, possibly supple-mented by some portion of stockholder earnings.
(See pp. 61 thru 67.)
--Federal reseatch and development assistance.
(See pp. 67 and 68.)
--Electric utility industry support.
(See pp. 69 and 70.)
l PROPERTY INSURANCE _ COVERAGE FOR THE INDUSTRY NEEDS TO BE EXPANDED An early stumbling block to the growth of the nuclear industry was its inability to obtain adequate third-party liability insurance.
The Congress took action through the Price-Anderson Act of 1957 to develop the necessary insurance coverage and thereby foster the growth of the industry.
Although both liability and property insurance coverage have grown since 1957, the TMI accident has demonstrated that the $300 million of property insurance available at the time of the TMI accident was inadequate.
vi
(
st&sh %
. x.JL \\
l Some increase in property insurance coverage has occurred since the accident at TMI, out nuclear units still remain underinsured because insurance companies have oeen reluctant to commit their resources to an industry that is perceived Different methods for providing as stagnant.
additional coverage are being explored by the utility and private insurance industries with some prospects for increasing coverage to
$1 billion.
(See pp. 81 thru 86.)
Mandatory utility self-insurance might be needed if the industry is unable to obtain the level of coverage determined to be adequate by NRC through voluntary means While there has been in a timely manner.
some favorable response to this proposal, congressional action will probably be required to give NRC authority to require (See pp. 87 and 88.)
such coverage.
Current legislative proposals would involve the Federal Government more directly in providing additional property insuranceThrough the formation of a qua coverage.
governmental insurance corporation, mandatory premiums would be collected from utility companies to cover ooth future accident losses The l
and part of the TMI-2 cleanup costs.
legislation provides that the corporation would be converted to a private mutual in-surance company at some future date.
(See pp. 59, 60, 88 and 89.)
NRC REGULATORY CHANGES NEEDED FOR l
l FUTORE ACCIDENT RECOVERY EFFORTS NRC's response to GPU's accident recovery needs was not as constructive as it might The initial priority given to have been.
its activities in responding to the accident diminished and lengthy delays in obtaining NRC approval for specific actions began to adversely af fect GPU's recovery ef forts.
The problems were compounded because NRC relied on GPU to initiate all the proposals as to how to proceed with the cleanup effort There was while NRC simply reacted to them.
little or no consideration given to the unique conditions that existed at TMI-2 and tne need for a departure from the routine way of its regulatory responsibilities.
carrying out vil Tur swet l
g o
0 NRC had approved GPU's plan for cleaning up the auxiliary building by mid-October 1979, the time of the but took 25 months from accident to make a decision on cleaning up the more highly contaminated water in the Although the proposed containment building.
technology was not new, NRC believed it had a responsibility to assess the environmental impact of the cleanup process and allow opportunity for public input into its final (See pp. 90, 93 and 94.)
l decision.
Current regulatory efforts appear to be responsive to GPU's needs but it is too early to tell whether the change will be sufficient to allow GPU to expedite the cleanup activities.
I The lessons learned from the TMI-2 experience I
should provide a good basis for a change in NRC's approach to a major accident recovery (See pp. 94 and 95.)
effort.
RECOMMENDATIONS TO THE CHAIRMAN, NUCLEAR REGULATORY COMMISSION Because another nuclear accident at an under-insured utility company could seriously affect public health and safety, GAO recommends that NRC closely follow the current efforts of the insurance and utility industries to increase insurance coverage to what it determines to GAO further recommends be an acceptable level.
1981, NRC assess that no later than December 31,This assessment should the progress being made.
include an evaluation of the insurance available in the private sector and a determination as to whether a mandated insurance coverage program is necessary.
To mitigate future regulatory constraints on nuclear accident cleanup activities, GAO recommends that NRC establish a set of guidelines that would facilitate the development of recovery procedures by utility companies in the event of other nuclear reactor accidents.
RECOMMENDATION TO THE SECRETARY OF ENERGY To assure the availability of funding needed to complete an expanded research and develop-ment program at TMI, GAO recommends that DOE prepare a multi-year budget proposal for Federal viii
- A hAIMJ4.dedgfgdyg((
?
5~ ",
h D'TQ4WiWppmg ggs v
participation in the TMI cleanup effort and The budget pro-present it to the Congress.
posal should recognize the primary leadership role of State officials in working with GPU and the industry in the cleanup effort and within that parameter should clearly specify the objectives to be achieved by the Federal the work steps required in each involvement, the application of the program fiscal year, and the total funding needed to
- results, successfully meet the research and development objectives.
RECOMMENDATIONS TO THE CONGRESS _
Given past congressional support for the commercial development of nuclear power, the continued Federal regulatory oversight of nuclear reactor operations and the need and radioactive waste disposal, to reduce the economic burden imposed by the TMI accident as much as possible, GAO recommends that the Congress provide the required multi-year funding to DOE for its research and development program at TMI.
GAO further recommends that the Congress closely follow the current ef forts to resolve the funding problems for the TMI-2 cleanup through State and utility industry financing and DOE's research and development program.
If these State-led ef forts are not successful, GAO recommends that the Congress devise a l
f mechanism which would serve to obtain the required financial assistance to complete the TMI-2 cleanup.
AGENCY COMMENTS AND OUR EVALUATION GAO provided a draft copy of its report to NRC and both agencies responded and DOE for review, (See app. I and II.)
with comments.
The Nuclear Regulatory Commission agreed with GAO on the need to increase insurance coverage and stated that the NRC staff will keep abreast of the two current proposals outlined in the I
NRC pointed out that it has a proposed report.
if approved as a for comment that, rule out would require power reactor licensees final rule,
(
i l
iX
_ Tear Shatl
i i
to provide the maximum amount of property insurance available.
NRC did not comment on our recommendation that it determine an acceptabir level of insurance coverage.
GAO believes this to be a critical element in monitoring industry efforts to increase insurance coverage since the maximum amount of insurance available may not be sufficient to cover the full costs of an accident recovery effort.
NRC did not disagree with the GAO recommendation that it develop accident recovery guidelines but suggested that additional clarification was needed as to what the guidelines should include.
Accordingly, GAO expanded its previous discussion of this need by defining some of the matters that might be covered in the guidelines that would be useful in developing acceptable accident recovery procedures in the minimum amount of time.
(See p. 97.)
The Department of Energy disagreed that a multi-year funding proposal for its proposed 3-year research and development program is necessary.
DOE believes that the normal annual review and Congressional authorization and appropriation processes will assure the program's consistency with DOE's objectives and the cleanup needs.
GAO believes, however, that a commitment of l
Federal sector support for the TMI-2 cleanup is extremely important in eliciting the support of other interested parties and that such support can best be expressed through an approved financial commitment for the entire effort rather than simply a multi-year plan with no total funding commitment to insure its successful completion.
l X
'N
<* h u ij n ! Lar aqwfgg3 *Q g
}
Contents Page DIGEST i
CHAPTER 1
INTRODUCTION 1
Overview of GPU 1
Agencies with regulatory responsibility for GPU 2
Our other related work 4
Objective, scope and methodology 4
2 GPU'S CURRENT POWER SUPPLIES ARE ADEQUATE BUT LONG-TEhM RELIABILITY IS QUESTIONABLE AND MORE COSTLY 6
GPU's system planning has changed since the accident 6
GPU service has become more costly and less reliable 8
Changes in load growth and capacity additions affect GPU system 9
Adequacy and reliability of GPU's transmission network 17 Conclusions 18 3
GPU'S FINANCIAL CONDITION CONTINUES TO DETERIORATE 20 Comparative income and earnings data for the GPU system hignlignts its problems 21 Current rate orders have only partially alleviated cash flow problems 24 GPU has long-term financing needs that must De met 29
[
Financing difficulties may affect completion of future construction projects 33 Conclusions 35 4
FUTURE FINANCIAL VIABILITY OF GPU MAY BE CONTINGENT ON RESOLUTION OF FUNDING FOR TMI-2 CLEANUP COSTS 40 f
Current cleanup cost estimates are i
still being refined 40 GAO assessment of cleanup cost estimates 41 Estimated cost could vary given differing circumstances 45 l
~
&cga CHAPTER Cleanup coat cavings will bo minimal I
47 if TMI-2 is decommissioned GPU is unable to fund the cleanup cost without jecpardizing its 47 financial viability Bankruptcy is a questionable way to resolve GPU's financial problems 49 55 Conclusions SEVERAL OPTIONS EXIST FOR FINANCING THE 5
56 TMI-2 CLEANUP COSTS 56 New ownership of TMI units 58 Nuclear fuel enrichment surcharge 59 Mandated insurance ascessment GPU funding from increased rate revenues and stockholder earnings 60 Federal research and development 66 assistance 68 Industry contribution 69 Conclusions Recommendation to the Secretary of 73 Energy 73 Recommendations to the Congress 73 Agency comments and our evaluation CHANGES ARE NEEDED TO FACILITATE FUTURE 6
75 ACCIDENT RECOVERY EFFORTS Increased utility property insurance 76 coverage is needed NRC needs to establish accident 89 recovery procedures 95 Conclusions Recommendations to the Chairman, 97 Nuclear Regulatory Commission 98 Agency comments and our evaluation APPENDIX I
Letter dated August 7, 1981, from the Executive Director for Operations, 99 Nuclear Regulatory Commission II Letter dated August 7, 1981, from the Assistant Secretary of Management and 102 Administration, Department of Energy 105 Definition of Moody's rating symbols III 106 IV Bibliography of bankruptcy studies 1
W h p ay_,+Lw<>ygf g
-a
'r y
l ABBREVIATIONS ANI American Nuclear Insurers APIC American Power Insurance Corporation DOE Department of Energy EEI Edison Electric Institute ERA Economic Regulatory Administration FERC Federal Energy Regulatory Commission GAO General Accounting Office GPU General Public Utilities Corporation GwH gigawatt hour kWh kilowatt hour MAELU Mutual Atomic Energy Liability Underwriters MAERP Mutual Atomic Energy Reinsurance Pool Met Ed Metropolitan Edison Company MW megawatt NEIL Nuclear Electric Insurance Limited NJBPU New Jersey Board of Public Utilities NML Nuclear Mutual Limited NRC Nuclear Regulatory Commission O&M Operations and Maintenance l
PaPUC Pennsylvania Puolic Utilities Commission PEIS Programmatic Environmental Impact Statement Penelec Pennsylvania Electric Company PJM Pennsylvania-New Jersey-Maryland Interconnection RCA Revolving Credit Agreement SDS Submerged Demineralizer System SEC Securities and Excnange Commission TMI Three Mile Island L
~-
m=mm
--am mmmm an
n
MMMEnww _
1 CHAPTER 1 INTRODUCTION eight Congressman jointly In a December 12, 1980, letter, future requested that we analyze several issues related to therole of th provider of electric power in Pennsylvania and New Jersey.
We were asked to respond to a number of questions relating ile the costs of cleaning up the nuclear-powered Three M Island (TMI) generating unit; (2) the financial status of to (1)
(3) reorganization alter-GPU and its subsidiary companies;the effects of the accident on the natives for GPU; (4) and (5) alter-companies, their shareholders, and consumers; natives available to the Federal government in the event one or more companies become insolvent.
OVERVIEW OF GPU GPU is an electric utility holding company owning the outstanding common stock of its three operating companies:
Jersey Central Power and Light Company in New Jersey, theand Pennsylvania Electri (Met Ed),
Metropolitan Edison Company (Penelec) in Pennsylvania.
GPU's investment in the
$1.4 billion, Company common stock of the three companies is about or 28 percent of the $5 billion in total assets.
GPU issues its own Under normal operating conditions, common stock to the public on which it pays dividends from and some small unsecured The operating company dividends, represent all of GPU's cash lines of short-term credit, resources since GPU is generally prohibited from issuing long-The operating companies receive capital term debt securities.
contributions from the parent and obtain other capital by issuing long-term debt securities and preferred stock.
The GPU System's normal operating methods were severely 1979, and a affected by the accident at TMI-2 on March 28, order to keep subsequent Nuclear Regulatory Commission (NRC) shut down until mand the undamaged TMI-1 1/
The loss of these two units made and certified by NRC.resulted in greatly expanded replacement power p The utility companies needs.
economically meet consumers' allowed to recover these purchased power costs were not 1 had been down for refueling and was ready to restart 1/ Unit on the day of the accident.
1
rates for some time after the accident and they had to be paid for through short-term bank borrowings.
All costs associated with TMI-2 that are normally collected through customer rates were disallowed almost immediately following the accident.
Similar costs have been disallowed for TMI-l since the second quarter of 1980.
In addition, the companies have had to pay from their earnings all costs to date for NRC-mandated changes to TMI-l and for the non-insured cleanup items for TMI-2.
As a consequence of the financial drain on their resources, only Penelec has made dividend payments to GPU 'on its common stock.
GPU has not made dividend payments for six successive quarters, has no market for its common stock, and none of its companies can sell bonds or preferred stock.
AGENCIES WITH REGULATORY RESPONSIBILITY FOR GPU Three Federal agencies and the Pennsylvania and New Jersey public utility commissions exercise jurisdiction over various segments of GPU System acti/ity.
GPU's e f forts to restart TMI-1, proceed with the cleanup of TMI-2, and remain financially viable have all been particularly affected by the regulatory controls exercised by these entities.
The Nuclear Regulatory Commission NRC is responsible for licensing and regulating activities at nuclear facilities, including TMI-l and 2, I
l under the Atomic Energy Act of 1954, as amended, and the Energy Reorganization Act of 1974, as amended.
This re-sponsibility includes providing reasonable assurance that the use of nuclear reactors does not result in undue risks to the health and safety of the public.
In accordance with this responsibility, NRC is conducting restart hearings for TMI-l and approves and monitors all cleanup activities at TMI-2.
NRC is also responsible for establishing specific waste storage and/or disposal criteria and regulations, consistent with the Environmental Protection Agency's criteria and general environmental standards, and for licensing and regulat-ing long-term, high-level waste storage or disposal facilities.
The Department of Energy The Department of Energy (DOE), in consultation with the Federal Energy Regulatory Commission (FERC), has the respon-sibility for assuring the reliability of electric bulk power supply throughout the United States.
The basic authority for Federal regulation of electric utility companies comes from the Federal Power Act of 1935.
The DOE Organization Act of 1977 (P.L. 95-91) divided the responsibilities held by the Federal Power Commission until September 30, 1977, between the Secretary of Energy and FERC.
The Secretary in turn delegated to the t
2
~M=AEWW R c711%Gt8an.QsiMk j
NIY[I"WDShrtsmwm m._
w responsibility for Economic Regulatory Administration (ERA)
FERC has Jurisdiction assuring the adequacy of bulk power supply.over the filed tariffs for i
It also power and approval of wholesale rates for electricity. interstate transmission has jurisdiction over facility agreements, rates, and capacity and ene power pools.
DOE was given additional authority in the electric power area by the Public Utilities Regulatory Policies Act (P.L.95-617). provid ERA was empowered to (1)
(2) and support FERC on ratemaking and cost of service m (3) l national energy policy issues, supplies and reliability, and (4) monitor State reviews of various rate structures and standards.
In addition to these general responsibilities related to i the electric utility industry, DOE is responsiole for develop ng disposal waste disposal methods and for long-term storage and/or l pro-of botn Federal and commercial high-level wastes and Federa gram transuranic contaminated waste.
The Securities and Exchange Commission administers The Securities and Exchange Commission (SEC)1935 (15 U.S.C.
the Public Utility Holding Company Act ofThe purpose of the Holding Com investors, and consumers from abuses 79, et seq.).
to protect the public, associated with the control of electric utility companies In part, it is a by use of the holding company device.
specialized antitrust statute with the objective of reor-ganizing and constraining the operations of utility holding d
companies, and a regulatory statute providing for continue surveillance of the corporate structure, financial trans-actions, and operational practices of public utility holding company systems.
State public utility commissions The Pennsylvania Public Utility Commission (PaPUC) have and the New Jersey Board of Public Utilities (NJBPU) k l
the GPU operating companies.
State regulators may review a utility's expenses, set tneand amount of revenues the utility will be able to collect, I
determine the allowable rate of return it can earn on itsthe regulators deter-Through these mechanisms, investments.
mine tne amount of profit a company can make.
3
OUR OTHER RELATED WORK We have issued several reports closely related to the questions addressed by this report.
Our report, "Three Mile Island:
The Financial Fallout" (EMD-80-89, July 7, 1980) provided the basis for the congressional request that ini-tiated this assignment.
In that report, we examined the financial status of GPU and the problems f acing the utilities that needed resolution.
We recommended that NRC expedite the restart hearings on TMI-l and that DOE continue the assessment of GPU started by us and report to the Congress on the need for any external assistance.
Our report, "The Nation's Nuclear Waste--Proposal for Organization and Siting" (EMD-79-77, June 21, 1979) discussed tne failure of the Federal Government to develop a publicly acceptable nuclear waste disposal system.
A letter report, " Analysis of the Price-Anderson Act" (EMD-80-80, Aug. 18, 1980) dis-cussed the need to reassess the provisions of the Price-Anderson Act 1/ as they relate to liability insurance pro-tection afforded the puolic and the nuclear industry in the event of a nuclear accident.
OBJECTIVE, SCOPE AND METHODOLOGY The broad range of questions raised by the congressional requestors required an analysis of the current situation at TMI and GPU as well as some assessment of what the next few years hold for the companies and their customers.
The time period covered in our analysis goes back to mid-1980, and in some instances, extends forward to the year 2009.
Although each of the three operating companies func-tions as an independent utility, much of the administration, technical support, and documentation for their operations are maintained at the GPU headquarters at Parsippany, New.
Jersey.
Consequently, almost all of our work with the companies was done at that location.
We held numerous meetings with corporation officials, obtained and analyzed documents, reports, studies, rate filings, demand and generating capacity forecasts, and related data.
We also developed the kilowatt nour (kWh) costs that would oe needed to regain and maintain some measure of financial viability.
1/The act was passed by the Congress in 1957 and is in section 170 of the Atomic Energy Act of 1954.
It provides for insurance coverage of up to $560 million for of f-site personal and property damage claims resulting from a nuclear accident.
4 m.e 4
- a.. -
- s. q;g.g.
' ;~ >
We also visited the TMI plant site, and discussed re-start and cleanup issues with responsible GPU and Met Ed officials, contractor representatives, and on-site repre-sentatives of NRC and DOE.
Discussions were held with the president of the NJBPU and the chairwoman of the PaPUC and their staf fs concerning GPU's financial problems and the role of the State commissions.
We met with the key staff person in the Pennsylvania Governor's office and with Pennsylvania tax officials on the GPU insol-vency issue.
We contacted key NRC officials responsible for both the TMI-l restart and TMI-2 cleanup.
Copies of pertinent NRC documents were obtained and analyzed.
We also met with DOE's Nuclear Energy staf f and obtained information on their proposed participation in the TMI-2 cleanup effort.
us in making an independent assessment of f
To assist the reasonableness of GPU's proposed construction program we arranged with DOE for the over the next few years, necessary engineering staf f to develop computer simulations of the GPU system using 19 mutually agreed on scenarios of facility construction and load growth.
The simulations were j
DOE run on the GPU computer using our own assumptions.
technical staff assisted us by analyzing the results and discussing them with us and GPU.
Load flow analyses of the transmission system were also provided. DOE staf f made revenue requirement computations for us based on the con-struction activities simulated in the model, and we reviewed these computations for reasonableness.
During the course of the audit, we met with officials of the banks holding the short-term loan notes for GPU, bond trustees, investment firms, private consulting firms, engineering firms involved in nuclear plant con-struction, other utility companies, and insurance companies.
Each of the officials contacted were considered to be experts in their field, and they shared with us their perceptions and/or the results of studies or analyses done on the issues included in our assignment.
limited our scope of work in several areas--scenario We and analyses, utility bankruptcy and reorganization issues, options for funding TMI-2 cleanup costs.
An explaration of these limitations is provided in the body of the report '
where applicable.
5 1
CHAPTER 2_
GPU'S CURRENT POWER SUPPLIES ARE 1
ADEQUATE BUT LONG-TERM RELIABILITY IS QUESTIONABLE AND MORE COSTLY The accident at TMI-2 and the lengthy delays encountered in restarting the undamaged TMI-1 generating unit have adversely affected the companies' pre-accident plans for maintaining of the next 2 decades.
The I
system reliability through mors loss of the TMI generating capacity has necessitated an un-usually heavy reliance on purchased power to economicallyThe aoandonment of the meet the System's energy requirements.
Forked River facility and slippages in other planned con-struction projects will continue this trend.
Although GPU was carrying the maximum property damagethe Syst insurance available, in its future planning because of the financial Durdens resu The financial aftermath of the accident is one of tne major factors contributing to the cancellation and deferment of of TMI-1.
This ongoing construction of needed future capacity additions. situa provided to GPU's customers and cause the power provided to customers to be more expensive.
The System's ability to continue providing reliable, economical power to its customers is strongly influenced Dy how fast consumer demand for electricity grows over the 1981-94 study period.
In fact, the rate of load growth has more influence on power costs and system reliability than other more obvious factors such as fuel prices andOther factors influ construction delays.
the return supplies of reliable, economic power are (1)the continued avail-to service of the undamaged TMI-1, (2) the ability of external firm power purchases, and (3) maintenance of a strong transmission network.
GPU's SYSTEM PLANNING HAS CHANGED SINCE THE ACCIDENT Prior to the TMI-2 accident, GPU ranked as the 14th The total invest-investor-owned electric utility.
in the System was about SS Dillion, and it collected largest The GPU System was ment about $1.3 billion in annual revenues.
experiencing an increasing growth in electrical demand prior Electricity sales had grown about 4 percent to the accident.
annually since 1976, and an ambitious construction program 6
mgy mg4 (w
3 muy gg.,
4
'~
~
N nad been initiated to meet the anticipated future customer In addition to the 8,281 megawatts (MW) of winter demands.
generating capability already installed on the GPU System, the utility planned to bring at least 6 new major projects These on line beginning in 1983 and continuing until 1992.
additions would have increased tne net winter generating capacity of the GPU System to about 10,952 MW.
I Following the accident at TMI-2, GPU significantly
[
curtailed its plans for future generating capacity additions.
This curtailment was caused by a number of factors including t
a reduction in consumer demand for electricity consistent i
(1) with national trends, (2) increased financial ooligations resulting from TMI-2 cleanup, (3) decreased access to financial markets, and (4) loss of revenues caused by removing the TMI rate bases.
The curtailed activity units from the companies' involved deferring new project completion dates for several The proposed 625 years, or cancelling the projects entirely.
MN Seward No. 7 coal plant, for example, was anticipated to Due to financial uncertainties, be in service by December 1985.
reduced energy cemands, and regulatory delays, GPU has deferred the completion of this unit until 1989.
The following table compares the current planning for the GPU System with the plans in effect immediately prior to the accident at TMI-2.
Table 1 Schedule of Slippages and Deferrals on GPU System Before and After TMI-2 Accident In-service date prior Current in-Months service date_ deferral Unit Type Capacity to accident _
Forked River Nuclear 1,120 M 1983 Canceled Seward No. 7 Coal 625 MW 1985 1989 48 0080 Coal 625 m 1988 1991 36 Undesignated Coal 625 MN 1989 1993 48 Undesignated Coal 625 MW 1990 1994 48 Undesignated Coal 625 Md 1992 1995 36 Un$esignated Coal 625 W 1994 1996 24 Undesignated Coal 625 MW 1996 1997 12 Pcmped Storage Hydro 850 M 1991 1994 36 Pumped Storage Hydro 1,000 MW 1993 Canceled 1998 Undesignated Coal 625 W 2000 Undesignated Coal 625 MW 1985 Ontario Hydro gPurchase 1,000 M gAlthoughthiscapacityadditionhadbeententativelyconsidered prior to the accident, considerations were formalized as a result of the accident.
7
e GPO SERVICE HAS BECOME MORE COSTLY AND LESS RELIABLE The accident at TMI-2 not only nad a significant impact I
on the financial integrity of the GPU System, but also placed it under stress from a technical standpoint as well.
The h
outage of the two units at TMI reduced System capacity by 1,656 MW, or about 21 percent of total net capacity.
To offset the loss of this source of relatively inexpen-sive energy, large amounts of replacement power have been pur-chased at costs ranging f rom $20 to $25 million a month.
Power purchases in 1980 totaled over 12 billion kWh--nearly three times the amount purchased in 1978.
The GPU System currently has short-term purchase contracts for over 1,600 MW of capacity and associated energy with utilities outside its Because of the current excess generating capacity in System.
neighooring utility systems, as much as 1,200 MW in additional capacity may be obtained through these purchase contracts.
In planning for future System requirements, GPU is cur-rently negotiating for the purchase of l',000 MW of capacity and associated energy under either a fita 10-year contract with a Canadian power supplier or with other potential long-term power suppliers.
If the negotiations are successful, facilities may have to be constructed before tne power can De brought into the GPU system.
Under the proposed schedule for the Canadian project, this additional source of power could be available by January,1985.
Other measures affecting reliability--in addition to the power purchases--have been required as a result of the loss of the TMI units and cancellations and delays of proposed generating facilities.
Because of continuing cash constraints, a program designed to curtail construction and maintenance expenditures has been instituted at each of the companies.
A separate austerity program reflectingThe reduced budget levels has been instituted for Met Ed.
program is based upon meeting minimum System needs, and according to company officials, provides less than prudent levels of cash outlays necessary to maintain System reli-ability, provide acceptable emergency response, and serve the economic and social interests of Met Ed's service territory.
