ML20244D871
| ML20244D871 | |
| Person / Time | |
|---|---|
| Issue date: | 05/13/1986 |
| From: | Eisenhut D Office of Nuclear Reactor Regulation |
| To: | Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20237J265 | List: |
| References | |
| FOIA-87-416, RTR-NUREG-CR-3971, RTR-NUREG-CR-4012, RTR-NUREG-CR-4568 NUDOCS 8605220345 | |
| Download: ML20244D871 (26) | |
Text
..
pR Rf G
'o UNITED STATES f
8' NUCLEAR REGULATORY COMMISSION n
WASHINGTON, D. C. 20555 3
s I
/
May 13, 1986 MEMORANDUM FOR:
All NRR Employees j
FROM:
Darrell G. Eisenhut, Acting Director l
Office of Nuclear Reactor Regulation I
l
SUBJECT:
NRR OFFICE LETTER NO. 16, REVISION 3 -- REGULATORY ANALYSIS l
GUIDELINES i
l This revised office letter supersedes the October 3, 1984, version (Revision 2) of NRR Office Letter No. 16.
In this version the letter and enclosures have been j
generally updated and specific changes were made to:
l (a) Accommodate the NRR reorganization j
(b) Incorporate NRR endorsement of NUREG/CR-3971, " Cost Estimating Handbook" and NUREG/CR-4568, "A Handbook for Quick Cost Estimates" l
(c) Incorporate NRR endorsement of NUREG/CR-4012 " Replacement Power Costs" (d) Incorporate interim guidance for the preparation of backfit analyses for proposed rulemakings or the imposition of new or different regu-latory staff positions interpreting the Commission's rules as required by the new backfit rule (10 CFR 50.109).
The Executive Director for Operations (E00) has issued NRR-wide regulatory analysis guidelines.
These are contained in NUREG/BR-0058, " Regulatory Analysis Guidelines for the U.S. Nuclear Regulatory Commission."
(The current version of NUREG/BR-0058 is Revision 1, dated May 1984).
NRR will prepare regulatory analyses, in accordance with NUREG/BR-0058, in support of covered submittals for review to the Committee to Review Generic Requirements (CRGR), or to the Deputy Executive Director for Regional Operations and Generic Requirements (DEDR0GR), or for decision to the i
EDO or the Commission.
Covered submittals include two categories, as specified l
in NUREG/GR-0058.
c I
1.
Major proposed and final rules, meeting threshold criteria stated in l
NUREG/BR-0058, Section IIA.
There the guidelines apply fully.
j 2.
Other rulemaking actions and non-rulemaking generic requirements or
{
guidance (referenced in Section IIB of NUREG/BR-0058).
For these much 4
less detail is required.
The extent of detail should be commensurate f
with the safety importance of the issue, the estimated magnitude of the j
proposed action's impact, and the complexity and analytical tractability of the issue.
l On September 20, 1985, the Commission published a final rule (10 CFR 50.109) l commonly called the "backfit rule." The Commission decided that a new, more spe-cific standard and related procedures should be applied to backfitting decisions.
i N.,
e-i f b. T) ( %
f}
(
1 i
M xo t
~
e May 13, 1986 I
the unusual case that the new requirement or guidance is to be " forwarded fit" only, then a backfit analysis is not needed.
The "backfit analysis" is to be
)
provided with the rulemaking package or the CRGR package when the package is submitted for publication in the Federal Register for public review and comment.
NRR experience with the development of a "backfit analysis" is very limited, hence, final guidance for the content of the "backfit analysis" can not be pro-vided to you at this time. to this office letter provides interim guidance for the preparation of a "backfit analysis." As more hands-on experi-ence is gained through the actual preparation of "backfit analysis," final guid-ance will be developed.
This office letter will be revised to provide final guidance to the staff when it is available.
As a part of the backfit rule, paragraph 50.54(f) of the Commission's regulations pertaining to requests for information from licensees has also been revised.
If extensive licensee effort is anticipated, the staff must show that the burden imposed by the information request is justified in view of the potential safety significance of the issue to be addressed.
The revision further provides for management control and accountability by requiring that the staff evaluation be reviewed by the ED0 prior i
to the issue of the request.
Accordingly, a regulatory analysis package should now be j
provided for all 50.54(f) information requests, specifically addressing the nine j
factors identified in the backfit rule, to the extent practical, as indicated above.
1 The revision to 650.54(f) exempts information sought to verify licensee compliance i
with the licensing basis for that facility, from the above requirement.