Tne curtailment in expenditures covers a broad range of System functions, from reductions in expenditures for opera-tions and maintenance in generation, transmission, and distri-Since technical bution to restricting maintenance at TMI-2.
operations have been curtailed, other reductions in Met Ed's workforce have occurred which may affect service to customers.
Personnel reductions have touched even the most routine 8
L w--mmwswmpagg
service activities such as tree trimming.
For the period 1976 through 1980, this item alone has resulted in a 12-fold increase in the average customer outage time due to tree-related causes.
Tne austerity program has also affected new customer nookups, causing them to take longer than normal.
These changes, while immediately visible throughout the GPU service area, will not probably become more apparent as the austerity program continues, and could cause future service to be somewhat more expensive and less reliable.
CHANGES _IN LOAD GROWTH AND CAPACITY ADDITIONS AFFECT GPU SYSTEM DOE staff performed a detailed analysis of the future re-liability and cost of power for the GPU System as part of our study.
Tne approach used by DOE, and agreed to by us, developed a base case scenario using the currently planned GPU capacity-addition schedule and load growth forecasts through 1994.
The base case assumes a financially-sound GPU System.
Eighteen additional scenarios were developed by varying the base case capacity additions and load growth assumptions.
For each of the 19 scenarios, GPU's incremental revenue requirements were calculated and the overall System reliability was assessed.
DOE and GAO selected the scenarios to indicate the possible operating conditions and revenue requirements for the System under differing assumptions regarding fluctuations in the consumer demand, variations in the planning and scheduling of additional generating capacity, and tne availability of both TMI generating units.
A major focus of the study was on the economic value of Since many of the capital the TMI units to the GPU System.
projects planned by GPU will have economic lives beyond 1994, and the operating licenses for both units are scheduled to expire in 2009, we calculated GPU's incremental revenue requirements--costs associated with building new facilities and providing power--to 2009.
We did this to better illustrate the useful value of the TMI units to the System and allow comparisons between various System configurations over the economic lives of the projects considered.
The incremental revenue requirements shown in our scenario analyses, therefore, reflect the value of the production facilities installed between 1981 and 1994 but operated through the year 2009.
For comparability, all costs in the scenario analyses were "levelized" to reflect the magnitude of change from the base Levelizing illustrates what the average annual cost would case.
be--taking into account the time value of money--if the cumulative present value of revenue requirements were spread evenly over The actual amount of revenues each year of the life of the asset.
9
l collected through rates each year will vary from the levelized figure because near-term annual cost increases are small, with larger increases occurring in the latter years of the study period due to the escalating cost of doing business.
GPU is a member of the Pennsylvania-New Jersey-Maryland PJM's purpose is to provide, througn Interconnection (PJM).
contractual agreement among the members, the service, reli-ability, and economy that would result if PJM were one company In daily while recognizing individual company constraints.
operation, tne amount of interchange power flowing Detween GPU and the rest of PJM is a function of several factors on both and avail-systems such as system demand, fuel type and cost,The reliability of the GPU ability of generating capacity. electric system is a function of the reliability main throughout the entire PJM generating and transmission systems.
However, an indication of GPU's System reliability may be provided oy analyzing its generating capacity reserve margins and the amount of electric power GPU sells to and buys from PJM (net interchange).
Our analysis of GPU's current and projected reserve margins under the various capacity additions and load growth scenarios showed a relatively consistent relationship between reserve margin levels--which normally indicate a level of systemreliability--
We noted that the quantity of energy interchanged tended to Therefore, in increase as GPU's reserve margins decreased.
our scenario analyses, we have used the energy interchange levels as an indicator of the re.lative reliaoility of the GPU System.
GPU's planned System configuration and scenario modifications The base case, used to determine the effects of changes in future System configurations, was developed from the latest available load growth and capacity addition forecasts published The forecasts show a compound annual in October 1980, by GPU.
load growth rate of 2.6 percent over the period 1981 to 1994.
In order to meet this projected demand for electricity, the company anticipates returning TMI-1 to service in 1982 and TMI-2 to service by 1988 as well as relying on firm power Although some small capacity additions purchase arrangements.are anticipated by 1985, no new major generating units expected to be in service before 1989.however, GPU sees a need to plac 10
, sw&;
ggg N
wg
w generating units and one 850 MW pumped-storage hydroelectric facility in service.
Other capacity additions to the System Table 2 shows the are expected, but these occur beyond 1994.
base case assumptions as well as the variations to the case case that were used in this analysis to assess potential System reliability and cost impacts.
Change in consumer demand and capacity additions affect cost of power and reliability Based on an analysis of the various case scenarios, changes in the rate of load growth, i.e., consumer demand for electric energy, greatly influence the cost of power (or revenue require-and reliability of tne GPU System.
As demand increases, ments) more costly sources of energy must be tapped, whether they are the utilities' own less efficient generation or higher-cost A decrease'in demand from purchases from other utilities.
the present forecasted levels permits the load to be met with more efficient, less expensive GPU generation or outside pur-Given that construction activities planned by other chases.
PJM utilities proceed as scheduled, interchange transactions could produce some cost savings which would be passed through to the consumer.
As an example, the scenario analysis indicates that if the GPU System growth rate were reduced from the currently the levelized revenue projected 2.6 percent to 1.6 percent, I
requirements could be reduced by an average of about $469 million per year in 1981 dollars as compared to projected Conversely, a 1/2-percent increase costs in the base case.
in the load growth rate to 3.1 percent could raise the annual revenue requirements by an average of about $311 million.
l l
These projected savings and increases in power costs would l
occur during the period 1981 through 2009.
Variations in consumer demand have a similar effect on the As indicated earlier, projected reliability of the System.the amount of interchange power can ne used a the relative System reliability because for GPU, decreased purchases generally indicate the utility is satisfying more of its consumers' demands with its own economical generating units.
Reduced levels of interchange tend to imply a more reliable system.
For example, the 41,329 gigawatt hours (GwH) 1/ of interchange power in the base case would be reduced to 4,248 GwH if the rate of growth in consumer demand were reduced by i
Conversely, an increase of 0.5 percent in the 1 percent.
growth rate of annual consumer demand would require 61,360 GwH l
I of electric power interchange.
1 1/One gigawatt hour equals 1,000 megawatt hours or 1 million kilowatt hours.
l 11
/
Table 2 t>scr aptson of Scenaraos For the General Publsc Utalatses System
'l 880 Pld mtar ao load Forced
'!N !- 2 625 PE 625 PW pJM growth outage Fuel Cost of hr t tor Hyfro Case rate rates se nces capatal 1N1-1 Pa(Me purc% se kaystown 1NI-2 Seward 7 coal _
coal I cnal 2 Peakang aMaticms n-1 Base Base Base 14.254 1982 1982 1985 1985 1988 1989 1991 1993 1994 Base y
c 2
It Hagher 3
1/24 Hagher 4
1983 1983 1986 1989 1989 1992 1994 1995 1983 1983 1986 1989 1989 1992 1994 1995 5
19 tower 6
1/24 Hagher 1983 1983 1986 1989 1990 1992 1994 1995 7
u returned 1991
(
Not 1991 h
8 18 lower returned 9
1/2n hagter Not i
returned 1991 t;
Canceled 1
IJ H
lo mt
(&
11 returned 12 Canceled Not r
i returned 13 Not Canceled Not g.
h returned returned Y
~
14
+5%
1986 g//-
15
-54 16
-104 17 1966 Q
h-[$
18
+5%
teley rsaclear 1A Hagh cost V
for cleanup Case I as consadered the base case. Capacaty aMatsons and load
- p i.
mtes:
1.
growth are based on tre GU load and Capacaty Forecast dated October 17, 1980.
Cases 2 through 18 arm? 1A show generatang capacaty plans, loads,
('j 2.
forced outaje rates and fuel praces only as they dif fer from the Fealang across the pape for a gaven case wall show base case.
all assanptaons Witch have been changed from the base case (Case 1).
ll pt
W,._ 5.:l w w w G W kCukran,
' ~~~~
Cnanging the presently planned System configuration by delaying or cancelling planned generating capacity additions affects power costs amd reliability in a manner similar to that Cancelling produced by changing load growth rate assumptions.
or delaying capacity additions increases GPU's own cost of producing power and requires greater reliance on power purchases from other utilities to economically meet customer needs in Delaying,all expected capacity additions in its service area.
the base case by 1 year, for example, would increase annual revenue requirements by $81 million and increase interchange power by about 42 percent or 17,500 GwH over the base case.
The same delays, coupled with the assumption of increased load however, can quadruple the incremental annual revenue
- growth, requirements, and cause an additional 33 percent increase in interchange power.
Table 3 shows the impact on revenue requirements and interchange energy of changing the load growth rate and capacity additions on the GPU System.
t TMI restart decisions will affect System costs and reliability The base case assumptions shown in table 2 on page 9 include the restart of TMI-l in January 1982, and the return to service of TMI-2 in 1988.
The following scenario analyses demonstrates the value of restarting these units to the GPU System and its customers.
TMI-l restart GPU never envisioned a 3-year outage of TMI-l when the unit was taken out of service for its annual refueling on Although scheduled for power generation on February 16, 1979.
April 2, 1979, TMI-l was kept out of service, voluntarily by GPU, and later by NRC orders, following the TMI-2 accident.
to oe restarted later in 1979, GPU initially expected the unit but numerous NRC orders requiring certain technical and operat-(
and protracted ing changes, safety improvements initiated oy GPU, issues have kept the unit unavaila public hearings on restartThere are still a numoer of technical requirements for service.
that must be successfully completed before NRC can authorize re-However, GPU now expects a favoracle NRC decision on restart by October 1981, with the unit returning to commercial start.
operation by early 1982.
Because of the expected return to service of the unit I
a scenario deviating from the base case only in early 1982, by the assumption that TMI-l would never be restarted wasHowever, by examini not specifically modeled.ence between System configurations 10 and 12 shown in l
13
9 IIl Table 3 Changes From Base Case _ Revenue Requirements
)
and ' Ictal Interchange _ Energy Resul_ing From t
Varying Load Growth Rates and/or Revising System Configuration Annual Incremental Total net compound load Annual Revenue interchange System configuration growth rate Requirements energy (note a)
(percent) b/($, millions)
(GwH) 41,329 1.
Base case 2.6 2.
Unchanged 1.6 (468.7) 4,248 3.
Unchanged 3.1 311.3 61,360 4.
Capacity additions delayed 1 year 2.6 80.6 58,883 5.
Capacity additions delayed 1 year 3.1 382.6 78,789 6.
mI-2 replaced with 880- m coal unit 2.6 191.9 53,980 7.
TMI-2 replaced with 880- m coal unit 1.6 (305.3) 16,582 8.
M I-2 replaced with 880- m coal unit 3.1 494.5 68,999 9.
M I-2 not returned I
to service 2.6 397.7 72,801
- 10. MI-2 not returned to service and no Ontario-Hydro purchase 2.6 658.5 118,118
- 11. MI-2 returned to service in 1986 2.6 (29.7) 33,123
- 12. MI-l&2 not returned to service, no Ontario-Hydro purchase 2.6 1,079.4 172,635
- 13. MI-l not returned to service 2.6 420.8 More than 118,118 but less than 172,635 3 he sum of GPU's annual purchases less sales for the period 1981-94.
/T b/1981 dollars.
(
14 t
^
page 14, we can approximate the cost to the System of not restart-ing TMI-l (Item 13, page 14).
Under this scenario, TMI-l is retired by 1985 with no capacity replacement.
As shown below, this would increase the levelized revenue requirements by nearly $421 million per year (in 1981 dollars) as compared to the base case assumption of a January 1982 restart.
Annual levelized incremental Case description revenue requirements (millions)
Base case without TMI-1, TMI-2, or Ontario-Hydro
$1,079 Base case without THI-2 and Ontario-Hydro (658)
Increased annual cost attributable to not restarting TMI-1 S
421 Assuming the resultant costs of permanently closing down TMI-l were to be shared in the same proportion as the companies are now buying replacement energy, the individual companies' shares of the $421 million would be as follows:
Current share of Estimated annual snare replacement energy of incremental costs (percent)
(millions) l l
Met Ed 45.4
$191 f
Penelec 18.2 77 Jersey Central 36.4 153 Total 100.0 S421 I
in annual levelized revenue requirements The increase of $421 million is only meaningful when compared to the base case and other scenarios.
Actual annual revenue requirements l
would differ from year to year and could be increased or decreased by changes in the load growth rate, the construction i
of a replacement generating plant, inflation, and by utility commission decisions on how amortization and decommissioning costs of the retired unit would be recovered.
15
TMI-2 restoration The return to service of the damaged TMI-2 unit by 1988, as assumed in the base case, is much more uncertain than the date of restarting TMI-1.
Consequently, several scenarios involving TMI-2 were developed to assess the probaole conse-quencms of decisions that might ce made regarding its future use.
In the following sections, tnese scenarios are com-pared to the base case and to each other in order to assess the cost and reliability impact of various dispositions of TMI-2.
Comparison A (TMI-2 out, and TMI-2 back in prior to 1988)
This comparison evaluates the permanent retirement of TMI-2 relative to restoring it to service in 1986, two years earlier than planned.
If TMI-2 is abandoned and decommissioned, and the other base case elements remain as projected, GPU's levelized revenue requirements could increase by about $398 million per year over the. base case for the period 1981-2009.
In addition, GPU would have to increase interchange energy purchases by 31,472 GwH over the base case.
The completion of the Ontario-Hydro project in 1985 as scheduled assumes added significance if TMI-2 is decommissioned.
If the Ontario-Hydro project is cancelled, considering that TMI-2 is permanently retired, annual revenue requirements are projected to increase by S658 million over the base case for the period 1981-2009 and interchange energy purchases would increase by 76,789 GwH for the 1981-94 period.
The early restart of TMI-2 in 1986, on the other hand, could reduce GPU's costs and enhance System reliability compared to the base case.
If IMI-l is restarted in 1982, and the Ontario-Hydro project is completed in 1985, a 1986 TMI-2 restart date could result in a decrease in annual revenue requirements of $29.7 million from the base case.
Furthermore, GPU's reliance on interchange power would be reduced by almost 8,000 GwH to about 33,100 GwH for the period 1981-94.
Comparison B (TMI-2 retired but replaced with a coal unit)
This comparison evaluates the base case with a System config-utation in which TMI-2 is decommissioned and its capacity is replaced with an 880-MW coal-fired unit in 1991.
With an annual load growth rate of 2.6 percent, levelized revenue requirements would increase by about S192 million and interchange energy purchases would increase by about 13,000 GwH when compared to a
tne base case.
16
Tne dramatic effect of shifts in load growtn rates on System costs and reliability is particularly noticeaole in this comparison.
Reducing the annual load growth rate by 1 percent and replacing TMI-2 with a coal-fired generating unit decreases GPU's annual revenue requirements by over S305 million as compared to tne Tne reduced consumer demand would also result in base case.
reducing interchange energy purchases by 24,747 Gwd compared to the base case and 37,398 GwH when compared to the 2.6-percent growth rate scenario.
If, however, the load growth rate increases by only 0.5 percent per year, the annual revenue requirements would increase from S192 million to $495 million and interchange energy purchases would increase by 15,019 GwH when compared to the 2.6-percent load growth scenario.
Comparison C (TMI-1, TMI-2, and Ontario-Hydro out)
This comparison evaluates the GPU System configuration over the next 14 years with no TMI units in service, no Ontario-and an annual compound load growth rate of 2.6 Hydro project, Under these assumed conditions, annual revenue require-percent.
ments could increase by about $1.079 billion when compared to This higher cost results from the need to increase the base case.
interchange energy purchases to a possible total of 172,600 Gwd to economically meet consumer demand.
ADEQUACY AND RELIABILITY OF GPU'S TRANSMISSION NETWORf All three of GPU's subsidiaries are interconnected witn the PJM power pool.
The PJM pool also includes interconnec-tions with other utilities in the region, and operates the entire interconnected system as one large system in order to obtain the most economical oalance between consumer demand Projected power flows on transmission and production costs.
lines were analyzed using computer simulation as a means of determining the adequacy and reliability of GPU's System and its relationship with the PJM power pool.
These techniques can identify potential problem areas by altering tne configuration of the transmission system and noting changes in various para-meters of the Systems' operation.
The geographic location of TMI's generating capacity in PJM System had supported the PJM voltage prior to the accident.
trie flow analysis indicated that under some conditions, The power the lack of the TMI generating capacity in the eastern portion of the PJM power pool could result in some reductions in voltage levels.
The analysis also indicated that some isolated minor power overloads and voltage fluctuations may occur, bdt these could be corrected oy system operators.
The corrections would require, however, that some generating units may have to be operated uneconomically, resulting in some incremental increases in power production costs.
17
Based on the results of the load flow analysis for the GPU and PJM systems, it appears that the integrated transmission network could respond to single major facility disturbances without adversely affecting the adequacy or reliaoility of power supplies to GPU's customers.
The loss of the TMI genera-ting capacity has placed a greater ourden on some aspects of PJM's operations, but only at the expense of a slight reduction in overall system reliability.
CONCLUSIONS The energy production costs and reliability for the GPU System result from the interrelationships of the following fluctuating and critical factors--(l) consumer demands, (2) generating capacity used to meet consumer demand, and (3) the extent to which GPU must purchase interchange elec-tric power from other systems in order to provide adequate and reliable service.
Based on the analysis performed in conjunction with DOE and interviews with GPU officials, and in view of the financial and regulatory constraints placed upon the company, we believe that GPU (1) can presently provide adequate power to its cus-tomers and (2) is planning for future energy needs in a prudent manner.
While it is true that the costs for electric power have increased since the accident at TMI-2, and will probaoly continue to increase over the near-term, to a great extent they are attributable to the measures that GPU has had to take to maintain System reliability.
For economic reasons, the GPU i
System has been forced to purchase large amounts of power from neighboring utilities and utilities outside its service area, and that power, due to the types of generation facilities required to produce and transmit it, has been more expensive than if it had been produced within GPU's own service area with the TMI units available.
Because of the operational constraints that have been placed upon GPU, it has had little recourse but to rely on out-side power purchases.
We believe that continued long-term usage of this practice is less than prudent, however, because it makes GPU's System reliability subject to the ability of out-side sources to supply power.
Similar concerns exist regarding GPU's reliance on PJM.
There is no guarantee that the contractual agreements which have served GPU's needs will be renewed because the other parties to the agreements may not be willing to continue devoting a disproportionate share of the total system's capability just to meet the needs of GPU.
We do not believe it is in the best interest of GPU as an electrical generating utility to rely on outside power sources to maintain adequate and reliable long-term power supplies for its System.
18
$G M M 4 E s ii?L i M @ M & $ G & w & Q j
The measures that GPU has taken to provide reliable service in the long-term future appear to be reasonable, yet are greatly dependent on the ability to quickly restart theis accomplished, the undamaged unit No. 1 at TMI.
Once that amounts of purchased energy needed by GPU should diminish, thereby decreasing the System's reliance on outside power As indicated by our sources and its revenue requirements.
scenario analyses of past and future capacity additions, it appears that GPU has planned its service system in a prudentIf the com manner to date.
to resolve the challenges currently facing it, we see no reason why GPU should not continue to provide this same levelAs long a of service in the future.
relying on outside power purchases, however, tne cost of the service it provides will be somewhat higher, with a greater probacility of service interruptions.
I l
(
l l
19
CHAPTER 3 GPU'S FINANCIAL CONDITION CONTINUES TO DETERIORATE the GPU System Two years after the TMI-2 accident, continues to experience a number of serious financial problems The financial indicators of long-term and uncertainties.
stability within the System have continued to decline fromConsolidated net inc their pre-accident levels.
No dividends are ceing Met Ed experiencing a net loss for 1980. Earnings levels preclude access to paid on GPU's common stock.
the financial markets and security ratings have been downgradedThis to unfavorable levels.
Met Ed faces the most of the System in a precarious condition.dif ficult and pressing problems of the three operating companies.
The short-term cash requirements precipitated by the need to make large expenditures for TMI-l and 2 replacement power pur-Short-term chases were initially met by the operating companies.
cash needs borrowings expanded rapidly to meet the companies' Favorable State until adequate rate relief could be obtained.
utility commission decisions on recovering the purchased p cash flow problems experienced oy Met Ed and Jersey Central.
Millions of dollars from stockholder's earnings on the non-TMI units, however, are being used to pay for unrecovered costs for i the TMI units and this continues to adversely affect the compan es No customer revenues or stockholder earnings have been used to pay direct cleanup costs as these have been financial recovery.
covered by insurance proceeds.
Each of the companies has a continuing need to ootain money for refinancing long-term debt and making capital improvements this could Over the next 5 years, in its operating facilities.
amount to over $2.7 billion, much of which will have to beThe inability of obtained from sources outside the GPU System.the companies to o long-term operation of the System.
At the present time, the only accessible source of external cash for the companies is short-term borrowings throu,ih a line This arrange-of credit agreement with a consortium of 45 banks.
ment has limited application to the companies' needs, nowever, and Met Ed has had its access to these funds severely curtaile in recent months.
access 1981, may see a further drastic change in the companies' to external cash resources.
0 20 t
ik 35'Zh~jHaik?_$ $___ i 4.
4 m.. m '
COMPARATIVE INCOME AND EARNINGS DATA FOR THE GPU SYSTEM HIGHLIGHTS ITS PROBLEMS Although the financial soundness of each of the individual the consolidated pre-accident position of the companies varied, Outside experts testified GPU System appeared to be favorable.to tne fact that the companies were Our analysis of the system financial statements capitalized.
for the period 1975-77 showed a steadily increasing earnings picture. 1/
This situation changed rapidly following the TMI-2 accident.
income and return on common equity declined for all the The favorable pre-Net companies with Met Ed experiencing losses. Dividends to GPU's common accident earnings levels eroded.
stockholders were discontinued and the companies lost tneir ability to attract long-term capital.
The following financial statistics provide a general overview of how the operating companies and tne GPU System We have included data have fared during the last 3 years.
for 1978 as a measure of company performance in the last Tables 4 and 5 full year preceding the TMI-2 accident.
show the effects of the TMI-2 accident on two key elements income of the operating companies financial situation--net and return on common equity.
Table 4 Schedule of Net Income (Loss) (note a) 1978-80 Year 1980 1979 1978
(000s omitted)---------
GPU System (note D)
$20,591
$95,783
$138,774 (9,979) 15,585 48,31d Met Ed 24,068 42,045 40,296 Penelec 22,770 50,621 56,561 Jersey Central a/After paying taxes and preferred dividends.
b/ Net of parent company expenses.
1980 annual reports.
Source:
"Three Mile Island: The 1/U.S. General Accounting Office, Financial Fallout," EMD-80-89, July 7, 1980.
21
Table 5_
Schedule of Return on Average Common Equity 1978-8Q Year 1980 1979 1978 Company
(percent)--------
1.5 6.9 10.4 GPU System (2.7) 4.2 12.9 Met Ed 5.8 10.1 9.9 Penelec 3.4 8.2 10.1 Jersey Central i
Source:
GPU.
t ck Table 6 shows the declining pattern of those common s o i
money statistics which reflect GPU's reduced ability to ra se in the equity market since the accident.
Table 6_
Schedule of General Public Utilities Corp._
Common Stock-Related Financial Statistics 1978-8Q Year 1978 1979 1980 Common stock statistic
$106,424,000
$73,385,000
$1.77 Dividends paid
$1.20 77.0%
Dividends paid per share 76.9%
S2.30 Dividend payout ratio 50.34 S1.56 Earnings per share 1979 and 1980 annual reports.
Source:
h limited by the continued omission of its cash divide in To date, stockholders have lost about $200 million t r of omitted dividend payments. During the first quar e d
1979, GPU stock was selling at nearly S19 per share an t
per stockholders received quarterly dividends of 45 cen s for In the second quarter of 1981, GPU stock sold for a high of just under $6 per share and stockholde share.
22 5
nro;gnsm -;.;y y
j^%% ;--,
M:+p.w
~ -
Because over half of the operating companies' capital is the trends of two other financial provided by long-term debt, statistics--coverage ratio and bond ratings--are of vital The bond indentures and SEC regulations require importance.
that a company's earnings be at least two times the interest cost for any 12 of the 15 months preceding a bond issue for new As indicated in the following table, only Penelec met money.
31, 1980.
the legal requirement for issuing bonds as of December Table 7 Schedule of Interest Coverage Ratios For GPU Subsidiaries 1978-80 Year 1980_
1979_
1978 Comoany 1.02 1.99 2.44 Met Ed 2.06 2.73 2.41 Penelec 1.86 1.95 2.38 Jersey Central Source:
GPU System Statistics, 1980.
By February 28, 1981, however, Penelec's coverage ratio had declined to tne point where it could not meet the required SEC provisions and it also was precluded from issuing any long-term GPU estimates that tne recent PaPUC and anticipated NJBPU debt.
oase rate orders will increase earnings sufficiently for Penelec and Jersey Central to achieve the minimum coverage requirement starting sometime after 1981.
A second factor affecting a company's ability to issue long-term debt is the quality rating given to the proposed secur-The two most frequently quoted ratings are those ity offering.
given by Moody's, and Standard and Poor's. At the time of the all but one of the companies' securities had a "Baa"
- accident, This represents a medium-grade security which or better rating.
is considered to be neither highly protected nor poorly secured.
rating represents a lack of desirable investment charac-A "B" teristics.
securities Table 8 shows the downgrading of the companies' that has occurred since the accident.