However, new, or modified interpretations of what constitutes compliance would not fall within this exception, and an analysis would be required.
It should be noted that 650.54(f) j does not by its terms apply to the review of applications or amendments.
j i
I In accordance with the Commission's Policy and Planning Guidance for 1984 (NUREG-0885, Issue 3), safety goals and associated numerical guidance will be used only as the Commission may direct in the future.
j i
i 1
I l
drre is nh c ing Director Office of clear Reactor Regulation l
Enclosures:
- 1. Supplementary Guidance for NRR
- 2. Decision Making Using Value-Impact Analysis
- 3. Decision Factors Supplementing Value-Impact Ratio
- 4. Procedure for Requests for Cost Analysis Group Participation in Regulatory Impact Analysis Process
- 5. Interim Guidelines for the Preparation of a Backfit Analysis cc:
V. Stello J. Sniezek R. Minogue J. Taylor J. Davis 1
w o
s
[(pa ncg[o UNITED STATES g
g NUCLEAR REGULATORY COMM'SSION 7;
aE WASHINGTON, D. C. 20555
%*****/
May 13, 1986 MEMORANDUM FOR:
All NRR Employees FROM:
Darrell G. Eisenhut, Acting Director Office of Nuclear Reactor Regulation
SUBJECT:
NRR OFFICE LETTER NO. 16, REVISION 3 -- REGULATORY ANALYSIS GUIDELINES This revised of fice letter supersedes the October 3,1984, version (Revision 2) of NRR Office Letter No. 16.
In this version the letter and enclosures have been generally updated and specific changes were made to:
l l
(a) Accommodate the NRR reorganizati_on 1
(b)
Incorporate NRR endorsement of NUREG/CR-3971, " Cost Estimating l
l Handbook" and NUREG/CR-4568, "A Handbook for Quick Cost Estimates" l
(c)
Incorporate NRR endorsement of NUREG/CR-4012 " Replacement Power Costs" (d)
Incorporate interim guidance for the preparation of backfit analyses for proposed rulemakings or the impositico of new or different regu-latory staff positions interpreting the Commission's rules as required by the new backfit rule (10 CFR 50.109).
The Executive Director for Operations (E00) has issued NRR-wide regulatory analysis guidelines.
These are contained in NUREG/BR-0058, " Regulatory Analysis Guidelines for the U.S. Nuclear Regulatory Commission."
(The current version of NUREG/BR-0058 is Revision 1, dated May 1984).
NRR will prepare regulatory analyses, in accordance with NUREG/BR-0058, in support of covered submittals for review to the Committee to Review Generic Requirements (CRGR), or to the Deputy Executive Director for Regional Operations and Generic Requirements (DEDROGR), or for decision to the EDO or the Commission.
Covered submittals include two categories, as specified in NUREG/BR-0058.
1.
Major proposed and final rules, meeting threshold criteria stated in NUREG/BR-0058, Section IIA.
There the guidelines apply fully.
2.
Other rulemaking actions and non-rulemaking generic requirements or guidance (referenced in Section IIB of NUREG/BR-0058).
For these much less detail is required.
The extent of detail should be commensurate with the safety importance of the issue, the estimated magnitude of the proposed action's impact, and the complexity and analytical tractability of the issue.
On September 20, 1985, the Commission published a final rule (10 CFR 50.109) commonly called the "backfit rule." The Commission decided that a new, more spe-cific standard and related procedures should be applied to backfitting decisions, w
)).
- s
(
1 N - ) x %' o '
a
l May 13, 1986 The new rule sets forth restrictions on imposing a new or revised provision-in the Commission rules or t5e imposition of a new or different regulatory staff
_j r
position interpreting the Commission rules after:
(i) the date of issuance of i
the construction permit (CP) for the facility for plants having cps issued after i
October 21, 1985; or (ii) six months before the date of docketing of the oper-ating license (0L) application for the facility for plants having cps issued before October 21, 1985; or (iii) the date of issuance of the OL for the facility for plants having OLs; or (iv) the date of issuance of the design approval under l
Appendix M, N, or 0 of 10 CFR Part 50.
This regulation sets forth the following requirements:
- 1. [S50.109(a)(2)] "The Commission shall require a systematic and documented analysis pursuant to paragraph (c) of this section for backfits which it seeks to impose."