23
Table 8 Moody's Quality Ratings of GPU Company Securities (note a)
Company 3/28/79_
4/19/79 6/29/79 3/2d/80 3/28/81 Jersey Central:
Bonds (note b)
Baa Baa Baa Ba Ba Debentures Ba Ba Ba B
B Preferred stock "caa" "Daa" "baa" "o"
"o" Met Ed:
Bonds (note b)
A Suspended Baa B
B Debentures Baa Suspended Ba B
B Preferred stock "a"
Suspended "ba"
" b" "b"
Penelec:
Bonds (note b)
A Baa Baa Ba Ba Debentures Baa Ba Ba Ba B
Preferred stock "baa" "ba" "ba" "o"
"b" a,/ Definition of Moody's rating symbols listed in app. III.
I b/ Includes pollution control bonds, i
Source:
Moody's Public Utility Manual.
The significance of the ratings assigned to the securities and the investors' perception of their desirability was demon-strated at Penelec.
On October 1, 1980, Penelec submitted an application to SEC proposing a $50 million bond offering for private placement.
Although the company met the legal coverage requirements, the bond offering was withdrawn wnen it could not be sold.
CURRENT RATE ORDERS HAVE ONLY PARTIALLY ALLEVIATED CASH FLOW PROBLEMS The cash flow problems that have confronted Met Ed and Jersey Central on an almost continuous basis since the accident were again alleviated by the April 1981 PaPUC and NJBPU rate orders.
Current replacement power costs and prior energy costs paid for by the companies are now being recovered from consumers in the rates.
The PaPUC and NJBPU orders that removed the costs of TMI-l and 2 from base rates, however, continue to have an adverse impact on the companies' cash flow.
Certain expenses such as interest, preferred stock dividends, and operations and maintenance that are major components of the base rates continue to accrue for 24
i these units and must be paid from non-TMI stockholder earn ngs.
require cash In addition, extensive capital improvements that Until outlays are being made on TMI-l prior to its restart.these cos
- ates, the cash flow problems are likely to continue.
Ratepayer contributions to accident recovery costs have been limited to paying for replacement energy i
The unanticipated loss of the TMI units in 1979, and the r d for continued outage over 2 years later, precipitated a nee t of GPU to make extensive energy purchases to replace the outpuThere is no the lost generating units.
l ment TMI-l will be ready to restart until early 1982, and rep ace il then.
energy purchases for that unit will have to continue unt As is not expected to be returned to service until 1988.
replacement energy costs for sales subjecttotal about $60 TMI-2 indicated in table 9, to the jurisdiction of PaPUC and NJBPU through June 1981.
Table 9 Estimated Cost of TMI Replacement (note a)
Energy--April 1979 througn June 1981 Year 1981 1979 1980 (9 mos.)
(12 mos.)
(6 mos.)
To ta_1_
Company
--(millions)--------------
Met Ed
$100
$110 S 60
$270 25 115 55 35 Penelec 220 Jersey 95_
5_0_
75 Central Total
$210
$260
$135
$605 include Pennsylvania and New Jersey gross receipts a/Does not tax.
Source:
GPU.
/ used by The levelized energy adjustment clause provisions 1 osts the PaPUC and NJBPU to account for fluctuations in energy c The fuel did not provide for the rapid increase in energy costs.
b t 0.4 cost for generating electric power by the TMI units was a ouf rom PJM follow cent per kWh and the initial replacement costthe accident ra Because h sed 1/A ratemaking mechanism to allow for changes in fuel an power costs.
and every 6 months in New Jersey.
25
the companies could not immediately pass this increased cost to consumers, a large outlay of cash was required by tne companies during the first year following the accident.
Funds were ob-tained from bank borrowings and internally generated sources.
The companies' deferred energy balances--funds expended but not collected through consumer rates--increased from S103 million on December 31, 1978, to a high of $261 million in March 1980.
Based on several rate orders issued by the PaPUC and NJBPU subsequent to the accident, the companies have been allowed to recover the TMI-related replacement energy costs on a current basis and to begin recovering their deferred energy balances.
As a result of these orders, the companies have had the replacement energy cost burden lifted and placed on their ratepayers.
Rate-payers, however, have not been assessed for any of the cleanup costs.
The following table shows the dates and amounts of the companies' authorized collections for TMI-related energy costs.
Table _10 Schedule of_ Replacement _ Energy Cost Recovery-TMI Related April 1979_through June _1981 Date of Order Jersey Central Met Ed Penelec Total n
(millions)----------------
S'148 June 13, 1979
$148 S 98
$ 42 140 June 19, 1979 73 73 Feb. 8, 1980 42 Apr. 1, 1980 42 a/82 a/48 130 May 23, 1980
~
~
(6)
(6)
Apr. 9, 19 81 1
Apr. 23, 1981 1
Total collections 191 247 90 528
~
Less: Base rate exclusion (note c)
(88)
(160)
(78)
(326) l Net cost to consumer a/$103 b/S 87 o/S 12 S202 a/ Includes $50 million and $20 million in deferred energy cost l
collections for Met Ed and Penelec, respectively.
b/ Includes revenue taxes of $12 million, S4 million and $0.5 million for Jersey Central, Met Ed, and Penelec, respectively.
c/ Includes tax surcharge.
Source:
GPU.
i l
l 26
i have Under the energy clause provisions, the compan es from ratepayers i
$528 million (including taxes)timated $605 million collected about through June 1981 with the balance of the esfor replacemen in The net id cost to consumers, however, has only been$326 million of TMI-r by consumers was deleted from the rates.
cbout Unrecovered expenses affecting cash
_ requirements are paid by stockholders payments for replacement energy costs areident-related costs Consumers' being complemented by payments for other accAs in the case of consumers, earnings.
used to pay for from common stockholders' however, no stockholder contributions have been1981 these costs were still As of July 1, direct cleanup costs.
being covered by insurance proceeds.
ident recovery The stockholders' contributions to the acc i
fixed expenses ef fort have been used primarily to cover certa nfor gene ate total bill.
debt issues, portion of the consumers' interest payments to bondholders for long-t dividends on preferred stock,In addition, a return on stockholders' ates.
equity is normally allowed as part of the base r and depreciation.
these costs for TMI-l were Similar costs for At the time of the accident, included in all three companies' base rates. collected by Penelec and Jers but 19, 1979, By PaPUC and NJBPU orders of Junerespective TMI-2 were being/
2 1
not by Met Ed.1979, costs for Penelec and Jersey Central were reprevious and June 18, to collect these i
As a result, llowed to collect costs was withdrawn.in revenues that they would normally have been ain thei on an annual basis if TMI-2 costs had been relying on company estimates that TMI-l left 1979 or early 1980, PaPUC and NJBPU, ies' base rates until would be returned to service in late the fixed costs of the unit in the companNJBPU removed $18 million for On April 1,
- 1980, A similar tes.
early 1980.TMI-l expenses from Jersey Central's base ra1980, when it removed Penelec's action was taken by PaPUC on May 23,
$27 million from Met Ed's rates and $12 million fromAs a result of t he TMI units rates for TMI-l costs. annualized revenue from base rates for t 65 million.
was effectively reduced by a total of $1 panies' but rates had not 1/PaPUC had approved Met Ed's request, yet been increased.
1 1
27 l
l Additional fixed cost increases of $19 million since 1980 were disallowed in the April 1981 PaPUC order.
The NJBPU order on Jersey Central's latest filing was issued on July 23, 1981, and it disallowed similar cost increases amounting to about
$20 million.
In summary, the elimination of TMI costs from the companies' rates is estimated to have the following effects on common stock-holders' earnings through 1981.
Table 11 TMI Costs Eliminated From Rates and Estimated Effects on Stockholder Earnings-April 1979 through June December 1981 (note a)
GPU Company Jersey Cost component Met Ed Penelec Central Total
(millions)--------------
Operations / maintenance S 19
$ 10
$ 10
$39 Depreciation 30 15 17 62 Reserve capacity credit (28)
(12)
(7)
(47) j Taxes 53 24 24 101 Interest 34 17 17 68 Preferred dividends 9
5 6
20 Common equity return 43 19 21 83 Total costs and return eliminated 160 78 88 326 l
Less:
Tax reductions (86)
-(42)
(46)
(174)
Net reduction in stockholder earnings S 74
$ 36
$ 42
$152 a/Does not include effects of possible accounting changes for TMI depreciation and reserve capacity costs.
l Source:
GPU.
l In addition to the costs charged against stockholder earnings shown in table 11, the companies have incurred other I
unrecovered costs.
For example, PaPUC and NJBPU have not allowed the companies to pass through the interest costs on the 28 1
-u._
w.-= q q A ~ y m ;.,
h s
short-term borrowings required to pay for the energy purc ase.
d at These costs have been tied to the prime interest rate an From June 1979 through times exceeded 23 percent per annum.
illion December 1980, stockholders have absorbed over $52 m of these interest costs.
d Certain costs for TMI-l activities have also not been passeNRC tart.
number of changes for TMI-l that are being made prior to res on to ratepayers.
f about 520 These changes have resulted in capital expenditures o lion in 1981.
million and GPU expects to spend an additional $21 milt l cost These costs, however, will presumably be added to the to a i tion of TMI-1 and collected from the ratepayers through deprec a Until that is allowed, however, the expenditures es.
represent a current drain on the companies limited cash resourc expense.
d by Jersey Central's finances have Deen adversely affecte The company was an added side effect of the TMI-2 accident.
cted constructing the Forked River nuclear plant with an expeConstruction of the facility l
completion date in the mid-1980s.
l because was suspended shortly after the TMI-2 accident part y On of its impact on Jersey Central's financing capability.the project wa i
tely company requested NJBPU approval to amortize its approx main the faci November 6, 1980, in annual
$400 million investment 23, 1981, but $14 million t d in This was approved on July interest charges and preferred dividends on funds inves eafter r the project, Jersey Central's other earnings.
i l
~GPU HAS LONG-TERM FINANCING
_NEEDS THAT MUST BE MET that The recent PaPUC and the anticipated NJBPU rate orders insolvency, provide the rate relief necessary to avoid near-term the com-particularly for Met Ed, are not sufficient to improve d
financial posture to the point where they can obtain fun s There are two areas of panies' in the long-term capital markets.
financial need that are nearly always met through these long-d f.nancing i
term borrowings--debt refinancing / redemption an5-year period, the total Over the next eas capital financing needs of the companies in these two ar ccr. '.ruction costs.
amount to more than $2.7 billion.
refinancing / redemption Debt
~ requirements the tnree operating companies have k issues Between 1981 and 1985, over $400 million in long-term debt and preferred stocTaole 12 shows tnat that come due and/or have to be redeemed.
- 1984, most of the replacement / redemption will occur in 1983, and 1985.
29
Table 12 Schedule of tong-Term Debt Replacement and Preferred Stock Redemption 1981-35 Year 1981 1982_
1983 1984 1985 Company (millions)
Met Ed:Iong-term deot 3/$15.23
$10.03
$52.23
$17.19
$46.98 Penelect Long-term debt,
6.93 11.43 14.43 58.93 1.84 Preferred stock 2.85 2.85 2.85 2.85 2.85 Jersey Central Long-term debt 10.28 11.30 41.07 20.64 58.99 Preferred stock 1.25 2.50 2.50 2.50 2.50 GPU Service Corp.
.53
.58
.42 4.13
.15 39.00 GPU Corporation Total
$76.07,
$38.69
$113.50_
$106.24_
$113.31_
a/ Includes $13millioninbondsmaturingOctober1,1981.
Source: GPU If new bonds are issued and the funds used to retire bonds that are maturing, the two times interest coverage ratio require-It would ment commented on earlier generally does not apply.
apply if the interest rate on the new bonds is lower than theold bo replacement is greater than 2 years--an unlikely situation for However, if the companies' earnings do not demonstrate to investors that the bond offering is a sound investment, GPU.
~
potentialis unlikely that the new bonds can be sold even if all then it The consequences of the failure legal requirements are met.
The cond to retire the debt on the due date is uncertain.
trustee would presumably evaluate the company's financial prospects at the time and determine how the interests of the bondholders would best be protected.
30 I
o-t"aHm9sJ.ggsm
Construction funding
} requirements The System's ability to continue furnishing reliable power to its customers requires a large annual outlay of funds. In addition to constructing generating units to replace older, inefficient units, new units must be added to meet load growth.
Existing units also require certain capital improvements.
Transmission and distribution systems require expansion and upgrading to allow new customer hookups and to prevent systemMaintainin degradation.
company basis is important to the continued integrated operation of the GPU System and to the PJM power pool.
During the 5-year period ending December 31, 1980, the operating companies expended over $1.78 billion in connection New generation capacity with their construction programs.
costs--such as those for TMI-1 and 2--are usually shared by the companies, while other construction costs are independ-As shown in table 13, the ently financed by each company.
companies reduced their construction outlays in 1979 and 1980 principally as a result of the cash flow drain resulting from the accident.
Table 13 Schedule of Construction Costs for GPU Operating Companies 1976-80 Year 1980 1979 1978 1977 1976 Cortpany (000s omitted)-
Met Ed S 43,182 S 53,559
$ 87,657 S101,503
$ 90,324 Penelec f6,347 68,615 87,246 121,721 133,752 Jersey Central 132,889 221,086 225,749 185,019 141,354 7btal S244,418
$343,260 S400,652
$408,243 S365,430 Source: 1980 Annual Reports.
Planned construction expenditures for the 5-year period 1961-0 smount to over $2.3 billion.
The details of these 4
expendi.ures are shown in table 14.
31
Table 14_
Schedule of Proposed Construction i
Expenditures by GPd Operating Companies 1981-85_
Met Ed Penelec Jersey Central
(millions)---------------
New generation:
Sayreville coal
$50 conversion (note a)
S 18 Raystown (Hydro) 130 12 Seward-7 (Coal) 5 14
$28 4
Coal #1 3
1 Coal #2 6
4 6
Pumped storage 92 156 Other 33 Total Existing generation:
341 Oyster Creek (Nuc) 34 35 71 18 TMI-l (Nuc) 189 33 223_
394_
Other 104_
Total Transmission:
213 Ontario Hydro 78 119 45 332 78 All other 45
~~
Total 296 199 145 Distribution:
124 26 50 Nuclear fuel:
13 12 10 General:
$694 S1,251 Sjg3 TOTAL If no a/ Assumes 50 percent of cost paid by Government. assistance probably not occur.
Source:
GPU.
Proposed construction expenditures for new generating 1981-85 capacity tend to oe relatively low during theperiod but are ex the 1986-95 period.
Seward-7
(
32
^
h 4
g
--e
,r -Aw
i addition planned for completion during that period.
Originally scheduled for service in 1985, this date has now slipped to 1989.
Completion of the remaining new units is now planned for the early 1990s.
Consequently, most of the new capacity expenditures are expected to occur between 1986 and 1994 as expenditures are made to complete Seward-7, the two coal units in 1991 and 1993, and the pumped storage project by 1994.
Planned generation expenditures for 1986-88 will be further increased if the decision is made to recommission TMI-2.
Current cost estimates (including the replacement of the nuclear fuel) are over $500 million when adjusted for a 10-percent annual inflation factor.
Recommissioning funds are expected to come from the capital markets as opposed to the companies ' expectations that the estimated $600 million for cleanup activities will come from other sources.
FINANCING DIFFICULTIES MAY AFFECT COMPLETION OF FUTURE CONSTRUCTION PROJECTS At the present time, the only sources of money for meeting the System's capital needs are its internally-generated funds from non-TMI earnings and its short-term borrowings.
These sources are expected to be sufficient to meet the planned capital requirements of $389 million in 1981.
Utility company financing plans rely on short-term borrowings to " smooth out" deficiencies in internally generated funds for the next Access to the long-term capital market will even-year or two.
tually be essential, however, if planned capital expenditures are to be funded after 1982.
Internally generated funds are generally meeting current capital requirements The continuing costs for TMI-1 modification; generation, transmission, and distribution system construction and/or upgrading; and expenditures for bond redemptions and sinking fund requirements have had to be met primarily from internally generated funds since 1980.
Shortly after the accident, Penelec issued $50 million of first mortgage bonds in June 1979.
Jersey Central later raised $97.5 through two first mortgage bond issues to reduce short-term bank loans and refinance a maturing bond issue.
Since 1979, however, no financing through l
long-term debt obligations has been possible.
The companies spent $279 million in 1980 to meet their capital obligations.
Of this amount, S246 million was for construction an'd $33 million was needed to retire matured securities and for sinking fund purposes.
A portion of l
the approximately $20 million collected in base rates for TMI-l through May 1980, provided part of the required 33 1
The balance of the money came from depreciation funding.
expense and earnings on non-TMI assets plus deferred taxes While the $33 million for and deferred energy collections.
retiring securities and sinking fund purposes met pre-accident budget expectations, the $246 million available for construction was 52 percent less than the $515 million budgeted before the accident.
Capital expenditures for 1981 and 1982 are again expected to be primarily financed from internally generated funds.
Construction activities costing $259 million--and comprising only 36 percent of the $720 million for the pre-accident budget--are proposed for 1981, including $21 million for TMI-l modifications and $54 million for accrued construction lia-Bond retirement and sinking fund requirements bilities.
The source of are budgeted for an additional $76 million.
these funds continues to be the non-TMI revenue receipts plus short-term borrowings.
GPU expected to meet part of its 1982 capital needs from external sources since the return of TMI-l to service in January 1982 should increase the interest coverage ratio to This would have allowed an increase the required levels.
in construction expenditures as well as funding the $77 million It does not needed for bond retirements and sinking funds.
appear, however, that the companies can realistically count on selling any planned bond issues until after 1982 and capital expenditure levels will again depend on revenue receipts and short-term borrowings.
Short-term borrowings remain GPU's only external source of funds Short-term bank borrowings have been the key element in keeping the GPU system current with its cash requirements.
Although less reliance is placed on short-term debt in future it remains a necessary component of continued budget plans, System viability.
On June 15, 1979, GPU officials negotiated a Revolving Credit Agreement (RCA) with 43 banks 1/ to finance the unrecovered cost of purchased replacement energy and other current cash obligations.
A maximum borrowing level of $412 million was authorized for the System but thus far, the banks have limited the amount of outstanding loans to
$292 million.
Each of the three operating companies and the parent corporation had individual sublimits set with the aggre-The loans have 6-gate borrowing limited to $292 million.
month maturity dates but thus far have been extended each time 1/Two other banks added later.
34
.;4 g: Wh&
The RCA expires on October 1, the renewal period has come up.
1981, at whicn time the banks will have the option of continuing, amending, or terminating the agreement.
The RCA provided a ready source of funds to cover the time lag between payments for power purchases, and PaPUC and NJBPU approval This was reflected to collect the addej costs in customer rates.
in the increased borrowings that occurred during the early part of 1980, as shown on the graph on page 36.
Penelec made limited use of RCA and by 1980 had repaid As indicated oy the graphs on page 37, its outstanding loans.
however, Jersey Central, Met Ed, and the GPU Corporation relied extensively on RCA for cash supplements to their depleted finances.
Although Met Ed and Jersey Central had nearly reached their credit limits in April 1980, Met Ed is the only company Jersey is currently being constrained by the terms of RCA.
that Central was able to reduce its outstanding borrowings when tne NJBPU granted the company $60 million in interim emergency Het Ed's request rate relief in May 1980, subject to refund.
to PaPUC for $34.1 million in interim rate relief, however, 28, 1980.
The banks viewed this as was rejected on August leading to a further deterioration of the assets supporting Met Ed's loans and limited Met Ed's credit to the value Starting in late 1980, liquid assets of its liquid assets.
were defined as the sum of Met-Ed's deferred energy balance, 80% of Customer Accounts Receivable balance, and an assigned of pledged uranium stocks at the end value ($20 million)
In June 1981, Met-Ed's cash needs were such of each month.
that they had to pledge to the Danks tneir coal inventory The banks in order to increase their liquid assets formula.
have accepted the coal pledge but have reduced Met-Ed's liquid assets formula by $5 million for July and $10 million The historical and forecast components of thereafter.
Met-Ed's 1981 liquid assets credit limit is shown by the graph on page 38.
CONCLUSIONS that the financial condition of the GPU 1
It is apparent Cfstem has deteriorated significantly since the TMI-2 accident and that the rate relief granted to date by PaPUC has done little more than maintain the companies in a solvent con-dition.
Tne present inability to obtain external sources of capital, coupled with the need to increase capital expend-(
raises serious questions as itures over the next 5 years, to the System's continued viability as a provider of electric I
power.
i 35
i I
I I
I I
I I
I ms z!
g o e
~
- 8 mi 4
ce
,1 l
4 m
o m
r*
z M
i ea o
z i
c e CE i
F u,
e e
> e a%
m i
o zl p
g 3
o e m
i ua 3
ma o
W q
g a
e o
9 on w
93 y
ri 9
E i
c:
o ac e
c m
g.
i ss
~
e a
w ai l
g
=
oi m.
m s
ma e,
m 9-E i
ea n.
o u*
y
,n g
o
\\4 z!
cle
=lg l8 k
>sB I
i 1
1 I
I l
4 f
o o
o o
o o
o o
o o
m o
m o
m o
m o
m l
v v
m m
ru na w
w l
twaawoz aosseam l
t t
36 r
t
'N.
%YW9 4 een g pg.y. 0 ' z*
- mie & & b &,[ * '
em V'
- %~'
4 I
I l
5 8
I I
l 4
l l
5 I
I l
g g
a, a.
g g g.
o8
-- o o
g
,3 g
g
==
,3 e
k
=
E i
. l l
.H E
=
g m
m a
4, g
,4, E g E.
p
- o' en
=.
p o.
en W
w A
G
=.
-,1l 4
E l
c w
- r. -
,,8
.c b
g E 8
=%
.g %
g o%
m a:
x e'
o.
'==
8 o
'a=
==
g e
a-
=
a a
g
-..e..
w u.
e2 p
m.
u
., (-
U r
P g
3 m.g %
i.-
i-i i
i i
i i
i i
I i
I i
i i
i i
3 3 0 2 8 2 3 ? 2 2 3 3 2 8 2 3 V 2 r
.J o a o c.J a. m o C
.J a-o r a o.J.J.m e i
I I
I I
l l
I I
I I
4 4
I I
I - a.
g
.g E g g
- w f8 o8
.o.
.t W
g g
g
-,3
=-
o g
c.
I
- =:
-=.,,
o i
- a; g
r.
8
=
r.
og o
w 3
=,
x
=
r
, y, E
- =.,.
- =.
g
=
f s:
A.
- t..
- z :
g o
- m, y.
3 m.,;.
A 1
o u
s.
z.
r ol
=
y 9
2 g.
g m
g Y
c
=
,g-k'
- c..,
..e s: e
-..i u
c m,. 8 1
1 I
i i
l i
I I
I I
I I
I i
1 -
% u) o o o o o o o o,
o 8
o 8
o, o.
8 o o o
o o,
v da ru w
a m
n m
w se e
e r
.a.J o z a o.J.a. er e c
s a oz oo s a am 37
J SHORT-TERM DEST & COMPONENTS OF R.C.A. CREDIT
.)
MET-ED:
1 120 g
g g
g g
g g
g g
i g
110 DEFERRED EMERGY INCLUDES NOTE 8 II A PLEDGE OF COAL PILES FOR JULY-SEPT an A ss nittIon CREDIT REDUCT!on IM JULY 31 A S19 MILLION CREDIT REDUCTION IN AUC-MPT 100 TOTAL R.C.A.
CREDIT LIMIT g go I
q L
4r JL h
L 80 I
URANIUM 9;
O 70 n.c.A.
4.i N
v ExPInEs
-m to u
'l y
D 60 uf Accounts
"'""" "87 4
0 RECE!UABLE I
L 8'E 1
k.l 50 1
L y
A l >s; R 40 e.+
s h
3 30 k1 w
DEFERRED
.p 20 ENERGY (SEE NOTE) 1.l _
2 V
,,,'~~
10 3(j I
I I
I I
I I
I I
I I'
I Na J
F M
A M
J J
A S
0 N
D 0
h-
/....... --------------------ggg1......----------------------s Source: GPU
=.='y3)y y,aw.,,..:.-w w ane- # A M~"-
'~
Barring the possibility of losing another major generating station or a disruption in the System's transmission grid, customer service could probably be sustained at an austerity budget level for some time before brownouts and possible New customer hookups would extended power outages occur.
probably have'to be restricted, however, and continued access to outside power purchases would be essential.
Of greater concern is the need to obtain funds to retire long-term debt, beginning in 1983, and to continue meeting Strong positive signals from sinking fund requirements.
the State regulatory commissions that the System's financial Without them, we viability will be maintained are needed.
seriously question the ability of the companies, and Met Ed to sell any kind of long-term borrowings that in particular, would provide funds to meet these obligations.
Two events appear to hold the key to the continued existence of the GPU System--the restart of TMI-l with the appropriate rate increases and the development of an ac-ceptable way to f und the uninsured portion of TMI-2 cleanup The importance to the System of restarting TMI-l costs.
was discussed in chapter 2.
Chapter 4 provides our analysis of the cleanup cost problems.
39
CHAPTER 4 FUTURE FINANCIAL VIABILITY OF GPU MAY BE_ CONTINGENT ON_ RESOLUTION OF FUNDING FOR TMI-2 CLEANUP COSTS The accident at TMI-2 on March 28, 1979, left GPU with the longest and most costly nuclear accident recovery pro-cess experienced to date by the electric utility industry.
Through December 31, 1980, GPU has spent about $180 million, largely from insurance proceeds, for the cleanup effort with little progress in decontaminating the reactor contain-ment building.