- 2. [S50.109(a)(3)] "The Commission shall require the backfitting of a facility I
only when it determines, based upon the analysis described in paragraph (c) l of [S50.109], that there is a substantial increase in the overall protection of the public health and safety or the common defense and security to be derived from the backfit and that the direct and indirect costs of imple-mentation for that facility are justified in view of the increased protection."
In order to reach this determination, Paragraph S50.109(c) sets forth nine specific factors which are to be considered in the analysis for the backfits it seeks to impose.
The required content of CRGR review packages and procedures for submitting them
)
are described in NRR Office Letter No. 39.
Regulatory analyses constitute a part 1
of such packages (Item 5.1 in Section II of Office Letter 39).
Preparation of the regulatory analyses is the responsibility of the lead Division within NRR for CRGR submittals, in accordance with NRR Office Letter No. 39.
The Division of Safety Review and Oversight (DSR0) will., as requested by the lead Division, provide 1
assistance in interpreting the guidelines and consultation on regulatory analysis techniques and serve as focal point for NRR requests for assistance from the Cost Analysis Group.
The lead Division will coordinate with DSR0 before submitting a regulatory analysis to the Director, NRR, in accordance with Office Letter No.
39.
The purpose of this letter is to supplement the NRC-wide regulatory analysis guidelines (NUREG/BR-0058) in some areas with more specific NRR guidance.
This l
supplementary NRR guidance is provided in Enclosure 1 through 4'.
In addition, we have interpreted the new backfit rule with regard to its impact I
on regulatory analysis.
The nine specific factors codified in the new trackfit rule are essentially specific statements of those facters already considered in the development of an NRR regulatory analysis package.
Henceforth all regulatory analysis packages developed in NRR should reference the backfit rule and specifi-cally address each of the nine factors identified in the rule.
For a rulemaking package and any staff position interpreting the Commission's rules which will be issued for public comment, such as Regulatory Guides, Standard Review Plans, Standard Technical Specifications, or orders (or revisions thereto), a "backfit analysis" is to be provided in addition to the regulatory analysis package.
In l
1
_. A
- I
. May'13, 1986 q
d the unusual case that the new requirement.or guidance is to.be " forwarded fit" only, then a backfit analysis is not needed.
The "backfit analysis" is to be i
provided with the rulemaking package or the.CRGR package when the package is submitted for publication in the Federal Register for public review and comment.
j NRR experience with the. development of a'"backfit analysis" is very limited, j
hence, final guidance for the content of the "backfit analysis" can not be pro-vided to you at this time.
Enclosure'5 to this office letter provides interim guidance for the preparation of a "backfit analysis." As more hands-on experi-ence is gained through the actual preparation of "backfit analysis," final guid-ance will be developed.
This office letter will be revised to provide. final guidance to the staff when it is available.
As a part of the backfit rule, paragraph 50.54(f) of the Commission's regulations pertaining to requests for information from licensees has also been revised.
If extensive licensee effort is anticipated,'the staff must show that the burden imposed by the information request is justified in view of the potential safety significance of the issue to be addressed.
The revision further provides.for management control and accountability by requiring that the staff evaluation be reviewed by the ED0 prior to the issue of the request.
Accordingly, a regulatory analysis package should now be provided for all 50.54(f) information requests, specifically addressing the nine factors identified in the backfit rule, to the extent practical, as indicated above.
The revision to 650.54(f) exempts information sought to verify licensee compliance with the licensing basis for that facility, from the above requirement..However, new, or modified interpretations of what constitutes compliance would.not fall within this exception, and an analysis would be required.
It should be rated that $50.54(f) does not by its terms apply to the review of applications-or amendments.
j In accordance with the Commission's Policy and Planning Guidance for 1984 (NUREG-0885, Issue 3), safety goals and associated numerical guidance will be used only as the Commission may direct in the future.
~
r f
i is nhe(M t,
ing Director rre Office of clear Reactor Regulation l
1
Enclosures:
i
- 1. Supplementary Guidance for NRR I
- 2. Decision Making Using Value-Impact Analysis
- 3. Decision Factors Supplementing Value-Impact Ratio
)
- 4. Procedure for Requests for Cost Analysis Group Participation in Regulatory Impact Analysis Process
-]
- 5. Interim Guidelines for the Preparation of a Backfit i
Analysis q
cc:
V. Stello f
J. Sniezek R. Minogue J. Taylor i
J. Davis J
1 a
I
ENCLOSURE 1 SUPPLEMENTARY GUIDANCE FOR NRR l
l NOTE:
Section numbers refer to correspondingly numbered sections of l
NUREG/BR-0058, Revision 1, " Regulatory Analysis Guidelines for l
the U.S. Nuclear Regulatory Commission," May 1984.
l j
III.B.1.
Statement of Problem 1
This section will include a clear definition of the scope and bounds of the issue to which the proposed action is addressed.
This is essential in arriv-ing at a sound and applicable analysis since issues are often complex and interrelated with other issues.
I III.B.4.a. Costs and Benefits of Alternatives i
i This will always include comparison of the' proposed action with the no-action alternative.
Other alternatives will be covered only where significant other alternatives exist.
The no-action alternative should reflect implementation i
i of all applicable requirements or guidance to date, even if implementation is j
not completed at the time of analysis.
I
.l i
Where the effect on different classes of plants varies substantially, separate j
estimates will be made and separate conclusions drawn for each class.
The analysis should show that the requirement or guidance is directed and focused as narrowly as practical.
Where additional information must be obtained from industry to help bound an affected class of plants, obtaining that information should be justified on the basis 01 its potential safety significance and cost.
The regulatory analysis should be reopened for reconsideration should the new information bring into question the soundness of the original conclusion.
l l
l
. The required analysis of costs and benefits constitutes a "value-impact" assessment.
NUREG/CR-3568, "A Handbook for Value-Impact Assessment," may be used as a reference on methods for performing such assessments.
Professional judgment will be required in choosing among the options presented in the Handbook, in adopting methods to the issue at hand, and in applying methuds not specifically covered when special aspects of the subject matter make that appropriate.
i A general discussion of decision making using value-impact 3ssessment is enclosed (Enclosure 2 to Office Letter No. 16).
l 4
The Costs and Benefits section should ordinarily be organized as outlined j
below.
This outline is primarily for safety issues, but for issues involving other subjects (environmental protection licensing and regulatory process improvement, etc.), following a similar (or analogous) outline is vtggested.
The outline follows:
l l
(i) Value -- Risk Reduction Estimates The value is the safety importance of an issue and is usually represented by the change in expected risk that resolution could effect.
Risk is ordinarily expressed in terms of the product of the frequency of an occurrence and the public dose (in man-rem) that would result in the event of the accident.
If more than one accident scenario is important, the risks are summed.
Value is attributed to reductions in the average, mean, or expected risk.
The dose is calculated for the 50-mile-radiur area around the plant.
The public man-rem based estimate may not be the only appropriate measure of an issue's safety importance in all cases.
Alternative measures of safety importance should be used when appropriate.
For example, when a possible core melt or other accident is involved but release outside containment would be minor, highly improbable, or highly uncertain in
. probability or magnitude, contribution to the core-melt probability should be estimated in addition to (or even in lieu of) the probabilistic public dose effect.
There is controversy as to whether such " contained accidents" are a sufficient basis for regulatory action.
Since core celts usually have a significant probability of failing containment, this is not an issue for most core melts.
However, other accidents can be contained.
The infor-j 1
mation on significant contained accidents should be displayed in the analysis, J
l but the analyst must weigh the significance of this factor in deciding to what extent, if any, it should influence the conclusion.
It should be noted that safeguarding against an excessive core-melt frequency may justify impo-l ition of a safety requirement, even without demonstration of offsite consequences, j
l where significant occupational exposure is incurred in implementing reso-
{
i lution f a safety issue, such exposure is taken into account, but stated l
separately.
Such exposure is viewed as a negative component in the net risk-reduction value.
j 1
I Where an accident may result in significant occupational exposure due to the accident directly or in post-accident plant cleanup, such exposure is taken into account and stated separately.
For plant-wide cleanup after l
a contained severe core-damage accident causing substantial wide-spread i
plant contamination, NUREG/CR-35681 and/or NUREG/CR-28002 may be used to estimate occupational exposure when no better issue-specific estimate is available.
l Usually public dose is an adequate surrogate for off-site contamination, j
However, the analyst should be alert to circumstances in which radiological l
effects not adequately reflected in the public dose as surrogate may be present. Examples include on-site damage, occupational exposure, liquid pathway effects, and supportive medical treatment, among others.
Long-term contamination of the environment including water bodies and other agri-cultural or industrial facilities should be taken into account where that is a principal effect.
NUREG/CR-3568 "A Handbook for Value-Impact Assessment," December 1983.