Current estimates indicate that it will take another $600 million--net of remaining insurance funds--and from 4 to 6 years to complete the job.
This cost estimate assumes that changes will be made in the regulatory process experienced to date so that expedited cleanup procedures can be implemented.
The need to complete the cleanup effort is not in question, but the source of adequate funding is.
Insurance proceeds have been covering most of the costs to date, but they will be exhausted by late 1983, even under the present limited rate of expenditure.
The current position of the PaPUC is that no ratepayer funds will be used to pay for cleanup costs.
Under current regulatory constraints, GPU does not have the financial resources to both continue the cleanup effort after the insurance money runs out and continue with even a limited capital improvement program to maintain System reliability.
Although a near-term bankruptcy action is no longer imminent, tne failure to resolve the cleanup funding issue could adversely af fect GPU's financial condition to tne point where a future insolvency could occur.
We believe, however, that there are too many uncertainties as to what mignt result from a bankruptcy action to make this a viable solution to the GPU System's problems from almost any point of view.
Moreover, the current GPU problems are creating substantial additional costs for many electric utility cus-tomers throughout the country and these have been estimated to increase further if GPU were forced into bankruptcy.
CURRENT CLEANUP COST ESTIMATES ARE STILL BEING REFINED Shortly after the accident at TMI-2, GPU contracted with the Bechtel Corporation to help develop a cleanup and reactor restoration cost estimate.
Accurate information on the conditions inside the containment building was generally not available, so numerous assumptions had to be made.
The initial 40 m2rmrmnnwashua:h
estimate assumed that the reactor core could be removed with-in 20 months after the first containment building entry and that TMI-2 restart could occur around mid-1983.
It was esti-mated the cleanup costs would be about $133 million in 1981 excluding all operations and maintenance (O&M) costs.
- dollars, Restoration costs were expected to be about $102 million plus
$60 million to replace the damaged core.
A contingency fund of $105 million brought the total cost of decontamination and restoration to $400 million.
The first major revision to the 1979 estimate was released in August 1980.
Prepared jointly by Bechtel and GPU, it was considered to be the first truly comprehensive cost estimate.
The time schedules for removing the nuclear core and restarting the unit slipped to 1983 and 1985, respectively.
The total cost of $952 million for cleanup and restoration All costs were computed included $97 million for O&M expenses.
in 1980 dollars.
The August 1980 estimate has served as the basis for all further analyses and revisions, not only by GPU consultants and Bechtel, but by Theodore Barry and Associates, to PaPUC, and also by Nuclear Energy Services, consultants to NRC.
The revisions to the August 1980 estimate incorporated changes to the basic assumptions used regarding the inflation factor and more particularly to the time schedule for core The adjustments made to the August 1980 removal and restart.
cost estimate and the assumptions used are shown in table 15.
The current cleanup cost estimate of about $701 million It reflects was prepared by GPU and Bechtel in April 1981.
the costs expected to be incurred between 1981 and 1987 to complete the cleanup effort and is the latest in a series of cost revisions that grew out of the first estimate prepared in June 1979.
A further revision to the April estimate is under study by Bechtel and is expected to be available by August 1981.
GAO ASSESSMENT OF CLEANUP COST ESTIMATES each with The development of a series of cost estimates, a different total cost was probably inevitable, given the lack of good information and the uncertainty about containment conditions that went into the initial estimates.
As additional containment entries were made and more data was available for l
l analysis, assumptions changed and the cost estimates began to vary.
It is notable, however, that very few of the cost variables changed during the early revisions, and it was not until recognition was given to a possible 2-year delay in removing the core and restarting the unit that the costs escalated appreciably.
41
Table 15 Summary of Total TMI-2 Recovery Cost Estimates Source /date of estimate Theodore GPU Nuclear
- Barry, Corp.,
Cost GPU, 8/80 9/80 GPU, 11/80 12/80 component (note a)
(note b)
.(note c)
(note d)
(millions)-----------------
Cleanup
$598
$598
$598
$937.1 O&M 97 90 90 169.0 Core replacement 70 70 70 123.9 Damaged core 37 Restoration 187 187 187 290.9 Restart 81.3 Inflation
_252 175 Total 952 1234 1,120 1,601.2 Less:
Insurance
_(300)
_300)
(
_300)
(300.0)
(
Net Cost
$652 S934
$820
$1,301.2 a/ Based on core removal in April 1983 and restart in 1985.
~
Estimate is in 1980 dollars.
b/ Based on GPU August 1980 estimate but added the book value of
~
the damaged core and inflated entire cost by 10 percent annually.
l 1979 O&M costs of $7 million were not included.
c/ Based on GPU August 1980 estimate but inflated restoration costs only by 10 percent annually.
1979 O&M costs of $7 million j
were not included.
d/ Based on core removal in 1985 and restart by December 1987.
All costs are in time-of-expenditure dollars.
Restoration means meeting pre-accident conditions.
Restart means capital modifications to incorporate new NRC requirements.
Source:
GPU.
42 1
_mmB_E_mRMr '1EM1r.ELsiMm " MWM
..... +:
1 +. g.p,.4 6.^s.,.....uy,..4.w w ym p :
. i w
The various methods used to account for inflation tended to confuse the estimated cost totals.
The Theodore Barry estimate in September 1980, for example, applied an infla-tion factor against all costs.
We also noted that while four of the estimates included restoration costs and one included restart costs, frequent mention was made of only the cleanup portion in public information releases or hearings.
When the cleanup costs were reported, O&M costs were usually included, although they are carried as a separate cost item in the estimates.
To assess the current financial needs of GPU to complete the TMI-2 cleanup, we believe a more realistic approach is to isolate the cost of cleanup from all other cost elements.
This is suggested for several reasons.
Previous attempts to aggregate all the costs incurred to restart TMI-2 have had to take into account too many uncertaintiec, thereby affecting the reliability of the estimates.
The number of containment entries made to date has improved GPU's knowledge of cleanup requirements over what was available in August 1980, and a cost estimate for this phase of the work can now be made with some degree of precision.
A detailed understand-ing of the requirements to restore and restart the unit, however, is still not available.
Consequently, we believe any attempt to incorporate cost estimates for restoration is premature and tends to confuse, rather than clarify, the needs of the utility.
Furthermore, it is possible that if and when restoration work is started on TMI-2, it will be viewed as l
new construction, with the necessary funds available through normal construction financing methods.
l We also believe that cleanup costs should not be aggre-l l
gated with O&M expenses.
These expenses--which are expected to average about $20 million annually over the cleanup period--are for activities that would normally be required if the plant were running.
Therefore, we view these costs as not attributable to the accident and subsequent cleanup Costs.
The cleanup cost revision prepared by GPU/Bechtel in April 1981, is probably the best representation of the financial needs of GPU to complete the TMI-2 decontamination l
process.
The estimated cost on an annual basis is shown in table 16.
43
i Table 16 Estimated TMI-2 Cleanup Costs Bechtel Reassessment, April 1981 (note a)
/
Actual expenditures Estimated future costs (note a)
Total 1979 1980 1981 1982_
1983_
1984 1985 1986 1987_ 1981-8_7_
(millions)
(millions)
$94.5 $86.7 S44.8 S117.4 S177.9 S131.4 S145.9 $80.8 $3.2 S701.4
_a/Does not take into account insurance proceeds.
b/ Estimate is in year-of-expenditure dollars.
Source: GPU.
Of the approximately $180 million in actual costs incurred in 1979 and 1980, about $170 million has been covered by insur-About S130 million of insurance proceeds remain ance payments.
to be applied against the estimated $701 million total cleanup leaving a balance to be funded of just under $600 million.
- cost, This estimate has the benefit of data and observations gained from the latest containment entry and the GPU experience in decontaminating the auxiliary building, and we believe it probably represents a reasonable expectation of what the total cleanup cost might be.
Our assessment of Bechtel's estimate is generally supported by the initial observations of a NRC consultant.
NRC contracted with a private consulting firm, Nuclear Energy Services, to conduct a technical assessment of the proposed The consulting firm reviewed the technical cleanup program.
aspects and associated costs of the GPU/Bechtel estimate with engineers from both companies.
Although the assessment is not yet complete, the consultant has tentatively concluded that the cleanup program is technically adequate and the costs reasonable if one accepts the GPU/Bechtel assumptions about worker productivity in decontaminating the containment building and the physical condition of the reactor vessel components.
DOE officials, however, consider the GPU/Bechtel estimate to be on the high side and believe the cleanup can ce done for less cost.
The major difference appears to be in the DOE feels that the radiation levels labor cost category.
need not be reduced to the levels proposed in the GPU/Bechtel Therefore, less decontamination effort would be estimate.
required before putting workers into the containment building d
44 1
- L4jpg;&,if W. A 4s --.L
'; ~
to prepare for core removal and fewer manhours are expended.
The tradeof f appears to center around the amount of worker exposure one is willing to risk in order to reduce the cleanup time and thereby the total cost.
ESTIMATED COST COULD VARY GIVEN DIFFERINC CIRCUMSTANCES Although we believe that the $600-million estimate for the remaining cleanup is reasonable and prudent for planning
?
this time, several events could occur that would purposes at tend to either decrease or increase the actual incurred costs.
One such event, while beyond the control of the utility, could af fect the time presently planned for core removal and possibly allow some cost reductions.
GPU/Bechtel engineers have assumed that considerable damage was done to the com-ponents within the reactor vessel due to the overheating of the core, and have planned their work schedule accordingly.
If the reactor vessel components are not as badly damaged as anticipated, core removal would be expedited and costs would be reduced.
Worker productivity is another area of cost uncertainty.
Bechtel/GPU labor cost estimates for decontaminating the con-tainment building are based primarily on the experience gained from cleaning the auxiliary building.
The engineers believe similar extensive manual scrubbing of the exposed surfaces that in the containment building with chemical additives will also be required.
The physical effort required, coupled with the protective clothing needed to minimize radiation exposure, was found to limit the time an individual could spend actually There is also some concern about processing the working.
waste water generated by the cleanup which would contain the various chemical additives used in the decontamination process.
During a containment building entry in June 1981, DOE and GPU conducted a decontamination experiment on a selected area using a hot water spray under 1000 lbs. pressure as a Initial results were mixed, but were sufficiently cleaner.
encouraging to warrant further testing.
If the hot water method for decontaminating the containment building surf aces proves to be feasible, it will eliminate the need to filter out any chemical additives.
The demonstration also showed that less water than anticipated was used which could further decrease the time and ef fort needed to process the cleaning The use of the hot water spray treatment could serve water.
to reduce both the time and labor costs of decontaminating the containment building.
An increase in the present $600-million estimate could also occur, even though we believe the estimate prudently recognizes the possibility of some adverse conditions occurring.
45
Just as reductions in the anticipated time schedules hold out i
the possibility of decreasing cleanup costs, slippages and delays in meeting the projected time schedules can result in As an example, revising the time schedule increased costs.
for core removal f rom 1983 to 1985 and restart f rom 1985 to 1987 increased costs by about $150 million.
/
Among Cleanup delays could result from several reasons.
a public perception that the radioactivity in these are (1)
TMI-2 is safely contained and can remain there for a long period of time and (2) legal requirements for public hearings The combination of these on NRC decisions for TMI-2 cleanup.
two reasons could result in protracted public scrutiny of GPU officials each NRC decision through the hearing process.
estimated that public hearings on each major item could add Such a move could 6 to 12 months to the cleanup schedule.
extend the cleanup time well beyond 1987 with an attendant increase in costs.
is the general consensus of nuclear and other scientific experts that TMI-2 poses no immediate public It
- However, health and safety hazard in its present condition.
it is also their consensus that the cleanup process should In parti-be carried forward as expeditiously as possible.
cular, it is important to immobilize the radioactivity in gallons of water in the containment building and the 700,000 establish barriers to its release to the outside environment The only barrier to the radioactive as soon as possible.
water at the present time is the containment building itself.
Regardless of the decision made on the future of the f acility, the containment building has to be decontaminated and the The longer cleanup reactor core removed and disposed of.
is delayed, the greater the opportunity for TMI-2 to become a health and safety hazard.
NRC also sees a need to continue with the cleanup and waste disposal because the island is not a designated waste The NRC staff, in a Programmatic Environmental repository.
Impact Statement (PEIS) on TMI-2, stated that "The staff has concluded that TMI should not become If the a permanent radioactive waste disposal site.
damaged fuel and radioactive wastes are not removed, the Island would, in ef fect, become a permanent waste The location, geology, and hydrology disposal site.
of Three Mile Island are among the factors that do not meet current criteria for a safe long-term waste Removing the damaged fuel and disposal facility.
radioactive waste to suitable storage sites is the only reliable means for eliminating the risk of widespread uncontrolled contamination of the environ-ment by the accident wastes."
46 M jaGAD$OiMI$hd
l w
It is important, therefore, that undue delays in proceeding with the cleanup are minimized while giving full consideration to health and safety concerns.
CLEANUP COST SAVINGS WILL BE MINIMAL IF TMI-2 IS DECOMMISSIONED l
We found that cost estimates for cleanup will be about The same regardless of a decision to decommission TMI-2 to pre-accident condition and restart rather than rastore it Based on the present assessment of what happened during it.
the accident, GPU expects that the major components of the unit--primarily the reactor vessel and steam generators--
will be reusable and the unit can eventually be put back Consequently, all cost estimates for cleanup in service.
are based on that expectation.
A final decision on the condition of the reactor components will not be possible until the nuclear core is removed and the reactor vessel and steam generators can be examined and tested If too much damage has occurred, it is probable more closely.
the unit would be decommissioned.
However, such a decision would negate very little of the cleanup effort presently Our discussions anticipated as part of a restart program.
with NRC officials and statements included in NRC's PEIS, indicate that decommissioning TMI-2 would probably require an if not greater, decontamination effort than restarting Although the component and reactor support systems within
- equal, containment might be handled differently if the unit were being it.
decommissioned rather than restored, any savings would probably be offset by more stringent NRC decontamination requirements, In its March 1981 PEIS, NRC concluded that an early decision i
on the choice i
to decommission THI-2 would have little effect of alternatives for the cleanup tasks because essentially 3
is required to remove and dispose of 1
the same procedure Because of the importance of a restart the damaged fuel.
versus decommissioning decision, GPU officials said they are continually reviewing the situation.
IS UNABLE TO_ FUND THE CLEANUP COST GPU WITHOUT JEOPARDIZING ITS FINANCIAL VIABILITY The current estimated cleanup is beyond GPU's ability to fund on its own, given its present financial condition.
l in chapter 3, Met Ed operated at a net loss l
As we pointed out in 1980, and the System as a whole is just barely staying No long-term financing funds are available to the companies and even continuation of the short-term funding solvent.
arrangement is questionable.
I 1
l 47
The GPU companies have not been allowed to recover any of their fixed expenses and a return on investment in TMI-2 m
since March 1979, and in TMI-l since the second quarter of All of tne fixed costs are being paid from revenues 1980.
related to non-TMI properties.
In 1980, these unreimbursed The loss of stock-expenditures amounted to $78 million.
earnings and unrecovered depreciation expense on holders' i
the TMI investment amounted to an additional $66 million.
GPU has budgeted about $60 million for 1981 TMI-2 About $20 million is for O&M expenses and expenses.
$40 million for maintaining the plant in a safe condition and some cleanup, including processing the water in tne containment building.
The S40-million expenditure is At this continued largely covered oy insurance payments.
level of expenditure, the insurance payments will run out in late 1983 with much of the work required to clean up As shown previously in table tne reactor remaining undone.
16, about S100 to $150 million per year will be needed to complete the cleanup as scheduled.
The restart of TMI-l in early 1982 will provide some relief to GPU but is likely to have little effect on the availability of cleanup funds.
The 1981 unrecovered cash outlay for O&M expenses will presumably be recovered through Interest costs, revenues once TMI-l is allowed to restart.
preferred stock dividends, depreciation expense and a return The companies, however, on investment will also be recovered.
will not be able to use all of these revenues to help fund The payments being made on the unrecovered cleanup costs.
fixed costs for the TMI units have drained GPU's resources that would normally have gone into maintdnance and con-struction projects to maintain system reliability, improve customer service, and to pay common stock dividends.
The drastic cuts in construction programs necessitated by the loss of earnings on the TMI units cannot continue during the rest of the decade without seriously impairing GPU's ability to maintain reliable and economical customer service.
it will be necessary to allocate most of the Co nseque n tl'/,
restored TMI-l earnings to uses other than for TMI-2 cleanup.
A furcher complicating factor would be a continuation of PaPUC restrictions on tne use of operating revenues to PaPUC has held to the position taken fund cleanup costs.
shortly after the accident that ratepayers should not pay In its June 1979 order, PaPUC stated its for the cleanup.
opposition to ratepayer financing of TMI-2 cleanup.
"The Commission is of the view that none of the costs of responding to the incident including repair, disposal of wastes and decontamination are recoverable from ratepayers."
48
In a September 1980 rate order, PaPUC again reiterated its position against ratepayer financing of TMI-2 cleanup.
PaPUC suggested a possible role of GPU stockholders and the Federal Government when it ordered Met Ed to
" cease and desist from using any operating revenues for uninsured cleanup and restoration costs * * *those cleanup costs and expenditures not covered by insurance ultimately are the responsibility of the company's stockholders and/or the Federal government; however, tney are not the responsibility of the ratepayers."
GPU also has to be concerned about its ability to meet the heavy financial obligation for retiring maturing cond issues over the next 4 years.
As we pointed out in chapter 3, this will require new capital financing of about S400 million.
According to some investment analysts, it is extremely douotful that GPU will be able to access the capital market for these funds as long as financing the cleanup costs remain the sole responsibility of the company and its stockholders.
BANKRUPTCY IS A QUESTIONABLE WAY TO RESOLVE GPU'S FINANCIAL PROBLEMS The PaPUC approval of Met Ed's rate increase request in April 1981, and tne quick bank response to Met Ed's loan request for funds to meet its tax obligation, removed tne near-term possibility that Met Ed would be in def ault and, therefore, subject to possible bankruptcy actions.
A bankruptcy action is still possible, nowever, if GPU is unable to negotiate a satisfactory short-term borrowing agreement with the banks when tne RCA expires Octooer 1, 1981, and Met Ed again finds itself in a cash-short position when its tax payment comes due in April 1982.
If tnat obligation is met, the potential for default again exists I
if one or more of the GPU companies cannot meet bond refinancing obligations beginning in 1983.
l Several studies 1/ have been made of the effects that a
~
bankruptcy action for a GPU company might have on the parties involved--tne company, its ratepayers, its creditors, and even lt the utility industry as a whole.
To assess the possibility that a reorganization of the GPU System under bankruptcy would resolve the financial dif ficulties brought on by the TMI-2 accident, we examined tne various bankruptcy studies i
and discussed possible issues and consequences with knowledg-able people.
Our general conclusion is that there are too lI many uncertainties in a utility bankruptcy to make a strong 4
l'i l
i ll 1/See app. IV for a listing of these studies.
l l
l 49
I case for a bankruptcy action for GPU.
We believe that it is likely that costs to consumers will be higher under a bankruptcy--not only to GPU customers but to customers of many other utility companies.
We also oelieve that a bank-ruptcy will not necessarily resolve the financial issue surrounding TMI-2 cleanup that would most likely precipitate the action.
Tne following brief analysis of what a bankruptcy might entail for a GPU company is intended only to identify, not resolve, issues that would most likely arise.
It is also intended to point out how few precedents are available on whien to make a Judgment that a bankruptcy would be good or bad for GPU.
The nature of a bankruptcy initiation is quite clear In its simplest form, a company that is unable to meet its financial obligations as they come due may oe placed under the supervision of a oankruptcy court.
Just a failure to pay its bills, however, would not automatically render a company bankrupt.
A bankruptcy court would first have to oe petitioned for an " order for relief."
This step can ce taken on a voluntary basis by the company filing a petition or it can be an involuntary action forced upon it by three or more unsecured creditors jointly filing a petition to protect tneir interests.
The nature of the bankruptcy process is to provide for an equitable distrioution of the assets of the bankrupt company among its creditors and for the discharge of the bankrupt company from its debts. Ideally, the process should benefit both the creditors and the debtor.
This can be done through (1) a liquidation of the debtor's estate by filing the petition under Chapter 7 of the Bankruptcy Code or (2) a reorganization of its financial structure under a Chapter 11 filing so that it can continue its business operations.
In a Chapter 7 case, the court-appointed trustee is required to collect the debtor's assets and reduce them to money.
Thus, a Chapter 7 proceeding culminates in a sale of individual assets of the debtor or of the debtor's ousiness as a whole.
The purpose of a Chapter 11 proceeding, on the other hand, is to implement a reorganization plan for the debtor.
This goal implies that the bankrupt firm would continue its business operations during the reorganization proceedings.
50 W I M E 2ln+D S AEhil;5G}]d s & %
is generally assumed that a Met Ed or GPU bankruptcy It would be in the form of a reorganization, ratner tnan a liquidation.
In a situation wnere the debtor supplies tne public with a Dasic necessity, such as electricity, the public's needs would be better satisfied by reorganization.
The remainder of this discussion adopts this assumption and addresses only the Chapter 11 reorganization process.
issues an order for relief, tne interested Once the court parties begin the process of developing a reorganization plan to rehabilitate the debtor.
The commencement and continuation of most proceedings against the deotor and its property are stayed.
Committees representing the various classes of creditors and equity interests may be organized to protect their A trustee may be appointed to administer particular interests.
the estate and the reorganization process although the debtor company could be allowed to manage its own affairs during a reorganization.
Business continues, as normal as possible, out under the court's supervision.
During the reorganization period, the debtor is provided some special relief to allow it to initially suspend payment on its liabilities and protect its assets with a view towards For instance, most reorganizing its financial structure.
tne debtor and its property are automatically actions against stayed and the accrual of post-petition interest on unsecured indebtedness is suspended.
The company, acting under tne is allowed to reject " executory contracts" court's supervision, and avoid preferential transfers.
Creditors, of course, will petition the court seeking adequate protection for their Tnus, a bankruptcy court will be faced witn such interests.
issues as wnether (1) relief can be granted f rom the automatic there is sufficient collateral to pay interest stay, (2) on secured debt, (3) a particular contract is executory, I
(4) certain assets may be sold, or (5) particular expenses i
therefore, entitled to first priority are administrative and, for payment.
Possible effects of a bankru_ptcy difficult to assess is difficult to assess the consequences of bankruptcy It For example, the current bankruptcy for several reasons.
enacted in 1978, 1/ has not been interpreted by the
- law, Furthermore, because no courts to any great extent.
investor-owned electric utility has ever become bankrupt, ii there is little experience in this area on which to draw.95-598.
1/The Bankruptcy Reform Act of 1976, P.L.
51
i Tne nature of the bankruptcy proceeding itself is unpre-dictable, and most decisions made by a bankruptcy judge during a reorganization process are influenced oy the facts peculiar to tne particular situation.
A Met Ed or GPU reorganization would probably be compli-cated.
Not only would a oankruptcy court have to face the normal bankruptcy issues referred to previously, but it would I
also nave to address some rather complex issues related to Met Ed's status as a public utility and member of a holding company, as well as issues related to the TMI cleanup.
Some of the questions that would require answers follow:
--Are tne GPU subsidiaries so linked that a Met Ed bankruptcy would force the other companies into bankruptcy too?
--How will the TMI-2 cleanup costs be treated by the bankruptcy court?
--What is the effect of bankruptcy on the TMI operating agreement?
--What is the effect on the TMI operating license?
--What authority does the bankruptcy court have in relation to the PaPUC?
--How can electrical service to consumers be maintained?
It is unclear whether these would actually become is-sues in a reorganizution proceeding and, if they were raised as issues, how the cocrt would resolve them.
Because of the complexity of the issues, however, what can be predicted with some certainty is ti.ot a Met Ed reorganization would be a long and protracted proceeding involving numerous parties with diverse and often inconsistent interests.
Met Ed, its stockholders, and its creditors would not be the only partici-pants in the proceedings because of the company's status as a regulated public utility and memoer of a holding company.
The Securities and Exchange Commission would play a prominent role in the reorganization as a result of its regulatory responsibilities under the Puolic Utility Holcing Company Act of 1935.
PaPUC, NJBPU, and Federal regulatory agencies might be allowed to participate.
Bankruptcy proceedings could have a disrupting influence on customer service.
Each party would undoubtedly appear before the court to establish its claims and priorities and to intervene in management and operation matters in order to pro-tect its interests.
Efficient operation and management could 52 Ed$AbsY$5@hId M
Decome quite difficult.
The reorganization process would oecome expensive to administer.
The company's access to capital markets and its relations with its suppliers of fuel and materials could be affected.
All of this would undoubtedly translate into a higher cost to tne ratepayer.
Furthermore, tne NRC staff has concluded that bankruptcy would not solve tne problems which initially led to financial distress.
The site of the accident would still have to be decontaminated, and replacement power would still have to be purchased to meet the requirements of the puolic.
Some tentative costs of ginkruptcy_have been identified The uncertainty attached to the actions that could De taken oy the court in a bankruptcy proceeding involving a public utility affects any cost estimate attempting to show the possible conse-quences of a bankruptcy action.
Despite this uncertainty, the Arthur Young and Company bankruptcy study for the NJBPU on Jersey Central, did quantify some of the costs that might ce incurred by the different parties.
The costs of a bankruptcy proceeding for Jersey Central and its customers were estimated to range from $630 to $800 million over a 10-year period.
These costs include the loss of tne relatively low interest rate on out-standing long-term debt and preferred stock, a higher required return on common stock, the increased cost of deot and preferred stock for future capital needs, and the administrative costs of tne legal proceedings.