" Guidelines for Nuclear Power Plant Safety Issue l
2 Prioritization Information Development," February 1983.
j i
__o
I l l (ii) Impact -- Cost Estimates i
I l
NUREG/CR-3971, " Handbook for Cost Estimating:
A Method for Developing Estimates of Cost for Generic Actions for Nuclear Power Plants," may
)
be used as a methodological aid to the performance of the required cost l
analyses and a guide to some widely useful data sources.
In applying the estimating approach described, professional judgment will be required in choosing a degree of detail of cost analysis that is appropriate to the total costs involved, the complexity of the issue, and closeness of I
l the value-impact trade-off.
In addition, professional judgment will be I
i required in properly identifying and characterizing the functions to be j
estimated, in critical application of the data sources, and in seeking cost information from sources not identified in the Handbook.
l t
i l
NUREG/CR-4568, "A Handbook for Quick Cost Estimates," presents a quick, I
but comprehensive, introduction to the subject and can be used to l
describe a quick cost estimate and to highlight the portions of the cost breakdown that potentially introduce the greatest uncertainty.
In January 1984, the E00 established a charter for the Cost Analysis Group (CAG) formed in the Office of Resource Management.
NRR staff is t
encouraged to seek CAG advice and support, as needed and available, to I
1 help develop sound cost analyses for use in regulatory analyses.
Requests to CAG should be coordinated through the Division of Safety Review and Oversight.
The Safety Program Evaluation Branch is the contact point.
Procedures are described in Enclosure 4 to Office Letter No.16, " Procedure for Regtests for Cost Analysis Group Participation in Regulatory Impact Analysis Process." Dependability, in terms of guarding against omission of important or even dominant cost elements, or against l
inclusion of costs that are not in fact entailed, is more important than high precision of the estimates.
. It is particularly important to determine whether plant downtime (or prolongation of downtime incurred for unrelated reasons, such as refueling) will be involved and, if so, to estiinate well the duration of the outage.
Downtime costs, when involved, are often larger than all other costs.
NUREG/CR-4012, " Replacement Energy Costs for Nuclear Electricity Gener-
)
ating Units in the United States," presents plant specific replacement J
l energy cost estimates applicable to short-term outages in the 1985-86 time frame.
These estimates should be relied upon when the identifi-cation of specific reactors is warranted, and when a high degree of l
precision in cost estimation is required.
The information may also be generally used as input to estimates prepared for other periods and for longer outages, subject to required professional judgment in adjusting the estimates according to the circumstances.
A $500,000/ day approx-imate outage cost figure may continue to be used where more detailed j
estimates are not warranted.
i i
l It should be borne in mind that procedural changes and professional work are not cost-free, and may have costs that are substantial, i
Both industry and NRC costs (as well as costs to others when involved) should be estimated.
Transfer costs (e.g., insurance) need not be calculated in NRC regulatory analyses since our perspective is a net national impact.
In some cases, plant-damage costs averted by the proposed action can substantially affect the cost-benefit evaluation.
Estimates for such
)
averted costs are developed and used in separately stated calculations,
{
so that the results, both with and without adjustment for averted plant-l damage costs, are readily apparent.
The averted costs may include those of averted equipment failures, limited-time plant outage, or limited
)
plant-contamination cleanup.
In the extreme, they can also include averted permanent loss of use of the plant and plant-wide cleanup, I
multiplied, in each case, by the reduction in frequency of such events that would be brought about by resolution of the generic safety issue.
l l
. i I
In the absence of better issue-specific estimates, the estimated cost of' plant-wide cleanup, before discounting to present worth, may be taken as $1.2 billion (1984 dollars; based on TMI estimates).
While again a controversial issue, plant damage is to be considered an impact and not a value in NRR analyses.
Should a favorable conclusion about a proposed generic requirement be contingent on taking plant damage into account, the analyst should state that conclusion, together with a statement of the sensitivity of the conclusion to this factor.
j a
When the discount rate is an important factor in the evaluation of a j
proposed action, the sensitivity of the results to the discount rate l
4 should be tested, as suggested in NUREG/BR-0058 (III.B.4.a).
Alterna-tive real discount rates should include a 5 percent rate, in addition to q
the 10 percent rate specified in NUREG/BR-0058.
High discount rates I
I reduce the impact (present worth) of future accidents but also reduce t.he impact of the continuing costs assoc:ated with implementing new requirements or guidance.
(iii)
Value/ Impact Ratio i
The total net safety value of the proposed action, typically in man-rem of public dose avoided, is related to total net costs (NRC, industry, plus any other) in terms of a ratio, typically dollars / man-rem.