Since these costs are primarily debt related: comparable costs for Met Ed would likely be somewhat less since the outstanding deot and future capital requirements are lower.
The Arthur Young study also calculated that even under tne most favorable evaluation of bankruptcy using extreme but possible assumptions as to what might happen, costs could reach $190 million over a 10-year period.
The impact of a bankruptcy on financing costs would probably not be limited to the GPU companies.
The expected effect of a utility bankruptcy on the industry as a whole was considered by tne Arthur Young study, although it was not quantified.
In the study, it was pointed out that a Dankruptcy of Jersey Central (and a similar effect could be presumed for any other utility) would likely introauce new, precedent-setting concepts of risk bearing and return on investment in the utility industry.
The risk premium applied to the bankrupt company was viewed as possioly spreading to other utility companies across the country.
The study concluded that if tne risk premium is spread broadly enough, even a small differential in financing costs would have large overall costs to consumers of electric power.
s 53
l A recent study on the effects of a utility bankruptcy by a major securities firm, 1/ however, did quantify the additional financing costs resulting from the TMI accident.
The authors of the study concluded that THI-related issues have already added a risk premium of about 0.75 percent to recent utility debt and equity offerings.
They estimated that based on 1981 new-debt and equity-financing needs of
$16 billion, ratepayers will be paying about $170 million more each year in added financing costs.
The study postu-lates that given a GPU bankruptcy, at least an additional 1-percent risk premium would likely be assessed for utility financing, raising the total costs to ratepayers across the country by $400 million annually.
Some short-term benefits are possible but uncertain The potential costs of a bankruptcy need to be considered in light of possible benefits that might accrue if a bank-ruptcy action were an option.
Our assessment of such possible benefits tends to parallel the Arthur Young observation that although some benefits might emerge from a bankruptcy action, they could probably be achieved without such action at less cost and with greater efficiency.
It is important to note that the value that might be assigned to both advantages and disadvantages of bankruptcy is a reflection of the views of each interested party.
The Arthur Young study concluded that these views are dependent upon and influenced by factors such as judgments about equity, potential fault, public policy, and social values.
A proposed benefit from a bankruptcy could be an improve-ment in the company's short-term cash flow.
As we indicated earlier, one of the first steps the court would take would be to stop the payment of pre-petition debts and suspend interest and dividend payments.
Met Ed, for example, paid over $57 million for interest charges in 1980 and had current liabilities on the books of nearly $129 million at year end.
However, the court could decide that the company should con-tinue paying interest to secured debt holders, and the cash availability would be reduced accordingly.
The Arthur Young study also pointed out that the transfer of costs from rate-payers to creditors if interest payments were suspended does not create a new economic savings.
Any short-term advantage will likely result in additional costs to future ratepayers as investor's required higher risk premium and bankruptcy costs are recovered through rates.
1/M. Anthony May and Evan J. Silverstein, "The Electric Utility Industry:
The Cost of Capital Ef fects of the Three Mile Island Mishaps," L.F. Rothschild, Unterberg, Towbin, New York, Apr. 8, 1981.
l 54
- ~msw2:=Mns~M*P_'M=%MnM5
Other benefits that might result from a bankruptcy include (1) a new atmosphere for change that might trigger financial assistance from State or Federal legislators or regulatory agencies and (2) an opportunity for the orderly treatment of y
the TMI accident costs.
Whether these or other perceived advantages would actually materialize in a bankruptcy action is as uncertain as what the actual costs might be.
CONCLUSIONS GPU cleanup cost estimates have been plagued by the lack of accurate data on the conditions inside the TMI-2 contain-ment building and the uncertainty of the regulatory environment within which the cleanup activities will take place.
The $150-million cost increase that resulted when the restart was slipped by 2 years indicates the economic impact of extending the cleanup schedule.
We do not believe that the lack of immediate danger to the public health and safety from TMI-2 in its present state should We result in unwarranted delays in expediting the cleanup.
believe that the continued financial viability of the GPU System is vital in maintaining reliable service to Pennsylvania and New Jersey customers and that a dissolution of that viability is entirely possible until the cleanup cost issue is resolved.
Bankruptcy is always an option for a company that is facing severe financial distress, and there appears to be a potential for some short-term advantages to a GPU bankruptcy.
We believe, however, that the uncertainty that surrounds the treatment of the issues by a bankruptcy court, the poten-tially adverse effects on the utility industry as a whole, and the ultimate longer-range cost impact on GPU consumers, the lack of assurance that a bankruptcy would resolve the cleanup issue tend to disfavor bankruptcy as a means of alleviating GPU's financial problems.
55
l CHAPTER _ __5 SEVERAL OPTIONS EXIST __F_OI3 FINANCING THE TMI-2 CL_EA_NUP__C.OSTS The inability of GPU to finance Doth the cleanup costs and maintain system reliability requires that most of the necessary funding for TMI-2 cleanup will have to come from sources other than GPU System earnings.
Numerous options have been proposed for providing cleanup cost support.
These include (1) new ownership for the TMI units, (2) nuclear fuel enrichment surcharge, (3) mandated insurance assessment, (4) GPU funding from increased rate revenue and stocknolder earnings (5) Federal research and development assist-ance, and (6) industry contributions.
Tnese options are not all inclusive and each one has its limitations, some more so than others.
We believe, however, that they represent a cross-section of the kinds of solutions that are being proposed for funding the cleanup costs.
Within the options listed, we further believe that a combination of options rather tnan any single one has the greatest chance for successfully resolving the cleanup funding problem.
One such combination would be a sharing of the costs among ratepayers, the Federal and State.
Governments, and the utility industry.
GPU is currently working to develop such a shared funding arrangement and DOE officials have developed a proposed research and development program at the Federal level.
The successful resolution of the current impasse over cleanup funding will require leadership which we believe rests with State officials in Pennsylvania and New Jersey.
State officials should, in our view, take the lead in working with GPU and the utility industry in reaching agreement on a method for resolving the TMI-2 cleanup effort.
It will also take close cooperation and support of the Federal sector, however, since the key cleanup components of public nealth and safety and high-level radioactive waste management and disposal are within its purview.
In our opinion, the Federal sector response to requests for cooperation and assistance in con-nection with the TMI-2 cleanup will greatly influence the future direction of nuclear power as a commercial undertaking.
The following is our analysis of the six options enumer-ated above.
Probably no one option satisfies the concerns of all the potential contributors and some appear to have serious implementation problems.
NEW OWNERSHIP OF TMI UNITS The takeover of the TMI units by a non-GPU entity has Deen discussed in several studies as a means of relieving the GPU companies from the cleanup cost burden.
A numoer of possible 56 w
Q R y'M T [ 'i 3 QB? M E W W 2X 2 W new owners nave been considered, including State public power authorities, tne Federal Government, and another utility company.
It is obvious that the units would have to be sold as a package, since the new owner would assume the TMI-2 cleanup responsibility and TMI-l revenues would provide a source of funds for that purpose.
The Edison Electric Institute (EEI) Task Force on Nuclear Institutional Issues estimated that perhaps S80 to S100 million annually could be raised for cleanup purposes by selling TMI-l energy at some-The tning less than GPU's current replacement power cost.
ratepayers would not get the full oenefit of the lower-cost TMI-l energy but would get some reduction in current power Increases in operation and maintenance expense and costs.
capital cost since the EEI study, however, could reduce the expected level of cleanup funds available from TMI-l energy sales to S30 to $40 million.
In certain instances, either a lease or sale of the TMI units could become mired in regulatory red tape and Jurisdic-tional issues which raises questions as to the practicality of the option.
The transfer of ownersnip would involve botn PaPUC and NJBPU at the State level.
While the State regulatory agencies would lose control over any rates charged for whole-sale energy sales from TMI-1, PaPUC would probably exercise jurisdiction over any ownership transactions involving tne TMI units because of their location.
FERC would assume rate jurisdiction over future wholesale energy sales f rom TMI-l but would not have to approve any sale or lease arrangement.
At the Federal level, such approval would involve the SEC under the provisions of the Puolic Utility Holding Company Act.
Any transfer of authority for nuclear plant operations would of necessity involve NRC.
Met Ed is the licensed operator and it would be necessary for the new owner to assure NRC that it could safely operate the facility.
This could l
i involve an extensive licensing review by NRC.
The biggest problem of course, is finding a willing buyer.
The uncertainties attached to obtaining all the necessary approvals would certainly affect any evaluation of TMI as an i.1 vestment opportanity.
Our discussion with PaPUC indicated that a takeover of TMI that would exclude State jurisdiction over setting rates would not be viewed in a favorable light.
and The creation of a public power authority to buy TMI it is not a new idea, out it would prooably take operate considerable time to accomplish since Pennsylvania has no It is also provision for such an authority at this time.
questionable whether the State would back the neccssary bond issues needed to raise sufficient capital to buy the units.
57
NUCLEAR FUEL ENRICHMENT SURCHARGE The EEI task force considered as one option increasing the price DOE charges domestic utilities for enrichment of their uranium fuel.
The task force estimated that a 12-percent increase in cost would provide about $100 million annually, which would be used to provide cleanup funding.
It was assumed that since the increase would relate directly to fuel costs, it could probably be passed on to consumers as part of fuel adjustment clauses without formal public utility commission hearings.
Such a procedure would also limit participation in the cleanup to utilities operating nuclear facilities.
There are a number of problems with the concept, however, that would have to be overcome.
The present statute requires that enrichment charges by DOE are to be based on the cost of the service rendered. 1/
The statute is probably not broad enough to allow a surcharge for another purpose, such as TMI-2 cleanup.
It is not clear that additional legislation could be passed that would permit DOE to make such an assessment.
If the surcharge were applied to all customers, foreign utilities who presently receive enrichment services would also be required to bear part of the cleanup cost burden since DOE cannot discriminate in the prices charged to its customers.
This could involve serious contractual and diplomatic consider-ations.
We believe the assumption that an enrichment surcharge
(
could be passed to customers through a fuel adjustment clause without a hearing is probably invalid.
Opposition to using l
the surcharge for TMI-2 cleanup costs could easily result in regulatory determinations that such costs would not be allowed in the fuel costs.
The timing of the cost recovery would pose a major hurdle for some utilities.
The task force report pointed out that while most of the fuel enrichment surcharge would ultimately be recovered through higher fuel adjustment charges, such recoveries wcP13 not occur until the fuel is actually used to produce ' lo tricity.
This could be as long as 5 years.
The uti' M) tr,fustry, therefore, could be faced with financing as muci 4 -> 71*
million by 1986, before the added enrichment charges ;oute iegin to be recovered.
1/ Atomic Energy Act of 1954, CCetion 161.
1 l
58 1Iw
T y y [ L J ;;: w. a n. w L a w MANDATED INSURANCE AS_SE_S3 MENT GPU had the TMI units insured for $300 million--tne maximum amount available at the time of the accident.
The insurance proceeds are being used to cover most of the cleanup costs incurred, but they will not ce sufficient to cover all expected costs.
Current estimates indicate that about $600 million in excess of insurance proceeds will De requirea over to 6 years to completely decontaminate the unit.
the next 4 One option for meeting at least part of the estimated financial needs is to require an increased level of property insurance coverage for each nuclear reactor beyond what is presently available.
Part of the premium proceeds would then De used to help pay cleanup costs.
The argument advanced in support of this option is that if the utility industry had been aware of the magnitude of the financial exposure resulting from a TMI-type accident prior to March 28, 1979, it would have taken the appropriate steps to increase the existing property insurance levels above the $300-million limit.
Payments from insurance premiums for the TMI-2 cleanup are viewed as a corrective action to rectify a prior failure to adequately assess tne real need for insurance coverage.
Payments for damage from an insurance fund established in the after an accident has occurred has some precedent We think insurance industry and/or Federal government programs.
it highly unlikely, however, that the insurance industry woula voluntarily enter into such a program for the TMI-2 cleanup.
there is some question as to whether public Furthermore, utility commissions would allow their jurisdictional utility companies to collect any insurance premiums marked for TMI-2 cleanup costs in their rates.
l Total participation by utilities with nuclear generation I
would be required to make the program work.
Such participation could oe obtained either as a condition of a NRC operating license or through specific legislation. 1/
State utility commission approval of an expense passthrough to utility company consumers is also viewed as more likely if participation is mandatory.
l i
Two separate legislative bills proposing an expanded
.asurance/TMI-2 cleanup cost coverage have been introduced in House bill 2512 2/ proposes the establishment the Congress.
of a quasi-governmental National Property Insurance Corporation with mandatory participation by all utilities having nuclear In essence, annual premiums would go to pay 75 facilities.
1/If court challenges to such a retroactive insurance scheme its usefulness as an option may be impaired.
- occur, 2/ Nuclear Powerplant Property Damage Insurance Act of 1931.
59
1 percent of the non-insured TMI-2 cleanup costs if certain conditions regarding the financial integrity of GPU or its successor were agreed to by PaPUC and NJBPU.
Under the clean-up assistance plan, one-half of the insurance contributions would be a grant, with the remaining half repayable oy GPU over some future time period through an increased premium.
Senate bill 1226 1/ is essentially the same as H.R. 2512.
It, too, proposes mandatory participation in an insurance pro-gram, with payments to GPU for cleanup from insurance premium proceeds.
The principal difference between the House and Senate bills is that S. 1226 would provide all insurance pro-ceeds on a grant basis to GPU--no repayment is envisioned.
The bills have two predominate features.
They propose (1) a specific course of action regarding resolution of tne TMI-2 cleanup issue and the underinsured status of utilities with nuclear units, and (2) that funds for TMI-2 cleanup and any future nuclear accidents will be derived from those who benefit from the use of nuclear power.
1 Criticisms of the bills from some electric utility and insurance company representatives include (1) industry con-cern about the open-ended obligation of future utility contri-butions, (2) a reluctance ny private sector interests to have the Government involved in an insurance program when all private alternatives for providing such insurance have not oeen exhausted, and (3) a feeling that TMI-2 cleanup and increased insurance coverage should be addressed separately.
Proponents of the bills, however, point out that ceilings on utility contributions can easily be set by amending the pro-posed legislation.
They also point out that the mills do not exclude utility or private insurance industry participation in providing insurance but only provide an additional layer of insurance to increase the total coverage available.
GPU FUNDING FROM INCREASED RATE REVENUES AND STOCKHOLDER EARNINGS Revenue received from customers for electric service provides the only significant source of funds for the GPU companies.
As we pointed out in chapter 3, the revenue is used to pay the normal expenses of the companies and provide an equitable rate of return on stockholders' equity.
The level of revenue collections allowed is set by PaPUC and NJBPU within the general guidelines that rates must be just and reasonable and that the companies are allowed to earn a fair return on their investment.
1/ Nuclear Powerplant Property Damage Insurance Act of 1981.
60
1 i
One of the options for funding TMI-2 cleanup is to pass these costs through to the companies' ratepayers.
This could be done either as a direct surcharge which would be specifically marked for cleanup expenses or through some other ratemaking methodology that would indirectly provide the companies suffi-revenue to meet normal expenses and earnings plus enough cient additional funds to pay for the cleanup.
PaPUC has not been allowing any operating revenues collected Furthermore, TMI-Ed to be used for cleanup expenses.
by Met related rate increases granted by PaPUC and NJBPU since the accident generally have only been sufficient to cover replace-1981, order by PaPUC ment energy costs, although the April 9, included $16.5 million annually for TMI-l restart-related These increases were operations and maintenance expenses.
Several rate more than offset by reductions in base rates.
increases have been granted to Jersey Central for non-TMI expenses in addition to those that are TMI-related.
The ef fect of the PaPUC and NJBPU rate orders has been to increase average residential costs per kWh for GPU System customers at only a slightly higher rate than those experienced New by customers of neighboring utilities in Pennsylvania, This is illustrated in table 17.
Jersey, and New York.
Over the 1976-80 period, for example, the kWh costs for the GPU System's residential customers increased at an average annual rate of 11.13 percent while kWh costs for non-System increased an average of 10.66 percent annually.
1979-80 customers GPU System costs increased at a higher rate for the post-accident period, however, with the average annual increase amounting to 14.5 percent as compared to an 11.13 percent l
l for non-GPU companies during the same period.
annual increase In addition to comparing average residential costs on a costs can also be compared based on the percent per kWh basis, of disposable household income used to pay for electric energy.
,ercentages were computed for 1980 actual energy costs.
The computations show that energy costs as a percent of dis-Thesc from 1.4 percent to 2.6 percent with the posable income ranged GPU System customers in the mid to upper range--1.8 to 2.2 percent.
GPU has estimated that it would need about $150 million per year to expeditiously move forward with cleaning up TMI-2.
We estimated what GPU System rates would have been in 1980 if the entire $150 million had been financed through consumer rates.
Table 18, page 63, shows the effect of this additional for the three GPU companies, cost on average residential rates of disposal household income used for electric the percent and the relative ranking of the companies with energy costs, the cleanup costs added to their rates.
61
r Table 17 Averace Besidential Electric Rates for Met Ed, Ibnelec, and Jersey Central Customers Conpared to Customer Rates for Neighboring Utilities (1976-80) 1980 1978 Utility C/km Utility C/kWh Utility C/kWh 1976 Consolidated Edison Co.
Consolidated Edison Co.
of New York, Inc.
8.78 of New York, Inc.
9.60 of New York, Inc.
11.82 Consolidated Edison Co.
Orange and Bockland Orange and Rockland Utilities, Inc.
6.76 Utilities, Inc.
8.10 Utilities, Inc.
9.84 Orange and Rockland Public Service Electric Public Service Electric Public Service Electric 6.60 and Gas Co.
8.42 5.75 and Gas Co.
Iong Island Lighting Co.
5.19 Long Island Lighting Co.
6.27 Iong Island Lighting Co.
8.42 and Gas Co.
5.92 Central Hudson Gas and Philadel@ ia Electric Co.
4.94 Duquesne Light Cc.
Electric Corp.
7.72 Atlantic City Electric Co.
4.78 Jersey Central Power and 5.54 Jersey Central Power Light Co.
7.55 Central Hudson Gas and Philadelphia Electric Co.
5.50 and Light Co.
4.75 Philadel@ia Electric Co. 7.33 C
Electric Corp.
Central Hudson Gas and 4.73 6.83 Duquesne Light Co.
Electric Corp.
5.43 Duquesne Light Co.
L 4.71 Atlantic City Electric Co.
5.11 Atlantic City Electric Co. 6.71 Jersey Central Power and 4.65 Metropolitan Edison Co.
6.21 Light Co.
4.47 Metropolitan Edison Co.
- (
Pennsylvania Electric Co.
4.10 Pennsylvania Electric Co.
4.70 Pennsylvania Power Co.
5.94 E
Metropolitan Edison Co.
Pennsylvania Electric Co. 5.51 4.08 New York State Electric New York State Electric if Pennsylvania Power Co.
4.59 and Gas Corp.
4.56 and Gas Corp.
5.25 Bochester Gas and l
3.80 Pennsylvania Power Co.
Rochester Gas and Electric Corp.
Rochester Gas and i
3.71 Electric Corp.
4.28 Electric Corp.
5.09
(
New York State Electric Niagara Mohawk Power and Gas Corp.
Pennsylvania Power and 4.88 i
?
Pennsylvania Power and 4.06 and Light Co.
3.55 Light Co.
Light Co.
West Penn Power Co.
3.99 Pennsylvania Power and 4.34 g
Niagara Mohawk Power Light Co.
3.41 Niagara Mohawk Power 3.95 West Penn Power Co.
4.12 p
and Light Co.
1 West Penn Power Co.
3.27 and Light Co.
Statistics of Privately Owned Electric Utilities in the United States, 4'
h 1976 and 1978 and FPC Form 1,1980, Source:
d n
!1 j NI n
Q q g NNWU-"
h Table 18 and Jersey Central Estimated Met Ed, Penelec, Rates for 1980 With TMI-2 Cleanup Residential Surcharge of $150 Million Per Year Percent of Disposable Household C/kWh Income Utility Inc.
11.82 1.4 Consolidated Edison Co. of New York, 9.84 2.6 Orange and Rockland Utilities, Inc.
8.42 1.5 Public Service Electric and Gas Co.
8.42 1.9 Long Island Lighting Co.
7.8 2.3]
Power and Light Co.
7.72 2.4 (Jersey CentralCentral Hudson Gas and Electric Corp.7.33 2.0 Philadelphia Electric Co.
7.2 2.3]
l Metropolitan Edison Co.
6.83 1.5 Duquesne Light Co.
6.71 2.3 l
Atlantic City Electric Co.
5.94 2.3 Pennsylvania Power Co.
5.8 1.9l l Pennsylvania Electric Co.
5.25 2.0 l
New York State Electric and Gas Corp.
5.09 1.5 I
Rochester Gas and Electric Corp.
4.88 1.6 Niagara Mohawk Power and Light Co.
4.34 1.7 Pennsylvania Power and Light Co.
4.12 1.8 West Penn Power Co.
The addition of the cleanup charge would have had the greatest impact on Met Ed customers since it has a 50-percent share of TMI The 1.0 cent per kWh increase would raise the actual 1980 average i
residential rates by 15.5 percent and would increase electr c l
energy's share of disposable income by 0.3 to 2.3 percent--equa i
ther to or below the percentages experienced by customers of f ve oPenelec's an utilities including Jersey Central. rates would increase 6 percent and 3.8 pe ively, while the disposable income percentage would increase by customers' 0.1 percent for customers of each utility.
The resolution of the cleanup funding issue is important, its so is the continued ability of the GPU System to meetAs we pointed out construction and long-term financing needs.3, the present rate reve but i
the companies to meet these commitments on a continuing bas s in chapter il and present a favorable investment opportunity for the fina for cleanup costs will be required to maintain the system'sTo de community.
financial viability.
ld we also estimated what the companies' residential rates wouif the have been in 1980 The been allowed to remain in the companies' base rates.
63
inclusion of these costs would have added an additional 0.9 cent per kWh for Met Ed, 0.3 cent per kWh for Penelec, and 0.3 cent per kWh for Jersey Central ratepayers.
The following table illustrates what the relative residential rates would have been in 1980 if GPU ratepayers had financed $150 million of cleanup costs and the TMI units had been left in the base rates. 1/
Table 19 Estimated Met Ed, Penelec, and Jersey Central Residential Rates For 1980 With TMI Units in Base Rates and a TMI-2 Cleanup Surcharge Percent of disposable household Utility C/kWh income Consolidated Edison Company of New York, Inc.
11.82 1.4 9.84 2.6 Orange and Rockland Utilities, Inc.
8.42 1.5 Public Service Electric and Gas Co.
8.42 1.9 Long Island Lighting Company Jersey Central Power and Light Co.
8.2 2.4 8.0 2.6 Metropolitan Edison Co.
7.72 2.4 Central Hudson Gas and Electric Corp.
7.33 2.0 Philadelphia Electric Co.
6.83 1.5 Duquesne Light Co.
6.71 2.3 Atlantic City Electric Co.
6.1 2.0l l Pennsylvania Electric Co.
5.94 2.3 Pennsylvania Power Co.
New York State Electric and Gas Corp.
5.25 2.0 5.09 1.5 Rochester Gas and Electric Corp.
4.88 1.6 Niagara Mohawk Power Corp.
4.34 1.7 Pennsylvania Power and Light Co.
4.12 1.8 West Penn Power Co.
The addition of these two costs would increase rates to Met Ed customers by 29.3 percent over 1980 actual costs.
Penelec's and Jersey Central's residential rates would increase by 11.5 percent and 8.2 percent respectively.
In addition, the added costs would increase the share of disposable income paid by Met Ed customers for electricity to 2.6 percent--the highest level for all 17 companies.
Estimating future consumer rates is speculative.
While table 19 indicates the relative consumer rates in 1980 under We 1/These consumer rates included State and local taxes.
recalculated consumer rates after subtracting these taxes to determine the net cost of electricity.
We found no significant shifts in Met Ed, Penelec, and Jersey Central relative rates.
64
- zu6Mk&8AMSf6h
the conditions described, we cannot make any definitive state-All major GPU and non-GPU future consumer rates.
ments about New Jersey, and New York utilities have increased Pennsylvania, their consumer rates from 1976-80 by an average of about 11 We would expect this general trend to percent per year.There are a number of factors that will specifically continue.
affect future GPU consumer rates.
Among these are which will ultimately reduce annual
--TMI-l restart, purchased power costs;
--expiration of the deferred energy cost surcharges;
--continued availability of economical coal-based energy purchases;
--completion of a major purchase agreement; and
--completion of the TMI-2 cleanup;
--increased rates for non-TMI attrition, restoration of O&M to normal levels, improved return on share-holders' equity, Forked River amortization, and increased "non-TMI" energy costs.
The key parties in exercising this option are PaPUC and and it is PaPUC has made its position quite clear, NJBPU.
this time how much of a rate increase related questionable at Jersey to cleanup costs would be allowed in Pennsylvania.
Central serves an area heavily populated by fixed income NJBPU officials residents with electrically heated homes.
in rates have already imposed a heavy said that past increases burden on these customers; hence, they are reluctant to add Cleanup costs could any more to their customer's cost of energy.
Such be passed through to industrial and commercial users only.
a regulatory decision might have serious economic repercussions, particularly if the affected consumers have high energy
- however, consumption.
A possible offset to some of the consumer cost increases described above is to divert part of GPU's earnings to the cleanup effort.
As we pointed out in Chapter 3, stockholders have already lost over $150 million in earnings as a result of the accident and no dividends have been paid for the last six For future financing purposes, it is important that quarters. common stock dividend payments be restored as rapidly as possible.
that the restoration of GPU's earning capacity It may be, however, should be accompanied by a contribution from the stockholders to I
the cleanup funds.