This ratio, along with safety importance, can be used as a supplementary basis for comparing alternatives, including evaluation against the no-action alternative, and ranking for implementation priority in relation to other issues.
(iv) Uncertainty Bounds Major sources of uncertainty in the benefits and costs should be identi-fied and judgments as to their quantitative significance indicated, as information warrants.
However, particular attention should be paid to
e those facters that have associated uncertainties of a substantially different nature or magnitude than those encountered in typical risk or-cost analyses.
(v) Special Considerations i
The value-impact analyses should be as quantitative as the situation reasonably permits.
Items i to iv, above, emphasize the quantitative aspects.
However, while the calculated risks and value-impact ratios are often valuable aids to judgment, other considerations not adequately reflected or not reflected at all in the basic numerical formula employed i
are often helpful in corroborating or adjusting the results.
Decision i
making is helped by explicit identification of such other considerations and explanation of how they bear on the conclusions.
A partial list and discussion of decision factors which might supplement the calculated risk and value-impact ratio is enclosed (Enclosure 3 tc Office Letter No. 16).
l l
l III.B.S. Decision Rationale l
This section should include a recapitulation of the main points of the ration-l ale underlying the value-impact conclusion concerning the prcposed action, including critical information of importance to the safety (or other) values j
expected to be obtained, costs, value-impact relation, uncertainties, and any special considerations.
I Rules and other existing requirements are not, in themselves, justification for a decision; they can be changed (by the Commission or authorized staff) on just the sort of regulatory analysis basis as is involved here.
For requirements established by law, the regulatory analysis should, when appro-priate, weigh alternative means of implementing the law.
l
\\
l ENCLOSURE 2 i
DECISION-MAKING USING VALUE-IMPACT ASSESSMENT l
I.
General Principles of Cost-to-Benefit Comparisons The decision process and decision considerations are similar when cost-benefit comparisons are close and when they are not.
The value-impact approach, as a disciplined process, employed in conjunction with good engineering analysis, helps guard against errors, regardless of the rela-tive magnitudes of costs and benefit.
The calculated cost-benefit -- or value-impact ratio, whether nearly balanced or unbalanced, is a factor in the decision, but should not.be regarded as the lone or even the generally determinative factor.
Other factors such as special ricks or costs, subjective perceptions, time-g related factors, regulatory stability or implementation feasibility can i
be important.
These factors, when present to some particular extent, are more likely to be controlling when the value-impact comparison is
-j close than when values and impacts are unbalanced.
"i i
l All decisions based on risk or cost analyses must recognize that the uncertainty of the estimates of the factors can be -- and in most cases j
{
is -- large.
Uncertainty includes not only the variability of data, but j
also incompleteness, whether from ignorance or error.
When there is a large imbalance between safety values and impacts, uncertainty is less important.
For close balances, uncertainty is likely to be an important I
consideration.
The values considered should be strictly limited to those results affect-ing the protection of public health, safety and property which the NRC is authorized to regulate.
All other consequences of a proposed action should be included only as impacts.
Even if the net impact is favorable, this, by itself, is not a justification for imposing an action.
The
.g-absolute safety value needs to be taken into account.
Issues of small i
l or no safety value usually are not worthwhile considering, unless regula-tory requirements are violated, since there is an irreducible minimum cost associated with any issue and trivial safety values are therefore not cost effective.
In addition, while issues of small value may have j
i small overall costs (and therefore appear to be justified), these costs n
are not uniformly distributed.
For example, in most cases of small issues, these costs fall primarily on the licensing staff of the utility
)
and the NRC.
These staffs are limited and the proliferation of issues with small safety values would divert them from significant issues which may have much greater overall costs, but not much greater use of a limiting resource such as licensing staffs.
Reasons such as these for dismissing issues of small value should be explicitly stated in the.
analysis.
2.
No Set Formula i
l l
l The significance of various potential decision factors and their inter-j play vary according to the issue.
All relevant and significant factors I
l v
should be recognized and taken into account according to the facts'snd the issue.
Some general guidance can be developed, but a prescribed and ostensibly exhaustive checklist or weighting formula is likely to con-strain decision making in counterproductive ways.
The analyses. involved, especially when the decisions are close, are often complex and difficult.
They require exercise of judgment, for which there is no effective-simpler j
substitute.
But value-impact analyses, or at least one case for all I
analyses, should usually be done with a standard method and set of assump-tions to allow meaningful comparisons between issues; work is naeded to develop appropriate assumption sets.