In 1978, for example, the GPU System had a consolidated net income of over $138 million and paid common 65 f
stock dividends of over $106 million.
If GPU's earnings level were restored to comparable levels, some portion of the dividend payments could conceivably be withheld from stockholders and used to reduce the consumer's share of cleanup costs.
Support for this part of the option may be derived from the benefits that would probably accrue to the stockholders if the cleanup funding problems were quickly resolved.
In addition, since GPU's customers would be paying higher electric rates as a result of the accident, an argument can be made that the stock-holder should also bear a portion of the cleanup burden.
Allocating a portion of the stockholders' earnings for cleanup costs, however, could adversely affect the utility and the entire industry.
GPU investors may perceive that the contribution already made is sufficient, especially in view of the fact that while consumer rates have increased, GPU has not been allowed by PaPUC to apply any of the proceeds to the cleanup.
The attractiveness of continued investment in GPU, or other electric utilities, would probably be diminished because investors may perceive that the risks of utility investment outweigh the benefits due to the future uncertain nature of predictable dividend payments that existed prior to the acci-dent.
FEDERAL RESEARCH AND DEVELOPMENT ASSISTANCE There are several aspects of the accident at TMI-2 which have been offered as justification for Federal research and development assistance in cleaning up the facility.
The option is supported by the dual premise that (1) the Federal Govern-ment initiated and supported the shift to nuclear generation at electric utilities and (2) additional information can be obtained from TMI-2 which may advance reactor operating and waste disposal knowledge.
Information gained from a research and development program could also reduce the cost of nuclear power by improving the safety and reliability of nuclear units and helping to reduce recovery time from any future accidents.
Since 1980, DOE has been working with GPU, NRC, and the Electric Power Research Institute (EPRI) on a program to obtain and exchange information on the technical aspects of the accident results.
This program has been of value in providing limited information on the effects of a nuclear accident on reactor and containment building equipment, the distribution of radio-active contamination resulting from an accident, and the treatment and elimination of accident wastes.
Because of the data-gathering nature of the program, DOE's research and development activities have not directly contributed to the cleanup.
Appropriations for this initial program amounted to
$10.5 million in fiscal years 1980 and 1981 with $10 million requested for fiscal year 1982.
66 m
DOE has also proposed a fiscal year 1982 research and development program of $27 million (possibly costing $75 million for TMI-2 to (1) develop tooling and methods to over 3 years) inspect the damaged reactor core before removing the pressure remove the core and perform examinationsand (3) perform vessel head, (2) and after removal of the fuel, immobilization experiments on the radioactive wastes captured prior to, during, in processing the containment building water to demonstrateh the feasibility of various techniques for immobilizing suc This program is in addition to the on-going data $10 acquisition program which will continue at approximatelyExcept for wastes.
million each year.
include shipping or disposing of the the DOE plan does not Although TMI-2 does not appear to pose DOE is anxious damaged reactor core.
a current public health and safety hazard, i
d can to begin this program so that the information obta ne be generically applied to current and future NRC licensing for criteria as well as contributing to existing technology from processing and disposing of the unique wastes resultingNew engine obtained from accelerated access to the core could jus reactor accidents.
ft l
of nuclear powerplant operations.
Even though there has been fairly widespread supp d development tion officials have stated that Federal research anassistanc this l
tation I
of a comprehensive cleanup program.should be the participation of oth interest in the TMI-2 cleanup such as GPU,and other cognizant Fed
- industry, organizations.
Implementation of the expanded DOE research and development d research program will require GPU participation in the proposeand congressional appro NRC cooperation, and development effort, is expected that DOE's research It of the required DOE funding.
least a 3-year period and and development program will cover at ill depend that the successful acquisition of the research data wlarg f
GPU also needs to know reactor disassembly and core removal.
to a i
tion in the with some certainty the extent of DOE's partic pathe administration Consequently, ch and reactor disassembly effort.
and the Congress need to consider approving a total resea for other interested parties of the Federal Government's concern development in the TMI-2 cleanup project.
and interest DOE participation in a research and development program f the at TMI will require congressional budget approval oThe current emphasis ifficult proposed funding levels. Federal expenditures may make the required 67
to attain.
In addition to the budget restrictions, some
,I cojections have arisen to any Federal funding of the TMI-2 cleanup effort regardless of the purpose for which it would be used.
Resistance to Federal participation in the cleanup is also based on the assessment that current technology is sufficient to perform all cleanup activities and tnat DOE would actually gain little new knowledge by its participation.
j INDUSTRY COdfRIBUTION The electric utility industry is perhaps the most affected oy the accident at TMI on a long-term basis.
As indicated in chapter 4, some increase in the cost of borrowing money for construction and other projects has already been attriouted to the accident at TMI-2.
As long as the cleanup problems continue to be unresolved, high financing costs are likely to continue or increase over time and contribute to the upward spiral of utility rates.
Greater industry participation may also help reduce the threat of bankruptcy to GPO.
If the company is forced into bankruptcy, not only are higher costs for utility borrowing likely to result but the actual supply of funds may be severely limited.
Another incentive for industry participation in the cleanup is the potential to assist in tne development of a more effective and precise regulatory program which could be applied to future accidents and reduce generating plant recovery timeframes.
This option, therefore, proposes that it is in the best interest of tne industry and its customers to clean up the damaged facility.
The industry commitment can be in the form of monetary contrioutions, personnel assistance, or noth.
An aggressive industry role in the TMI-2 cleanup could produce long-term benefits.
Investment in electric utilities has typically been viewed by investors as a safe and reliaole avenue for receiving dividend returns.
The accident at TMI and its related regulatory uncertainty, however, seem to have been the most dramatic in a series of events that have oeen working to shake investor confidence in electric utilities--
especially those owning nuclear facilities.
Consequently, electric utilities in need of financing are finding that their access to the capital market is being restricted.
The continua-tion of the TMI-2 dilemma will probably only work to exacercate the difficulty for the industry in obtaining financing.
If GPO is forced into reorganization, the ripple effect on capital access for the industry will prooably create new and costly precedents in the interest rates electric utilities will ne forced to accept to finance construction projects.
It is less likely that long-term negative after shocks of TMI-2 will exist in the capital market if the industry is aggressive in quickly responding to the needs of GPU.
Electric utility industry response to the problems at TMI can come in several forms.
Greater assistance could De directed a
68 M M M EELM Wi-N;DE M
--l N
'I through EPRI, the industry's research organization.
While a reprogramming of its projects to assist TMI may require the approval of State utility commissions, such realignment of EPRI's activities can be could occur at other member utilities. justified because similar accidents from the EPRI involvement The knowledge gained and could accelerate other utilities'at TMI-2 could have direct application recovery process.
Another avenue for greater industry participation is the contribution of personnel to assist in the decontamination process.
Again, this may require State utility commission approval, but could result perienced industry personnel.in a nucleus of trained and ex-the cleanup process could also be used as a foundation forThe training received industry development of more effective procedures for working in high-radiation areas for eventual decontamination of retired nuclear generating plants.
in the resolution of TMI-2 is the potential to assist in theA possibl development of more precise and effective regulatory programs for accident recovery.
There is general agreement that the TMI-2 cleanup process can be accomplished using existing technology. Although available, considerable debate has occurred l
on the application--and therefore the regulation--of this technology.
Besides the direct knowledge obtained from the cleanup operation, industry participation in accident recovery programs could assist reducing regulatory timeframes.in upgrading recovery procedures and Such a program would in turn i
benefit utility customers who would have to rely on expensive j
purchased power for a shorter period of time in the event of an accident at another utility.
Edison Electric industry participation at TMIInstitute task force officials said that 6
does not of fer the degree of assistance needed by GPU.on a voluntary basis probabl believe that They State utility commissions responsible for the s
actions of the contributing utilities will be reluctant to
[
t l
allow them to voluntarily expend financial and personnel i
resources to assist GPU at Shareholders may also objectthe expense of their customers.
i f
to a utility making a contribu-L}.
tion from its earnings.
Industry support for the cleanup effort may, therefore, be limited unless some agreement is
- j reached with State utility commissions or a retroactive p3 insurance agreement can be negotiated.
pt
{}s CONCLUSIONS y1 There is little question that i
the first major nuclear accident has had a significant impact on a large utility j
system, its customers, both State and Federal regulatory j
agencies, and the electric utility industry as a whole.
Jh
-d h
We believe that the decisions made in the resolution--or lack of resolution--of the problems facing GPU will strongly influence the future course of the nuclear industry as it relates to the commercial production of electric power.
The resolution of GPU's problems wi'll require the although some application of more than one single option,of the individual options describe potential for funding all or a major part of the cleanup For example, it is possible to impose the entire effort.
financial burden of cleanup and maintaining the GPU System The large increase over pre-accident on the ratepayers.
rate levels, both in real terms and in relation to increases on other utility systems, however, could result in numerous An argument can be made that System customers hardship cases.
have benefitted from the low rates in the past--due in part to GPU's use of nuclear power--and that TMI-2 is a SystemIn cost that rightfully should be horne by its customers.in particular, addition, Pennsylvania residents and businesses, will reap the major economic benefits of having the cleanup issue expeditiously resolved.
On the other hand, there are benefits to be gained by both customers of other utilities with nuclear reactors and the Federal Government that would warrant some assistance to GPU for the TMI-2 cleanup.
The increased risk investors have attached to utility financing has already raised the A failure to successfully resolve cost of borrowed money.
the cleanup issues could add to this cost, particularly if The increased there were a bankruptcy on the GPU System.
not costs for long-term capital are passed on to ratepayers, only in the short run but over the life of all securities bearing the higher interest rates.
The Federal sector's involvement in nuclear power through the mandated regulatory activities of NRC and the responsibili-ties for the development of nuclear power and high level radio-active waste management of DOE, would also justify some Federal It is apparent participation on the TMI-2 cleanup activities.
that the uncertainties surrounding the accident, particularly as they relate to the conditions inside the reactor itself, have made it more difficult for NRC to provide the positive, active guidance needed by GPU to move forward with the cleanup Much can be learned about what actually occurred that will be beneficial in improving the regulatory require-process.
ments regarding design and construction of similar nuclear facilities.
The argument can be made that the cleanup is a matter of concern to only the private sector and that the marketWhile place should decide the value of cleaning up the unit.
it must also be remembered that perhaps valid in many caces, the nuclear power industry is heavily regulated and that GPU 70
.t.
sw-cannot make unilateral decisions that would then be judged by market-place responses.
Consequently, we believe that along with the responsibility on the part of the GPU System customer and the utility industry to participate in funding the cleanup, The there is also a Federal responsibility to participate.
Federal sector needs to use its regulatory authority in a manner that would expedite the cleanup while protecting the public health and safety, and participate in such a way as to enhance its knowledge of reactor accident consequences for providing guidance in future accident recovery efforts if necessary.
We believe a shared approach to funding the TMI-2 cleanup is fundamental and several potential participants have indicated a willingness to contribute to the project.
DOE, as discussed previously, has proposed an expanded research and development effort at TMI-2 for at least one year.
The utility industry, through EEI, has indicated a willingness to participate in the On July 9, 1981, the Governor of Pennsylvania cleanup effort.
proposed a $760 million comprehensive cost-sharing plan for the TMI-2 cleanup, including the establishment of a " National Energy Research Institute" as a channel through which financial and technological support would be funneled. 1/
We believe that the primary leadership role in resolving the current impasse rightfully rests with State of ficials--
the Governors of Pennsylvania and New Jersey, State legislators, and the PaPUC and NJBPU.
These State of ficials, however, will need the cooperation and support of the utility industry and regulators and the appropriate Federal entities--notably its NRC in its Federally-mandated role over nuclear facilities, and DOE, with its research and development capability and responsibility for high level nuclear waste handling and disposal.
The significance of nuclear power as a key element in the country's electric power generating mix over the next two decades would appear to justify the Federal sector's con-tinued demonstration of interest in nuclear power by actively providing the support needed by the States in resolving the TMI-2 problems.
The extensive studies performed on the TMI problems and issues should provide an adequate basis for a frank and open dialogue between GPU, industry representatives, l
and State and Federal regulatory agency officials on ways and means to resolve the cleanup impasse.
If the States accept the leadership role in resolving the cleanup issues, the Federal sector's cooperation could be provided in at least two ways.
One way is for NRC to use its 1/The proposed $760 million cleanup cost covers the period 1982-87.
It also includes O&M costs for the period, a cost element not included in the cleanup estimate shown on page 44 of our report.
71
available authority to provide a regulatory environment tnat would expedite the cleanup effort and thereby reduce overall Costs.
A complementary approach is to fund a multi-year research and development program using DOE's 3-year, $75 million projec-i tion as a benchmark for Federal involvement.
The tasks performed during the program would provide important research and develop-
+
ment data with some offset to GPU's reactor disassembly costs.
DOE estimates that under a $75 million program, about $25 million of GPU's costs would be of fset--or about 4 percent of the estimated $600 million needed to complete the cleanup.
Although DOE's proposed funding level may be adequate to support its research and development effort, it may not elicit the kind of financial support needed from other interested parties to allow any research and development activity.
At the present time, OMB has restricted the use of future COE funds until a program for funding the TMI-2 cleanup is developed.
We recognize that any requests for Federal funding will be matters for congressional consideration.
In our opinion, positive support for any such requests will likely oe inter-preted by the industry, oy State utility commissions, and oy the financial community as a reaffirmation of the support of nuclear power given by past Congresses and administrations.
A passive, or negative reaction, could have the opposite effect with a decided dampening of further nuclear power development I
j in the private sector.
Beyond the Federal involvement projected above, we believe the remaining ' cleanup costs should be a matter for resolution We believe by the involved States and the utility industry.
that the benefits that will accrue to all non-Federal entities should be sufficient to encourage an active participation in Various studies have indicated the cost resolving the impasse.
to the utility industry if the cleanup issues are not resolved.
It is also relatively easy to project the effect of various levels of cleanup funding on consumer rates.
Therefore, any final sharing of the costs will undoubtedly require extensive negotiations among all parties.
Strong leadership at the State level will be required, but we believe a precedent for handling future nuclear accident recovery problems can evolve from tnis experience.
The continued reluctance of the principal parties involved in the cleanup to initiate the actions necessary to resolve the funding problem may eventually require additional Federal sector involvement.
If no resolution has oeen agreed on after a reasonable time period, it may be necessary for tne Federal sector to develop a solution to tne cleanup problems and obtain its implementation by the involved parties.
i 72 mwn:sM 2tM Rs % 3Q Q [y;Qp f k "M
~ --
R I
m il RECOMMENDATION TO THE SECRETARY OF ENERGY To assure the availability of funding needed to complete an expanded research and development program at TMI, we recom-mend that DOE prepare a multi-year budget proposal for Federal participation in the TMI cleanup effort and present it to the Congress.
The budget proposal should recognize the primary leadership role of State of ficials in working with GPU and the industry in the cleanup effort and within that parameter should clearly specify the objectives to be achieved by the Federal involvement, the work steps required in each fiscal year, the application of the program results, and the total funding needed to successfully meet the research and development objectives.
RECOMMENDATIONS TO THE CONGRESS Given past congressional support for the commercial devel-opment of nuclear power, the continued Federal regulatory oversight of nuclear reactor operations and radioactive waste disposal, and the need to reduce the economic burden imposed by the TMI accident as much as possible, we recommend that the Congress provide the required multi-year funding to DOE for its research and development program at TMI.
The information gained will be helpful in (1) developing procedures that will mitigate adverse consequences of any future accidents and (2) enhancing DOE and NRC nuclear power oversight activities.
We further recommend that the Congress closely follow the current efforts to resolve the funding problems for the TMI-2 cleanup through State and utility industry financing and DOE's research and development program.
I f these State-led ef forts are not successful, we recommend that the Congress devise a mechanism which would serve to obtain the required financial l
l assistance to complete the TMI-2 cleanup.
AGENCY COMMENTS AND OUR EVALUATION DOE agrees that the rapid and safe cleanup of TMI-2 requires a cooperative effort of all concerned parties and that each one has a role to play in the process.
DOE further stated that it is moving forward with a research and development program that will provide valuable data related to nuclear safety and cleanup technology and which it believes will be of substantial value to the entire nuclear community, DOE does not believe, however, that it is necessary to seek multi-year funding to support its proposed 3-year program.. DOE officials feel that the normal annual review and Congressional authorization and appropriation processes will assure the program's consistency with DOE's 73
t i
They further believe that objectives and the cleanup needs. DOE's plans for a multi-year program a GPU and other parties to proceed with plans to fund and complete the cleanup.
We do not agree with DOE that a multi-year funding proposal for a TMI-2 research and development program is unnecessary.
Although the Governor of Pennsylvania has taken the lead in proposing a method to fund the cleanup costs, we still believe he that a commitment of Federal sector support for resolvi future of the support of other interested parties but for theWe believe that such the nuclear utility industry as a whole.
il support can best be expressed through an approved financ a i
commitment for the entire effort rather than simply a mult -year l
plan with no total funding commitment to insure its successfu Within the total plan, we believe that an annual review of progress and need is necessary and should be an completion.
integral part of the overall program.
4 74
+., - :;
- .
- 3
. x CHAPTER 6 CHANGES ARE NEEDED TO FACILITATE FUTURE ACCIDENT RECOVERY EFFORTS Our review of the factors affecting the GPU System as a result of the TMI-2 accident disclosed two areas of concern that need correcting if the problems that face GPU are to be mitigated or eliminated in the event of another major accident at a nuclear generating unit.
These concerns center around the (1) present underinsured status of utility companies which operate nuclear reactors and (2) changes needed in NRC's regulatory oversight of a utility's accident recovery efforts.
The early development of the nuclear industry was affected by the inability of the public utilities to obtain adequate levels of third-party liability insurance coverage for nuclear units.
Passage of the Price-Anderson Act in 1957 helped solve this problem by providing a means for utilities to obtain this coverage.
Property insurance coverage for nuclear units, however, was readily available to the utilities.
Over the years, both liability and property insurance coverage for nuclear units have increased.
The TMI accident in March 1979, provided the first major test of the adequacy of existing liability and property insurance coverage.
While the liability insurance coverage fostered under the Price-Anderson Act has been adequate in paying all claims to date, property insurance coverage has proven to be inadequate due to unanticipated decontamination expense.
While only
$300 million of property insurance coverage has been available, cleanup costs have been estimated at $1 billion.
The recogni-tion of this inadequacy has prompted an insurance and electric utility industry effort to increase property insurance coverage up to S450 million.
However, property insurance coverage remains deficient.
Nuclear utilities are presently considering two broad pro-posals to provide a substantial increase in property insurance coverage beyond that which is currently available.
In possible conjunction with existing sources of property insurance coverage, as much as $1 billion could be available to finance the cleanup of a nuclear accident under the proposed plans.
If the utilities are successful in obtaining the levels of property insurance coverage deemed adequate by NBC, we see i
no need for Federal Government involvement.
However, if the utilities are unsuccessful in obtaining such coverage, a Federal Government requirement for mandatory insurance should be considered to protect the Government from the possibility of having to finance the future decontamination of a disabled nuclear unit in the event of another accident.
l 75
1 NRC nas taken 27 months to allow GPU to proceed with the first phase of the cleanup efforts in the containment ouilding.
The initial priority given to the accident recovery efforts by NRC declined and by early 1980, criticisms were ceginning
/
to be made about the delays in moving forward with the cleanup process.
GPU's proposal for processing the contaminated water in the containment building were held in abeyance by NRC until the Programmatic Environmental Impact Statement on the proposal was completed in March 1981.
The uncertainties as to what cleanup methodologies would be appropriate, and public opposition to certain required cleanup activities, served to further limit or delay NRC regulatory approval of GPU's proposed efforts.
We believe that the experience demonstrates the need for NRC to develop a set of accident recovery guidelines that would be available for establishing precedures for any future accident recovery effort.
The lessons learned from the TMI-2 experience should provide a sound basis for the development of such guidelines.
INCREASED UTILITY PROPERTY INSURANCE COVERAGE IS NEEDED The present adequacy of third party liability insurance for covering offsite claims resulting from nuclear accidents to date is in contrast to the proven inadequacy of onsite property damage insurance coverage.
The voluntary increases in property insurance coverage since 1957 reached the $300 million level just prior to the TMI-2 accident and have increased even further since then.
The $450-million coverage level currently proposed is still less than what is needed, however, and the insurance organizations currently providing property insurance coverage are working to increase the level of coverage.
Plans are under consideration to offer additional insurance coverage up to
$1 billion through different insurance arrangements.
If tne present efforts in the private sector are not successful, the importance of, and need for, increasing property insurance coverage may require some type of Federally-mandated program.
Pre-accident growth of nuclear insur_ance coverage
(
l The development of the private nuclear power industry was aided by congressional limitations established for third-party I
liability in the event of a nuclear accident.
The offsite liability insurance coverage was accompanied by onsite property damage coverage designed to protect the interests of nuclear reactor owners.
76 m
,n-
'l Liability insurance had congressional backing The Congress passed the Atomic Energy Act of 1954 to promote the peaceful development of atomic energy.
It authorized private industry to construct and operate nuclear generating plants, subject to strict Government regulation.
The early private development of nuclear power was delayed, however, because electric utilities were unable to obtain adequate insurance coverage for proposed nuclear plants.
Insurance companies did not offer adequate third party liability coverage 1/ against a major nuclear accident.
The companies were reluctant to underwrite this coverage because they felt that while the risk was remote, liability losses of several billion dollars from a single accident could be incurred.
To foster the growth of the nuclear industry and in recog-nition of the problems encountered in obtaining adequate third-party liability insurance the Congress enacted an amendment to the Atomic Energy Act.
The amendment became known as the Price-Anderson Act of 1957.
The act currently limits offsite liability to $560 million in the event of a major nuclear accident.
Subsequent amendments to the Price-Anderson Act developed a complicated but effective means of providing this
$560 million of liability insurance coverage.
The Price-Anderson Act now divides the liability insurance coverage into three " layers":
--The maximum amount of liability insurance available from private sources (first layer).
--Required utility industry self-insurance (second layer).
--Federal Government indemnity (third layer).
First-layer liabililty insurance coverage of $160 million is provided by two insurance pools in the private sector--American Nuclear Insurers (ANI) and Mutual Atomic Energy Liability Underwriters (MAELU).
The two insurance pools are composed of insurance companies who have voluntarily pledged funds to either ANI or MAELU for insurance coverage.
Their collective commitment determines the maximum liability insurance coverage provided.
Currently ANI provides $124 million and MAELU provides
- 36 million of coverage.
l If a major nuclear accident were to result in offsite liability greater than the S160 million first-layer coverage, 1/ Insurance to cover offsite claims by the public for bodily l
injury and/or property damage caused by a nuclear accident.
77
the second layer of utility self-insurance would become opera-tive.
The second layer of insurance coverage is also managed i
and administered by ANI-MAELU.
If required, ANI and MAELU can order utility companies to pay a retrospective premium assessment of up to $5 million for each licensed operating nuclear unit to cover liability losses in excess of that provided oy the first layer of insurance coverage.
To protect the financial exposure of a utility, the Price-Anderson Act limits a utility's total assessment to S10 million per nuclear unit per year.
As of May 1981, second-layer coverage would yield a maximum of $360 million in insurance coverage from the 72 licensed commercial reactors.
In the event of an accident which exhausts both the first-and second-layers of liability insurance, the Federal Government is currently liable for the third layer of $40 million under the Price-Anderson Act.
Federal Government indemnity is the difference between the $560-million Price-Anderson Act limit and the sum of the first and second layers of coverage, wnich currently is $520 million.
If a catastrophic accident results in offsite liacility greater than $560 million, the Price-Anderson Act directs the Congress to review the incident and take whatever action is deemed necessary to protect the public interest.
Private sector financing of Price-Anderson liability insurance coverage has increased over the years by two means.
Both ANI ana MAELU have gradually increased first layer insurance coverage from
$60 million in 1957 to S160 million in 1980.
In addition, second layer insurance coverage has increased by SS million as each new nuclear unit has been licensed.
Under the Price-Anderson Act provisions, when total first-and second-layer liability insurance exceeds $560 million, Federal government indemnity will cease and the Price-Anderson $560-million liability limit could increase correspondingly.
There are two important characteristics of the Price-Anderson Act:
--Federal Government involvement in tne program is minimal, and
--Those who have benefitted from the use of nuclear power are also responsible for financing the potential liability resulting from nuclear accidents.
The largest liability obligation under Price-Anderson has been a tentative $25-million settlement for the accident at TMI.
This will be covered by the first layer of insurance coverage.
+
The precautionary second and third layers of insurance coverage have never been needed.
Although there is broad consensus that 78
.y-c~-
. a.c.....c m.;... w;..
s w..
the Price-Anderson Act is well structured and will work effectively, we have reported that the $560-million limit on liability provided under the act is arbitrary and should be realistically defined. 1/
Proper _ty insurance coverage was voluntary Property insurance for nuclear facilities has been available through the private insurance market since 1957 and through a Both utility-organized mutual insurance company since 1973.
insurers provide onsite property insurance coverage during tne construction and operation of nuclear units.
Private insurance market-ANI and Mutual Atomic Energy Reinsurance Pool (MAERP) are the two private companies providing property insurance coverage for nuclear generating units.
They have offered this coverage continuously since 1957 and now insure 37 nuclear units with operating licenses.
has received commitments from about 128 conventional ANI stockholder-owned insurance companies, such as Allstate Insurance Company and Hartford Insurance Company, to pledge funds--insurance capacity--for property or liability insurance.
In contrast, MAERP obtains its commitments for insurance capacity from about 100 mutual insurance companies, such as American Mutual Insurance Company and Nationwide Mutual Insurance Both ANI and MAERP also solicit reinsurance capacity Company.
from insurers abroad.
To date, approximately one-nalf of ANI-insurance capacity is derived f rom foreign insurers.
MAERP Although they are separate organizations and write separate policies, the two companies work closely together to maximize In practice, ANI and MAERP have combined insurance coverage.
their resources by extensively reinsuring each other to form,
(
a single, larger ANI-MAERP pool.
ANI-MAERP have in effect, gradually expanded their insurance coverage through the years--
f rom $63.88 million of coverage in 1957 to $300 million in 1979.
Utility-organized mutual insurance--Nuclear Mutual Limited
~fs T mutual insurance c6mpiny established by utilities (NML) with nuclear power plants.
NML has offered property insurance coverage since 1973 and now insures 15 utilities owning 32 units with operating licenses.
It was chartered in Bermuda to avoid State regulatory requirements and to take advantage of tax benefits.
As a result, NML is exempt from the cash reserve and other restrictive requirements of State insurance laws that would apply if NML 1/U.S. General Accounting Office, " Analysis of the Price-Anoerson Act", EMD-80-80, Aug. 18, 1980.
79
were chartered domestically.
According to a NML official, NML maintains a cash reserve which has grown steadily since 31, 1981.
The official also 1973 to S86 million as of May said NML has also been successful in obtaining $75 million of Nonetheless, reinsurance coverage from conventional insurers.
a major nuclear accident would exhaust both of these sources To finance the costs of financing property damage costs.
of such an accident, NML management has the authority to assess each member utility its proportionate share of the insured balance due the member owner of the disabled nuclear This is termed a retrospective premium adjustment.
unit.
To limit the financial exposure of each member utility, in any one year, the maximum yearly retrospective premium adjustment is limited to 14 times the annual premium rate Based on NML's stated annualized paid by the member utility.
premium level, this could provide an additional S580 million of insurance coverage above existing cash reserves and To date, current premiums have proven adequate reinsurance.
No to finance insurance claims and build cash reserves.
- NML, retrospective premium adjustments have been required.
like ANI-MAERP, has increased its insurance coverage since 1973--from an initial $100 million to $300 million in 1979.
Post-accident additions to insurance coverage The TMI accident demonstrated the dif ficulty f aced by a utility in paying replacement energy costs when a nuclear unit is suddenly put out of service for an extended period of time.
In 1980, the newly formed Nuclear Electric Insurance Limited NEIL is a (NEIL) began offering replacement power insurance.
mutual insurance company established by utilities with an insurable interest in nuclear power plants.
The insurance coverage provides a maximum of $156 million to owners of a disabled nuclear plant to purchase replacement power for its After a 25-week waiting period, the utility customers.
insurance coverage provides a maximum of $2 million per week for the first year of plant outage and $1 million per week for the second year. Most privately-owned utilities are memoers,of NEIL.
NEIL is an offshore mutual insurance company chartered in As an offshore insurer, Bermuda for legal and tax purposes.
A NEIL NEIL, like NHL, is er.empt from State insurance laws.
official said the company's current reserve of available funds to pay for replacement power is relatively small--about $75 It has had limited success in obtaining reinsurance million.
Consequently, NEIL would rely on from established insurers.
/
a retrospective premium adjustment to finance major insurance Total I
payments to any member owner of a disabled nuclear unit.
annual retrospective premium adjustments of any member utility l
l 80 i
is are limited to five times the annual premium rate paid by the member utility.
According to a NEIL official, retrospective adjustments at present premium levels could provide an additional $340 million of insurance coverage above cash reserves and reinsurance.
In 1980, the American Power Insurance Corporation ( APIC) was established to provide coverage for certain on-going expenses in the event of a prolonged nuclear plant shutdown.
Formed under the auspices of the American Public Power Association, the insurance coverage is provided by private insurers.
After a 120-day waiting period, it provides up to $50 million for prin-cipal and interest payments on plant deot and for the costs of plant maintenance and security.
As of April 30, 1981, only one utility was insured through APIC although several utilities have expressed an interest in obtaining coverage.
Property insurance coverage has been increased but needs further expansion The insurance industry has partially responded to utility requests for increased property insurance coverage.
In April 1981, two years after the TMI accident, total ANI-MAERP property insurance coverage was increased from $300 million to $369 million per site per policy.
ANI-MAERP repesentatives indicated that they plan to further increase property insurance coverage to aoout
$450 million in January 1982.
Correspondingly, NML coverage has increased from $300 million to $450 million since the accident. 1/
Coverage has not increased to the levels needed by many utilities, however, and they remain underinsured.
Electricity is now being generated by utilities with nuclear units that do not have adequate property insurance coverage.
Insurers, bankers, investment brokers and the utilities themselves have all noted that utilities with nuclear capacity, sucn as Met Ed, are now exposed to tne risk of bankruptcy in tne event of a nuclear accident.
A financial report by the investment firm of Paine, Webber, Jackson, and Curtis on the financial consequences of the TMI accident concludes:
"We would like to emphasize that many investor-owned utilities engaged in nuclear construction or operating on-line plants presently do not enjoy the pre-TMI financial status of Met Ed.
For many companies the nuclear involvement is so large that if a major nuclear accident were to occur, we submit that these companies could be in an even more precarious financial situation than the one in whicn.we find Met Ed." 2/
1/The $450 million coverage became available on August 1, 1981.
2/ Paine, Webber, Jackson, and Curtis, Inc., Metropolitan Edison:
Financial Repercussions of a Major _ Nuclear Accident, Novemoer 1980.
81
The Edison Electric Institute is also concerned about inade-l quate property insurance coverage and has appointed a task force composed of utility representatives to investigate possible methods of financing the cleanup of TMI-2.
The task force stated that "There is need for a substantial increase in property insurance coverage for nuclear power plants...EEI should promptly explore
/
the development of an insurance program which would provide substantially greater property insurance coverage." 1/
Problems with increasing coverage Due to regulatory and financial constraints, it does not appear that either ANI-MAERP or NML can significantly increase their existing full prepaid or retrospective-based property insurance coverage beyond what is currently available or planned in the near future.
Although this assessment was obtained from both insurance groups, the reasons for the apparent limitation differed.
American Nuclear Insurers--Mutual Atomic Energy Reinsurance Pool ANI-MAERP representatives do not believe that they can expand existing prepaid, full coverage property insurance to $1 billion in the immediate future.
They stated that the insurance companies from which they derive their insurance capacity are reluctant to commit that level of coverage funds to the ANI-MAERP insurance pool.
Two factors combined to limit the insurance capacity The first factor is the lack of provided by insurers.
actuarial knowledge of the risks involved in providing insurance coverage to nuclear units.
Insurers view nuclear insurance as only one of several possible investment oppor-tunities.
Since millions of people have insured themselves and their property, insurance companies have good actuarial knowledge of the risks involved in providing conventional coverage for life, auto, commercial, and homeowner insurance.
the limited experience with nuclear plants does In contrast, not provide a reliable actuarial basis for assessing risk, particularly since decontamination and other cleanup costs As a result, many insurers are not firmly established.
simply prefer to avoid this actuarial uncertainty and limit their commitment to the nuclear insurance pools.
The second factor involves the insurance industry perception that the nuclear industry is particularly vulnerable 1/ Report of the Task Force on Nuclear Institutional Issues to EEI Board of Directors, Cleanup of Three Mile Island Unit No.
2, March 1981.
82 i
l 1
--+s a 4mWMkh;f7Cg"WR$ E_8MC _q
to future Federal Government intervention which could threaten future premium flows.
To date, there are relatively few nuclear plants providing a relatively small annual premium flow to the insurers.
Despite the low probability of a major nuclear accident in any given year, insurance losses from a nuclear accident can easily exceed total premiums collected by the insurers in that year, as at TMI.
Consequently, an insurance technique known as inter-temporal loss spreading is applied to the nuclear industry.
This approach means that while insurers may recognize a loss in any given year, over a period of several years insurers expect to realize a return on their investment.
The TMI accident exhausted several years of previously accumulated premiums for ANI-MAERP insurers.
Many of these insurers would be exposed to a net loss on their investment if another major nuclear accident occurs in the next few years and the Federal Government subsequently takes strong regulatory action against the nuclear industry.
Government regulatory action could range from severe constraints placed on the industry to shutting it down entirely.
Such Government action would threaten future premium flows and, as a result, the overall return on investment to insurers.
Consequently, due to their fears of Federal Government regulatory action, insurers are hesitant to increase their insurance commitmen t to ANI-MAERP.
Given the uncertainties of assessing the risk and the regulatory climate that surrounds the nuclear industry, many insurers would rather commit their insurance capacity to conventional insurance coverage where the risks and returns are more predictable.
Nuclear Mutual Limited A representative of NML also stated that they do not anticipate being able to expand the property insurance coverage to $1 billion in the immediate future within the current company structure.
They explained that property insurance coverage cannot be rapidly increased to $1 billion without a corresponding increase in retrospective premium adjustments, reserves, or reinsurance.
Otherwise, a series of nuclear accidents could force NML into bankruptcy.
NML member utilities are concerned, however, about increasing retrospective premium adjustments because they are uncertain whether a State public utility commission would allow the cost of a retrospective premium adjustment to be passed through to the ratepayer.
Since retrospective premium adjustments have never been assessed, no precedents exist upon which to predict State commission's ratemaking decision.
We contacted State a
utility commission representatives in New York, Florida, Illinois, Wisconsin, and California.
They stated that utilities usually contact their respective state commissions to determine their opinion regarding membership in NML and possible retrospective premium adjustments in consumer rates.
The representatives speculated that the present commissions would probably be willing to include some or all of a retrospective premium 83
adjustment in the consumer rates if an accident occurred at a However, they could neither commit their NML member utility.
commissions to this course of action nor predict the actions of l
I any future State commissions.
If a State commission refused to allow a retrospective premium adjustment to be collected in the rates, the utility would be required to internally finance a retrospective premium adjustment.
Most of the NML's member utilities own more than Depending on the number of units owned, a future one unit.
retrospective premium adjustment could range from S5 million to over $100 million for each member utility in any given year.
For example, Jersey Central has insured its Oyster Creek unit As a NML member, the company is subject to an annual with NML.
retrospective premium adjustment of up to $19.6 million if a major incident occurs at a nuclear unit of any other member If the New Jersey Board of Public Utilities decides, utility.
Jersey Central could be required to internally generate some or all of a retrospective premium adjustment.
This possibility makes some utilities reluctant to expose themselves to increased retrospective premium adjustments.
Due to its limited membership, relatively small insurance reserves, and concerns about State regulatory actions, NML does not now appear to have the capability of offering $1 billion in property insurance coverage without imposing extraordinary risk exposure on its member utilities.
Options for increasing property insurance coverage The apparent inability of ANI-MAERP and NML to overcome the present inadequacy of nuclear property insurance coverage raises questions as to what other options might be considered.
Although insurance organizations are limited, there are ways the present in which the insurance and utility industries can meet the pro-If perty insurance needs of utilities with nuclear reactors.
these needs cannot be met voluntarily, some type of wandatory coverage may be required to provide sufficient funds from non-Federal sources to protect public health and safety in the event of another major nuclear accident.
Voluntary coverage by the insurance and utility industries The first alternative for increasing nuclear property insurance is to rely on the insurance and utility industries to eventually increase property insurance coverage to adequate As a result of TMI, some industry ccmmitment toward levels.
increasing overall insurance coverage has been demonstrated by the formation of NEIL, APIC, and the increased property insurance coverage now offered by both ANI-MAERP and NML.
84
As described earlier, neither ANI-MAERP nor NML are indi-vidually capable of increasing property insurance coverage to adequate levels in the immediate future.
However, the possiDility exists for the utility and the insurance industries to increase property insurance coverage to adequate levels.
Two possiole methods will be discussed.
They are (1) quota snaring and (2) primary / excess insurance coverage.
Quota Snaring--Quota sharing is a common practice in tne insurance industry.
In quota sharing, each insurer is responsi-ble for its proportionate share of any covered loss.
If applied to nuclear insurance, a utility could elect to purenase property insurance coverage from botn ANI-MAERP and NML.
Witn quota sharing, coverage f rom the combined resources of ANI-MAERP and NML would nearly double to about $675 million if a utility elected to obtain the maximum amount available.
In discussions with both insurers on quota sharing, however, several major problems surfaced.
These include
--reduced premium flows to both ANI-MAERP and NML,
--possible violations of antitrust laws,
--dif f ering safety and contract standards between ANI-MAERP and NML, and
--increased exposure of reinsurers who insure ooth ANI-MAERP and NML.
ANI-MAERP representatives have stated that they are willing to quota share witn NHL.
However, NML representatives have argued that quota sharing would reduce NML annual premiums which have been As the reserves increase, the financial its reserves.
used to build exposure of a member utility from a possible retrospective premium adjustment declines.
If quota sharing were implemented, reserves probably would not increase as rapidly and, if an accident occorrea, NML members would face a larger retrospective premium adjustment.
In addition, NML representatives emphasized that quota sharing could impair the healthy degree of competition that now exists between ANI-MAERP and NML.
Tney contend that this competition nas improved the contract terms and amount of property insurance Both insurers coverage offered to all of the nuclear utilities.
noted differing safety and contract standards as well as potential problems for reinsurers who provide insurance capacity for both Assuming all of these difficulties can somehow ANI-MAERP and NML.
De resolved, quota sharing can be viewed as one possibiliity for increasing future property insurance coverage.
Primary / excess insurance coverag_e--Primary / excess insurance is another common insurance practice.
It means that one l
coverage 85
insurer provides insurance coverage for the first layer (primary) while another insurer provides second layer insurance coverage (excess), up to a specified maximum, for losses in excess of the first layer.
For example, insurer A would provide for insurance losses up to $375 million while insurer B would provide insurance to a coverage only for those losses in excess of $375 million, The major problem with this type of insurance specified maximum.
coverage has been that both ANI-MAERP and NML wanted to provide the primary coverage to obtain the higher annual premiums.
In the past, neither company has been interested in providing the second-layer coverage.
Recent developments indicate this attitude may be changing.
The nuclear utility industry is currently considering too broad proposals to provide for a major increase in property insurance Under one proposed plan, NEIL would provide member coverage.
utilities $500 million of property insurance coverage for losses in excess of $500 million.
The proposed excess property insurance program would be similar to NEIL's replacement power program and NHL's property insurance program in that it would collect an annual premium to build reserves yet probably rely on retrospective premium adjustments to finance large losses.
It is hoped that widespread industry participation will lessen the financial exposure of any one member.
The relative amounts of coverage via retrospective premium adjustments and reinsurance from the insurance market have not yet been determined nor has any decision been made on possible quota sharing or primary / excess insurance layering with ANI-MAERP or other insurers.
A spokesman for the insurance group stated that he believes that these issues will quickly be resolved and the new company could begin offering insurance by the end of 1981.
A second broad proposal has been advanced by ANI-MAERP.
They are proposing three layers of property insurance coverage to the nuclear utilities.
The first layer is the existing coverage avail-able from either ANI-MAERP or NML.
A second layer of $350 million insurance would be utility self-insurance in which a retrospective assessment or adjustment would be collected only if the first layer of insurance coverage was exhausted by a nuclear accident.
1/
The third layer would be ANI-MAERP pre-paid insurance coverage 1/Present NML member utilities may be offered full second and third layer insurance coverage up to $1 billion for losses exceeding their NML coverage limit.
0 86
{
- - - - e smesacs s g.g g g
3 Despite the concerns of insurers about providing add insurance capacity, ANI-MAERP representatives believe they can obtain the insurance capacity from the insurance market 1982.
and begin to offer the additional coverage by January These alternatives appear to provide realistic options for increasing property insurance coverage to $1 billion withoutThe effectiven or near unanimous, participation Government intervention.
be measured by (1) the unanimous, in either insurance proposal by nuclear utilities and (2) the by prompt provision of the additional property insurance coverage rage.
these possible sources of additional property insurance cove Mandatory insurance nay become necessary If the utility industry is unable to obtain adequate property a mandatory insurance insurance coverage in a timely manner,One method for providing this coverage program may be required.would roughly parallel the first and second layer l aT i bility coverage now required under the Price-Anderson Act.
involvement of this requirement would be to minimize Governmentin pro d not margin of insurance coverage so that the Government neebe there would be two required layers Under this proposal, as part of its licensing
- NRC, of property insurance coverage. authority, would require commercial nuclear i
bly the maximum amount of property insurance coverage reasonaThis NML, or any other insurer.
available through ANI-MAERP,should not impose any additional burden bec nuclear utilities already have property insurance coverage. h If an accident exhausted property insurance coverage from t ethe affecte first layer of insurance, from the owners given the authority by NRC to assess and collecta prorated retrospective prem of each licensed nuclear, unit which would not to exceed some specified amount, be paid to cover additional property losses incurred from the assessment, Based on the current number of licensed operating this cecond layer could provide an additional accident.
S720 million property insurance coverage if the maximum assess-The congres nuclear units, ment established were $10 million.per nuclear unit will determine the total As under maximum assessment amount of second-layer insurance coverage available.NRC could requir the Price-Anderson Act, it will unit operator provide some measure of assurance thatAssuming a maximum $
not def ault on an assessment. total property insurance coverage could assessment per unit, 87
exceed $1 billion.
Both layers of insurance could be managed and administered by insurers in the private sector with minimal Federal Government involvement.
As the first layer of insurance coverage grows, and as new methodologies and a revised regulatory framework for coping with nuclear accidents are developed from the cleanup of TMI-2, the primary coverage from ANI-MAERP or NHL could be adequate to cover l
the costs of a major nuclear accident.
Under this proposal, l
however, the precautionary second layer of utility self-insurance would be available in the event of a nuclear accident with costs that exceed primary coverage.
This shifts the burden of financing any radioactive cleanup from the Government to the nuclear industry.
We discussed this mandatory second-layer, self-insurance proposal with representatives from ANI-MAERP, NML, and the electric utility industry.
Response to the proposal was favorable although it was stated that congressional action would probably be required to give NRC the statutory authority to require such l
property insurance coverage.
l The mandatory insurance option is comparable to the voluntary insurance options now under active consideration by the utility industry.
Both promote the idea of minimal Federal Government involvement in providing nuclear property insurance.
However a possible major difference between a mandatory and a voluntary insurance program is that total nuclear utility membership is assured if the program is mandatory.
Therefore, the maximum amount of nuclear utility funds would be available to finance a possible future cleanup if so required.
While widespread utility participation in any voluntary insurance program would also provide maximum insurance coverage, there is no assurance that such widespread industry participation will occur.
Consequently, the possibility would exist that a major nuclear accident could occur at a nuclear unit which was not a participant in the voluntary insurance program.
As a result, the Federal Government could become involved in financing decon-tamination efforts at a disabled nuclear unit.
Current legislation proposes more direct Federal Covernment involverent The need for additional property insurance coverage was recognized in the House and Senate 1/ bills discussed previously in Chapter 5.
Each of the bills proposed that a National Nuclear Property Insurance Corporation be established to provide for supplemental insurance to cover costs resulting from damage 1/H.R. 2512 and S.
1226, respectively.
I es i
t i
l ww.e to nuclear powerplants.
The corporation would be exempt from most taxes and from any general limitations imposed by statute on budget outlays of the United States.
The Federal Government would not be liable for any obligation or liability incurred by the corporation.
The proposed legislation establishes a Nuclear Property Insurance Fund on the U.S. Treasury books which would be credited with insurance premiums, interest, and charges and other monies collected or accrued to the f und.
The corporation established by the legislation would be authorized to provide insurance to supplement that which is available from private sources.
The proposed coverage would conpensate the insured utility for costs that exceed the greater of $350 million per accident or $50 million plus the amount of insurance available from private sources.
The maximum coverage for any one accident would be limited to the greater of S2 billion or an amount determined by the corporation's board of directors.
Utility participation would be mandatory with compliance linked to the issuance or validation of the nuclear operating licenses by NRC.
Premiums would be assessed for each utility so that they equaled at least S150 million annually, and this would continue until a reserve of at least $750 million had been accumulated.
If payments from the fund exceed the amount available, each insured utility would be liable for an Ldditional assessment to cover the obligation.
The legislation provides for the corporation to be converted to a private mutual insurance company at some future date.
As we pointed out in chapter 5, the corporation would be empowered to l
make partial payments to GPU for the TMI-2 cleanup from insurance l
premiums.
The conversion would occur when the TMI-2 cleanup is completed or the reserve fund has accumulated $750 million, whichever is earlier.
NRC NEEDS TO ESTABLISH ACCIDENT RECOVERY PROCEDURES NRC is responsible for regulating the operations of nuclear rnactors used to generate electricity in the private sector and l
tor protecting public health and saf ety with respect to radio-l active exposure.
This responsibility makes it mandatory that in the event of a nuclear accident, NRC must become directly involved in any accident recovery effort.
This involvement covers two major areas--approving the recovery methods employed by the reacter owners and responding to public concerns over radiation exposure resulting from an accident.
89
TMI-2 was the industry's first major nuclear accident that tested NRC's response capability.
The length of time NRC took to reach the point where it could approve GPU's plan for cleaning up the TMI-2 containment building indicates one thing:
the need for NRC to develop accident recovery guidelines for use in establishing procedures for any future accident recovery efforts.
The lessons learned from the TMI-2 recovery experience should provide a good basis for developing the guidelines.
NRC followed its normal procedures in responding to GPU's needs The restrictive environment in which a utility company operates a nuclear reactor made it difficult for GPU to immediately begin cleanup activities without the full co-operation and approval of NRC.
NRC did not have any specific guidelines or criteria pertaining to a nuclear accident recovery effort but followed its normal regulatory process.
Under this process, GPU is required to propose the methodology for cleaning up the accident damage and submit it to NRC for evaluation and approval.
If the NRC staff finds the proposal adequate, they send it to the Commission with their recom-mendation for approval.
If the utility's proposal is not adequate, it is returned by the staff for more information and development.
This process can take a long time to a
complete depending on how quickly the (1) utility submits a proposal, (2) NRC staff responds favorably, and (3) NRC accepts the staff's recommendation and gives its approval.
The accident resulted in over 500,000 gallons of inter-mediate-level contaminated waste water flowing into the auxiliary building tanks and over 600,000 gallons of high-activity waste water collecting in the basement of the containment building.
The continued release of radioactive substances from this water f
required that it be processed and removed before the rest of the decontamination work could proceed.
Shortly after the accident, GPU began designing the systems needed to process the contaminated water in both locations.
On May 25, 1979, NRC directed its staf f to prepare an environmental assessment regarding GPU's proposals to decontaminate and dispose of the radioactive waste water at TMI-2.
The first part of the assessment dealt with pro-cessing the water in the auxiliary building and was issued on October 3, 1979.
On October 16, 1979, NRC approved the use of GPU's proposed EPICOR-II system.
The decontamination and disposal of the containment l
building's radioactive water was scheduled to be considered in a subsequent staff assessment.
Before the assessment was com-pleted, NRC, in a Statement of Policy dated November 21, 1979, directed its staff to prepare a programmatic environmental 90
u.,
.. w a..j..
impact statement on the decontamination and disposal of all radioactive wastes resulting from tne accident.
Consistent with the Commission's November 21, 1979 directive, the Director of NRC's Office of Nuclear Reactor Regulation imposed a require-ment, as of February 11, 1980, that GPU not undertake the processing and discharge of water in the containment building and the reactor coolant system without NRC approval.
Lengthy delays affecting TMI-2 cleanup have occurred since November 1979 Tne initial emphasis given to tne accident recovery effort appeared to diminish after NRC's November 21, 1979, Statement of Policy was issued.
This was pointed out in a NRC staff task force report, utility company complaints, and in an industry report on the TMI-2 cleanup.
On February 28, 1980, a special NRC task force reported to the Commissioners on its evaluation of the cleanup activ-ities at TMI-2. 1/
Tne task force's findings are summed up as follows.
"The main thrust of our findings recommendation is that pro.npt action is needed by NRC to restore forward motion to the Three Mile Island cleanup process.
During our meetings with NRC staff, licensee management, and Pennsylvania State officials, we observed frustration with tne pace of tne cleanup, the lack of criteria, tne tedious decision process, and the erosion of what once was a high priority program.
We have.not ooserved strong initiatives to change these conditions.* * *.
"We believe the Commission should announce quickly a commitment to proceed as expeditiously as possible with tne cleanup.
Under the general umbrella represented by this commitment, we would expect to see increased priority given to cleanup-related activities and some reallocation of resources into tnese activities.*
Utility company correspendence with NRC during 1980 continued to express GPU's concern over the lack of importance attacned to the cleanup by NRC and the affect tnat regulatory uncertainty had on the cleanup timing and cost.
On June 30, 1980, f o r exampl e',
GPU responded to a May 28, 1980, letter from the Director, Office of Nuclear Reactor Regulation, concerning the cleanup activities.
In its response, a GPU official stated:
- - - ~ -
1/ Report of a special task formed by NRC's Acting Executive Director for Operations, " Evaluation of the Cleanup Activities At Three Mile Island, "Feb. 28, 1980.
91
l i
i "Your May 28 letter does not reflect the importance we attach to the prompt clean-up of contaminated water in the in fact, be read as It can, the containment structure.
l requiring both completion of the entire PEIS and extensive regard-consideration of all alternative treatment systems, before less of their availability or state of development,This would a decision is reached on operation of the SDS.be consistent with the Comm not, in our view, 1979, which recognized that the statement of November 21, might require early action in advance of the completionpu deliberate delay would foreclose the the PEIS.