Other cases with assumptions or methods most appropriate to the issue under study can then be done.
l L.___.__m._____
^
3.
Development and Display of Facts No decision involving value-impact considerations'can be dependable unless it rests on a sufficiently' comprehensive factual analysis of values and impacts.
This is, of course, especially true for clece decisions.
l "Sufficiently comprehensive" cannot he defined beforehand, but must be 1
l judged for.each issue.
Methods and procedures for verifying the accuracy and completeness of the factual analysis must be used.
These include internal and public review and comment.
This assurance of accuracy and completeness is most important when the decision is close.
However, a lack of data or uncertainty of the data used to determine j
values and impacts cannot be used as a basis to defer a decision since both action and inaction are decisions.
One course or another must be taken on the basis of the information currently available, no matter how l
good or poor it may be.
If the information base is weak, the likelihood
.l that inaction (as well as action) may be wrong needs to be carefully considered.
The analysis should include:
i (a) Identification of value and impact elements that may be significant for the issue.
1 (b) Realistic quantitative estimates, wherever estimates are reasonably possible.
(c) Estimates of uncertainties, where practical, and comments on any 1
unusual aspects in the nature and structure of uncertainties.
(d) Qualitative description of unquantified value and impact elements.
l
')
1 i
4 The bases and assumptions for all calculations and qualitative state-ments should be made clear since decision outcomes may be sensitive to them. Where important alternative bases and assumptions exist, sensi-tivity to them should be explored.
i i
4.
Conclusions l
l The basis for the decision should be stated.
Trade-off factors, what factors and analytical results were considered, rnd how the various considerations entering a decision were weighed should be. stated explicitly.
I Decisions to take no action are usually more difficult.
One, or a few identified possible sequences that contribute to a significant risk, may l
be sufficient to justify a decision to require action.
A no-action j
decision requires as much care as on action decision, in order to ensure that no significant factors have been omitted.
This is particularly true of close no-action decisions.
l I
5.
Reverse Decision Analysis Reverse analysis may be used to determine whether a value-impact imbalance could be upset by countervailing considerations.
(A " reverse calculation" coicerns the question, "What facts would justify the action?",
i rather than "WNat action do the facts justify?") The advantage of such j.
an approach is that it diminishes pressures for quantifying factors that t
i are problematical to quantify.
One need not determine what quantitative l
value to assign to an unquantified factor-it suffices to determine 1
whether the quantitative value should be above or below the value that would establish value-impact equipoise.
l t
l 1
P ENCLOSURE 3 a
DECISION FACTORS SUPPLEMENTING CALCULATED VALUE-IMPACT RATIO j
t i
A calculated valo 4"1 pact comparison generally reflects a necessarily simpli-fied evaluation, focusing on public dose and direct cost impact.
The possible l
importance of the factors and the limitations of an often incomplete and imprecise data base point to a need to corroborate or adjust the formula results by other considerations.
Some such effects -- occupational exposure, averted plant-damage costs, uncertainty' bounds -- require careful consider-ation for all issues.
Others thought to be significant should be identified _
j and considered-quantitatively when practical, but at least qualitatively.
Some special considerations may be quite specific to an issue.
This list is l
i not complete and the analyst should assure that all significant and relevant factors are considered.
'l l
l 1.
Special Risk and Cost Aspects l
J 5pecial risk and cost aspects potentially affecting net value or net l
impact, ~but not routinely included in numerical formula, are as follows:
Special Risks and Value Aspects (a) A significant net change in occupational doses.
[See discussion of occupational exposure in a preceding enclosure (Enclosure 1 to Office Letter No. 16)].
(b) Loss or severe degradation of a layer in the defense-in-depth concept (e.g., all but one mode of core cooling, or all modes of containment cooling).
l
~
1
-2 I
(c) Circumstances imparting unusual significance.to radiological or other accident consequences (such as ingestion pathway effects or great psychological stress) or mitigating measures (such as evacu-ation, sheltering, or supportive medical attention) that are not directly included in the public dose calculations, i
(d) Enhancement or impairment of the value of the proposed action by environmental or safeguards benefits or impairments in addition to its safety value.
I (e) Potential for substantial unusual off-site damage aspects not adequately reflected in the public radiation dose as surrogate.
Special Costs and Other Impacts (a) Any significant non-radiation related occupational risk.
l (b) Averted cost of plant damage and outage, not only from the postulated accident, but also its precursors.
L j
(c) Potential for substantial secondary cost impacts not reflected in the cost estimates.
t l
(d) Significant impact on small entities (protected by the Regulatory Flexibility Act).
l 2.