In fact, action we propose--namely, acceptable treatment of the containment water as quickly as possible."
1980, letter to the NRC Chairman, GPU In a September 12, In the letter, GPU again expressed concern over NRC actions. based on NRC documents relating it was led stated that, to conclude that:
"* * *we should not rely on any significant regulatory guidance or definition of criteria or approval to proceed with major cleanup activities until completion of the That completion had been scheduled for late 1980 but we understand that serious consideration is being final PEIS.
given to extending the period for comments on the draft Further, the PEIS with resultant delay in its completion. draft PEIS indicate we cannot expect to have the definitive guidance and criteria required for us to establish firm plans.
statement, Instead, much of the cleanup criteria apparently will be developed in the process of reviewing our proposals on We do not believe that such an a case by case basis.
approach permits timely, effective progress.
"We do not consider that this indicated regulatory approach provides the maximum assurance of protecting the public health and safety.* * *.
GPU recently prepared a synopsis of its accident recovery d the efforts, concentrating primarily on the costs incurred anThe synop progress made.nearly $200 million spent on the recovery, only a few key items in the overall scope of the recovery operations have beenwhen compared to some extent, GPU pointed out that,the magnitude of expenditures to date is very accomplished.
with actual results,much a function of delays in the regulatory process which have GPU hampered an expeditious execution of the recovery process.significa maintain the status quo while waiting for regulatory approval further stated that to proceed.
92 1
N '_' x h w
]
I A similar finding of regulatory delay was included in an Edison Electric Institute task force report on TMI-2 cleanup.1/
the task force concluded that the regulatory In its report, environment is uncertain and during 1980, NRC had not permitted the program to be performed via previously established regulations task force found that long delays occurred and guidelines.
The because minor technical and radiological problems were referred to NRC in Washington, D.C.,
for approval.
It reported that NRC has informally indicated significant possible changes in its waste disposal regulations and has been reluctant to approve significant site activities.
The report concluded that the present cleanup cost estimate will likely increase unless the regulatory process is accelerated.
The report also postulated that under a more expeditious program, some savings could be achieved.
Union of Concerned Scientists 2/ officials were also critical of NRC's actions but from a slightly different perspective.
Although they recognize the need to quickly complete the cleanup, that NRC has not followed a planned and deliberate they believe The officials believe course of action in the cleanup process.
They NRC may be acting too quickly, at least in certain areas.
pointed to the unresolved waste disposal issues resulting from processing the contaminated water in the auxiliary building as an example of NRC's approving a GPU action without having a well thought out plan for the entire process.
the delays imposed by Notwithstanding its concerns over NBC during 1980, GPU moved ahead with processing the water in the auxiliary building and decontaminating its interior surfaces.
GPU also proceeded to design and construct the Submerged to process the containment water Demineralizer System (SDS) it own risk since NRC had not given its although it did so at NRC finally allowed GPU to vent the approval to use the system.
accumulated Krypton-85 gas from the containment building in and subsequently a number of manned entries were made mid-1980, to determine conditions inside the building.
Actual cleanup activity in the containment building, however, had been delayed its approval to the final PEIS in its Statement until NRC gave of Policy issued April 27, 1981--17 months after the PEIS was ordered and 25 months from the date of the accident.
-1/ Report of the task force on Nuclear Institutional Issues to EEI Board of Directors, " Cleanup of Three Mile Island Unit No.
2,"
i 1
March 1981.
is a coalition of scientists, 2/A nonprofit corporation that engineers, and other professionals who are " concerned about safety, environmental, and national security problems."
- health, 93
Current NRC actions indicate a more responsive regulatory e_nvironment NRC recognizes that delays have occurred in the cleanup The use of the effort but believes it has acted properly.
SDS and procedures proposed by GPU, although represe The November 21, 1979, conditions that existed at TMI-2.
decision to require a PEIS was based on NRC's perceived need to comply with the National Environmental Policy Act pro-NRC al In sum-opinion in selecting the decontamination procedures.
visions.
marizing the actions taken regarding TMI-2, NRC has stated that
"* *
- simply put, the Commission has a responsi-bility to fully evaluate the environmental impact of decontamination, including involving the publicinto the ing environmental issues and alternatives befor_e commitments to specific choices are made."
Although the future regulatory environment imposed by NRC on cleanup activities cannot be predicted, the actions taken il-since the Commission approved the PEIS indicate that the ava in suc-ability of funds to proceed may be GPU's biggest hurdle cessfully resolving its problems.
Two of GPU's immediate concerns--processing the water in d
the containment building and disposing of the concentrate l
the NRC program director for been resolved.
On June 18, 1981, t
TMI-2 sent GPU the NRC order requiring the prompt com inated water in the auxiliary building tanks and the highly contaminated water in the containment building sump and in the On June 3, 1981, DOE reactor coolant system using the SDS.
notified NRC that it had proposed a waste immobilization re-i search and development program in its fiscal year 1982 fund ng request to the Congress that would include acquiring all the SDS waste products generated by the water-processing system.
NRC staf f are also developing criteria for disposing ofA radioactive wastes other than those generated by the SDS.
d change in the regulations for this purpose is being prepareIn addition, a i
for NRC's consideration. standing with DOE on the entire waste disposal pro prepared.
NRC forwarded Amendment No.16 to the On June 26, 1981, The amendment, which had been re-l operating license to GPU.
establishes the environmenta 10, 1981, quested by GPU on Junerelease criteria that will be used for any radioactive water effluent generated by the cleanup.
94
NRC staff departed from their normal reactive method of responding to utility proposals when they engaged in a 2-day review meeting with GPU staff on its proposed use of the SDS NRC arranged the meeting to process the containment water.
for f ace-to-f ace discussions of the health-and safety-related SDS issues in lieu of the time-consuming process of sending GPU a formal list of questions which would require a formal The meeting resulted in a mutually agreed upon response.
list of commitments for additional information needed by NRC to take appropriate action on GPU's proposal.
NRC has also sought authority to amend a utility's operating license prior to holding a public hearing on the amendment if it determines that the amendment involves no The need for legislation significant hazards consideration. Court of Appeals decision 1/
resulted from a November 19, 1980, that NRC may not issue a license amendment, even if it involves no significant hazards considerations, prior to holding a hearing requested by an interested person under section 189(a) of the Atomic Energy Act of 1954, as amended.
The case arose out of NRC's decision that allowed GPU to vent the Krypton-85 gas from the TMI-2 containment prior to granting a pending for a hearing on the proposed order.
request On March 11, 1981, NRC submitted proposed legislation to the Senate Committee on Environment and Public Works that would amend section 189(a).
The request addresses NRC's concerns in unnecessary the Court of Appeals decision could result that disruption or delay in the operation of a nuclear powerplant and could impose unnecessary regulatory burdens on NRC that related to significant safety benefits.
are not The actions taken to date by NRC are only the first of The many decisions that will be needed during the cleanup.
NRC staff now have authority to approve GPU activities covered Not all cleanup issues were covered in the PEIS, by the PEIS.
however, and determinations will have to be made as to what further NRC steps will be required to authorize action on these issues.
CONCLUSIONS We believe that the private sector has the resources to provide adequate property insurance coverage without 1/ Step _ hen Sholly and Donald E.
Hossler_ v. NRC, et al_.,
and et al., Nos. 80-1691 People Against Nuclear Energy v. NRC, United States Court of Appeals, District and 80-1783, The Supreme Court accepted this Columbia, Nov. 19, 1980.
case for review on May 22, 1981.
95
There is some indication that the Government intervention.
nuclear utility and insurance industries are voluntarily taking positive steps to markedly increase current coverage We believe that this is the preferable method and levels.
- However, should be encouraged by Federal and State regulators.
if the nuclear utility industry is not able to obtain adequate if not all, nuclear reactors, and prompt coverage for most, Government requirement for second-layer insurance coverage may be desirable to protect the Federal Government against the possibility of having to finance the decontamination of a disabled nuclear unit.
If mandatory coverage becomes necessary, we believe that it can be successfully managed by the private insurance and/or Therefore, the development of a utility industry sectors.
2512 and quasi-governmental corporation as envisioned by H.R.1226 to provide Because the effects S.not appear to be necessary at this time.
of a nuclear accident generally involve public health and safety issues, however, we believe that NRC should monitor the progress being made by the utilities to increase property insurance coverage to levels that NRC determines are adequate The electric utility industry's assessment that additionalto preclud coverage could be available by the end of 1981 can be used as a benchraark for NRC's evaluation.
NRC's response to GPU's accident recovery efforts could We recognize that the accident have been more constructive.
was not anticipated by either NRC or the utility company and consequently, no one was really prepared to respond to the The severity of the accident and its resulting conditions.
uniqueness, however, should have been sufficient cause for NRC to depart from its traditional regulatory practices of reacting to utility proposals and take a more active role in in developing an working with GPU and others, such as DOE, We do appropriate response to the accident recovery needs.not believe th and safety standards would have been compromised had more frequent face-to-face meetings been held between and among the various participants to expedite the resolution of the numerous problems generated by the accident.
Our review of NRC's actions point up the need for NRC to develop accident recovery guidelines that a utility company can use in proposing procedures for any future accident recoveryIn which of the existing standards will apply to an accident efforts.
Effluent release and man-rem exposure standards used for operating units, for example, may be adequate during situation.
most accident recovery efforts and need only be designated 96
l for use by the utility.
In other cases, deviations from these operating standards may be necessary, and these should be defined for use in developing recovery procedures.
Guidelines could encompass both procedural and technical aspects of the recovery effort.
It would be helpful to the utility if the " regulatory envelope" within which it wil.1 operate were clearly defined.
More specifically, the guidelines could address such matters as (1) how to request approval for special situations not covered in the regulations, (2) which of the existing regulations would apply in a given accident situation, (3) how detailed do proposed cleanup plans have to be, (4) what is the criteria for the content and detail of Technical Evaluation Reports, and (5) what is the role and responsibility of NRC staff assigned to an accident site, i.e.,
are they limited to reviewing proposed actions or can they provide technical guidance as well.
To the extent possible, it would be useful to have NRC-approved technical methodologies available for utility use so that lengthy stu-dies of alternatives could be minimized.
The knowledge that is being gained from the TMI-2 accident recovery process, and recovery efforts for less severe accidents at other utilities or nuclear installations, should provide NRC a good basis for the development of the necessary guidelines.
RECOMMENDATIONS TO THE CHAIRMAN, NUCLEAR REGULATORY COMMISSION Because another nuclear accident at an under-insured utility company could seriously affect public health and safety, we recommend that NRC closely follow the current efforts of the insurance and utility industries to increase insurance coverage to what it determines to be an acceptable level.
We further recommend that no later than December 31, 1981, NRC assess the progress being made.
This assessment should include an evaluation of the insurance available in the private sector and a determination as to whether a mandated insurance coverage program is necessary.
To mitigate future regulatory constraints on nuclear accident cleanup activities, we recommend that NRC establish a set of guidelines that would f acilitate the development of recovery procedures by utility companies in the event of other nuclear reactor accidents.
The preparation of the guidelines should be initially based on the lessons learned and experience gained from the TMI-2 cleanup and recovery i
efforts at other nuclear installations.
Because a number of years may pass before another comparable accident occurs, NRC should periodically assess the adequacy of its guidelines and standards and evaluate the state-of-the-art technology l
for decontaminating air and water effluent produced by a l
nuclear accident to ensure that it can quickly respond to 97
the needs of the regulated utility and adequately protect the public health and safety.
AGENCY COMMENTS AND OUR EVALUATION l
NRC agreed that there is a need to monitor the efforts of the insurance and utility industries to increase the insurance coverage to acceptable levels and that a recommendation toNRC pointed out th this effect would be useful.
it approved the publication of a proposed rule for public I
- 1981, if approved as a final rule, would required power comment that, reactor licensees t'o provide the maximum amount of property NRC staff will also keep abreast of the insurance available.
current proposals to increase coverage.
however, on our suggestion that it NRC did not comment, determine the acceptable level of insurance coverage needed We and measure industry progress based on that determination.
believe a requirement that licensees obtain the maximum coverage available is inadequate if that level of coverage is not enough We would, therefore, to fully cover accident recovery costs.
expect NRC to make such a determination and use it as a benchmark in its evaluation of industry progress in increasing insurance coverage.
NRC stated that it was unclear as to what guidelines were needed for future accident recovery efforts and that each accident has unique characteristics that make it difficult to establish a We agree that each general prescription for cleanup activities.
We believe, accident may have certain unique characteristics.
however, that general guidelines for use in formulating accident recovery procedures that meet the needs of a specific accident situation can be developed and would be useful to those respon-Accordingly, we have sible for formulating such procedures.
expanded our discussion of the need for guidelines on pages96 an This expanded discussion also responds to could encompass.
DOE's comments concerning the need for new guidelines and regulations governing cleanup activities from future accidents.
NRC also stated that our report implies disagreement with the course of action outlined in its Statement of Policy ofIn commenting November 21, 1979.
a PEIS prior to approving a cleanup procedure, we did not take DOE and GPU officials and a position on the need for a PEIS.NRC staff, however, did question the nee the results of the PEIS before allowing GPU to proceed with certain cleanup activities and attributed cleanup delays to the NRC decision.
0 98
APPDTDIX I APPCiDIX I
((Q*tG R
UNITED STATES
[g NUCLEAR REGULATORY COMMISSION
,e g
WASHINGTON, D. C. 20655 j
\\...../
AUG 7 1981 Mr. J. Dexter Peach Director, Energy and Minerals Division U.S. General Accounting Office 441 G Street, N.W.
Washington, D.C.
20E48
Dear Mr. Peach:
We appreciate the opportunity to comment on the draft GAO report entitled
" Greater Comitment Needed to Solve Continuing Problems at Three Mile Island."
The report makes essentially two recomendations to the Nuclear Regulatory Commission. The first recomendation addresses the current efforts of the insurance and utility industries to increase insurance coverage to acceptable levels and we find this recomendation to be useful to the agency. The second recomendation, which addresses the need to develop a set of guidelines to facilitate future post-accident recovery efforts, is somewhat unclear and it would be more helpful to NRC if the report and the recomendation were supple-mented to set forth a greater degree of clarification.
NRC staff coments on the report are provided below.
COMMENTS ON RECOVERY AND RESTART PROCESS The report implies that NRC has been a cause of lengthy delays in GPU's TMI-2 recovery process and in returning TMI-l to commercial operation.
In regard to the recovery process at TMI-2, the report is critical of NRC's reactive role.
We have been functioning under the provisions of NEPA and the Atomic Energy Act and, therefore, it is the licensee's responsibility to propose, initiate, end fund cleanup activities.
If GA0 believes the NRC should depart from this l
arrangement, as implied on page ivx, then we would appreciate more details.
Further, the pace of the cleanup has largely been determined by the timeliness and quality of GPU proposals to the NRC, the licensee's ability to meet construction schedules, and the general level of expenditures by GPU on TMI-2 I
cleanup.
NRC actions have not been the sole determining factors in the cleanup.
l The report also indicates that NRC made no decision on cleaning up the more contaminated water in the containment building for 27 months after the accident (page ivx). This is a factually true but incomplete statement that sets the tone of the report. The licensee did not submit their proposal, describing a l
modified system for processing this water until almost 24 months after the accident. Within approximately 3 months after receipt of this proposal, and following numerous NRC/ licensee meetings, NRC granted approval to the licensee to process containment building water. The licensee was simply not capable of processing containment building water until July of this year because system construction, preoperational testing, and processing methodology were not completed.
l 99
APPENDIX I APPENDIX I Mr. J. Dexter Peach With regard to delays in TMI-1 restart, the Comission decided in July 1979 that it lacked the requisite reasonable assurance that Unit 1 could be operated without endangering the health and safety of the public and that it was in theIn public interest to conduct a public hearing prior to the restart of TMI-1.
/
addition to the applicable technical fixes identified as a result of the accident at Unit 2, some unique concerns at TMI-1 that need resolution before restart are:
(1) potential interaction between Unit 1 and the damaged Unit 2 (2) the manage-ment capabilities and technical resources of Metropolitan Edison, including the impact of the Unit 2 accident on these, (3) the potential effect of operations necessary to decontaminate the Unit 2 facility on Unit 1, and (4) recognized deficiencies in existing emergency plans and station operating procedures.
The hearings that were initiated have taken longer than originally expected and. in that sense, the hearing process is a cause of some delay.
COMMENTS ON THE FIRST RECOMMENDATION The first recomendation addresses the availability and level of insurance coverage of utilities. You should be aware that on Thursday, July 23, 1981, the Nuclear Regulatory Comission approved publication of a proposed rule for public coment that, if approved as a final rule, would require power reactor licensees to provide the maximum amount of property insurance available. The NRC staff will evaluate the coments and will keep abreast of the two current proposals outlined in the GA0 report to increase the levels of property insurance being offered.
You should also be aware of the July 9,1981 proposal from Pennsylvania Governor Thornburgh to move forward with funding the cleanup.
The proposal includes (1) establishment of a private, non-profit institute as a conduit for financial and technological cleanup support, and (2) asking utilities, manu-facturers, and suppliers of the nuclear industry, the Federal government, States of Pennsylvania and New Jersey, and GPU for specified contributtuns to the cleanup. The Governor also is supporting the restart of Unit 1, contingent upon safety assurances, as a basis for raising additional funds to devote to the cleanup.
Finally, the property insurance payout at TMI of $180 million is several months out of date. The report should be updated in regard to these three issues.
Given these subordinate points, the recomendation appears reasonable.
COMMENTS ON THE SECOND RECOMMENDATION The report's second recomendation to the Chairman, Nuclear Regulatory Comission, recomends that in order to mitigate future regulatory constraints on nuclear accident cleanup activities, the NRC should develop a set of guidelines that would facilitate the development of recovery procedures by utility companies in the event of other nuclear reactor accidents.
100 M2W'
'N E S fk
\\
\\
APPDIDIX I APPENDIX I Mr. J. Dexter Peach The recomendation is unclear and not fully clarified by appropriate detail in th Each accident has had and will have unique characteristics body of the report.
which make it difficult to establish a general prescription for cle accident generated water will be a function of considerat activities.
and State and local views.
The GAO report implies disagreement with the course of action (outlined in the 21,1979) to implement its Comission's Statement of Policy of November responsibilities under the Atomic Energy Act and NEP The report under the laws or with the requirements of the laws themselves.
would be more helpful if it elaborated on the basic elements of the recomende guidelines.
Sincerely,
/
6 William'J. Dircks Executive Director for Operations t
l 101
APPENDIX II APPENDIX II Department of Energy Washington, D.C. 20585 AUG 7 1987 Mr. J. Dexter Peach Energy and Minerals Division U.S. General Accounting Office Washington, D.C.
20548
Dear Mr. Peach:
The Department of Energy appreciates the opportuntty to review and comment on the General Accounting Office draft report entitled " Greater Commitment Needed to Solve Continuing Problems at Three Mile Island." The Department of Energy believes the draft report is comprehensive and treats the subject objectively.
The Department of Energy agrees with the report's observation that the General Public Utilities Company, its stockholders and customers, the States of Pennsylvania and New Jersey and their utility regulatory commissions, the Federal Government, and the U.S. utfitty industry all have a role to play in assuring a rapid and safe cleanup. Thus, the job of implementing the cleanup must be a cooperative effort among all the parties.
The Department of Energy believes the final report should reflect that the Governor of Pennsylvania has recently taken a leadership role to secure a l
cooperative approach among all the parties in consonance with the General Accountin Governor'g Office's principal conclusion. The Department believes the s recommendation represents an excellent step toward resolving the funding issue.
i I
The Department is moving forward with a Three Mile Island research and development program to provide valuable data related to nuclear safety and cleanup technology. This program will also provide technical support to the General Public Utilities Company for prompt, safe, and efficient fuel removal and waste handling operations. The Department currently projects expenditures I
of about $75 million over 3 years (beginning this October) for research and I
development in the fuel and waste processing area, and about $10 million per year to acquire data on radioactivity distribution, electrical equipment perfonnance, and other areas related to nuclear safety.
j The Department believes that a.research and development program of this basic scope and size will be of substantial value to the entire nuclear community I
and has infonned the other parties involved in the cleanup of our plans.
1 l
All Department cf Energy research and development programs are nonna11y l
subject to an annual review process and Congressional authorization and appro-i priations are secured on an annual basis. This review process will be useful l
102 A
- 1
..,. a... 4M u...
APPENDIX II APPDiDIX II in assuring the Three Mile Island research and development program remains consistent with the Department's objectives and the needs of the Three Mile Island cleanup program. Therefore, the Department does not believe it neces-sary to seek multiyear funding as proposed by the General Accounting Office.
The Department does believe, however, that the current plans for the multiyear program described above offer sufficient bases for General Public Utilities and the other parties to proceed expeditiously with actions to fund and complete the cleanup effort.
The Department agrees with the r: port's conclusion that there are too many uncertainties in a utility bankruptcy to make a strong case for a bankruptcy action for General Public Utilities.
If anything, the report has under-estimated the potential adverse effects that a bankruptcy action for a General Public Utilities Company would have on the parties involved, including the company, its ratepayers, its creditors, and the electric utility industry as a whole. It is possible that a bankruptcy action could begin even before 1983, the point at which the report's introduction suggests the threat might arise again. The Department also agrees with the report's observation that any advantages that might accrue as a result of bankruptcy can be accomplished without such action at less cost and with greater efficiency.
In short, bankruptcy is an option to be avoided.
The Department agrees that it would be advisable for nuclear utilities to increase the levels of their property insurance coverage so that more funds would be readily available in the unlikely event of a future accident resulting in cleanup costs in excess of present insurance. The Department also agrees with the report's conclusion that the private sector has the resources to provide it without Government intervention and should be given a reasonable amount of time to increase coverage to acceptable levels. The Department will continue to monitor developments in this regard.
The Department does not believe that new regulations and guidelines governing cleanup activities from future accidents should be generated by the Nuclear Regulatory Commission. The Department's view is that standards which govern the allowable releases from nuclear plants to the environment are already adequately specified; delays occurring during the early phases of the TMI cleanup resulted from decisions to depart from these standards and to use more stringent requirements and new regulatory processes not applied to other nuclear powerplants. Rather than request the Nuclear Regulatory Commission to produce a new set of guidelines, the Department of Energy suggests a more appropriate action would be to assure that existing, proven standards are applied in the future.
103
1 l
APPENDIX IV APPENDIX IV BIBLIOGRAPHY OF BANKRUPTCY STUDIES
" Potential Impact of Licensee Default on Cleanup of TMI-2,".
NUREG-0689, Utility Finance Branch, Division of Engineering, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission (November, 1980).
A study designed to recommend steps NRC might take should the licensee become bankrupt and to minimize tne potential of bankruptcy.
" General Public Utilities Corporation Pennsylvania Operations:
Management and Operations Study," Theodore Barry and Associates (September 1980).
A study prepared for the Pennsylvania Public Utility Commission.
An
" Report on Analysis of the Potential Effects of Bankruptcy:
Analysis of Strategic Options for Jersey Central Power and Light Company," Arthur Young and Company (October, 1980).
A study prepared for the New Jersey Board of Public Utilities.
"The impact of a Cnapter 11 Bankruptcy Proceeding on Cnrysler Corporation," Professor Frank Kennedy, University of Micnigan Law School (April, 1980), an unpuolished memorandum.
O G
g (309337) 106 I
m%
APPDIDIX II APPDIDIX II The Department of Energy appreciates the opportunity to comment on this draft report and trusts the General Accounting Office will consider the comments in preparing the final report. Comments related to the text of the report and an annotated copy of the draft report have been forwarded separately.
Sincerely, m ' i ~ : 4. / ? / lgfo' %
William S. Heffelfinger Assistant Secretary Management and Administration l
l l
l 104
1 APPENDIX III APPENDIX III l
DEFINITICN OF POODY's RATING SYMBOLS Debt Preferred Stock Aaa = Best Quality; intere'st and "aaa" = 'Ibp quality; good asset principal exceptionally protection and least dividend imoairment.
secure.
Aa = High quality; margins of "aa"
= High grade; reasonable protection may rot be as assurance of well large as in Aaa bonds maintained earnings and asset protection in foreseeable future.
A
= Upper medi m grade; many "a"
= Upper medim grade; favorable investment attri-earnings and asset Iro-butes; security principal tection expected to and interest adequate but reain adequate.
may be susceptible to impairment in future.
Baa = Medim grade; neither "baa" = Medim grade; protection highly protected or poorly adequate for present but secured.
may be questionable over lorg term.
j Ba = 'Ihese have speculative "ba"
= Speculative el ments; elments; not well safe future cannot be consi-guarded durirg both good dered well assured; char-and bad times.
acterized by uncertainty.
B
= Lack desirable investment "b"
= Lack desirable investment characteristics; assurance characteristics; assurance of interest and principal of dividend payments anS pyments over any larg maintenance of other terms period of time may be small.
wer any lors pericx3 of time may be small.
l Caa = Poor standing; may be in "caa" = Likely to be in arrears default or may have danger on dividend gyments; with respct to principal does not rule out future or interest.
dividend payments.
Ca = Speculative in a high degree; may be in default.
C = Iowest rated borr$s; extreely l
poor prospects of ever attaining real investment standire.
'Ihese ratings may be modified by the " addition of a plus or minus sign to show relative standing within the major ratirg categories.
105 l
\\
-