Uncertainties i
Factors related to uncertainties stemming from an incomplete or impracise l
l data base for the value-impact formula are as follows:
1 (a) Uncertainty bounds, imbalance in uncertainty factors, certainty of cost to'fix versus uncertainty that safety is really improved and I
the true extent of such improvement.
l l
t j
. (b) Situations where uncertainty is extraordinarily large (in accident probability or consequences or in cost, or any or all of these),
for example, where unusually great plant-specific variability of risks and costs defeats dependability of generic estimates, or f
where data about a risk are lacking.
(c) The potential for a proposed change to affect more than one acci-dent or transient sequence, thus affecting risk to a greater or lesser degree than assessed in the current description of the issue; notably, the potential for a new safety decrement, or increase in l
risk, due to suspected unidentified effects of a proposed change, or added complexity, or for other reasons.
1 I
l (d) The value inherent in reduction of uncertainty.
(e) The potential for human intervention, using available equipment.
3.
Subjective Perceptions Perceptions and judgments that cannot (or cannot readily) be quantified are as follows:
(a) Public concern about a particular issue, or special Commission or Congressional concern.
The effect that a proposed regulatory action (or inaction) may have on such subjective' perceptions can be quirk-some:
additional safety measures may be perceived as confirmatory of a concern and heighten it rather than quell it.
Usually, NRR decisions should be based on objective factors, but a discussion of such subjective factors can be provided as supplemental information.
(b) Acute knowledgeable professional controversy concerning the impor-tance of an issue or modes of dealing with it, or about the imple-mentation costs of the proposed action.
l
4
- i 4.
Time-Related Factors l
1 Time-related factors are as follows:
(a) How rapidly the action can be implemented.
(b) The potential for a better resolution becoming available later, or j
for an issue disappearing in light of future improved knowledge l
(balanced against (c), below, when applicable).
]
(c) Potential substantial deterioration of the value/ impact ratio while j
awaiting a better regulatory resolution (e.g., a potential design' 5
fix that is inexpensive to apply'before construction, much more i
expensive after the plant is largely built, and extremely expensive and problematical to apply to an operating plant).
j 5.
Regulatory Stability Regulatory stability considerations are as follows:
(a) Consistency with the principle of keeping the number of alterations to existing requirements to a reasonable minimum.
(b) The extent to which the proposed action affords better prospects for subsequent regulatory stability and predictability.
(c) The extent to which licensee latitude is restricted'.without good reason.
___m____._____
4 6.
Implementation Feasibili'ty Implementation feasibility considerations are as follows:
(a) The ease of implementation from technical, human-factors, and I
regulatory standpoints.
)
(b) How readily compliance is verifiable.
i (c) Vulnerability of the proposed change to lapses in implementation.
l i
-l ENCLOSURE 4 PROCEDURE FOR REQUESTS FOR COST ANALYSIS GROUP PARTICIPATION IN REGULATORY IMPACT ANALYSIS PROCESS A.
BACKGROUND 4
By a memorandum dated January 4, 1984, the ED0 established a charter for I
the Cost Analysis Group (CAG) formed in the Office of Resource Management and defined the CAG's role in the regulatory impact analysis process.
The CAG will-serve as the center of cost-estimate development expertise for the NRC.
The CAG is now functioning, has work under way'to develop certain generic cost estimates, and is prepared to offer guidance and support to NRC offices, including NRR, in development and review of co'st estimates on a selective basis.
l B.
POLICY The NRR staff is encouraged to seek CAG advice and support, as needed and available, to help develop accurate and dependable cost analyses for use in regulatory analyses and other pertinent contexts, such as priori-l tizations of generic issues or multi plant actions.
l C.
LEVELS OF SUPPORT l
The following levels of support may be available from the CAG, depending on resource availability:
(a) CAG provides general guidance on cost estimating methods and/or cost information on a generic basis.
1
, (b) CAG is available on call to confer.and advise.
(c) CAG has advisory responsibility in development of the cost analysis.
(d) CAG provides the cost analysis.
CAG also has the following review and oversight function:
(e) CAG reviews the completed cost analysis of others.
D.
NRR C0 ORDINATION l
The Division of Safety Review and Oversight is designated as the focal point for coordination of NRR requests for CAG services.
The Director, i
DSRO, has identified the Safety Program Evaluation Branch as the contact J
1 point.
1 DSR0's responsibilities include coordination of NRR priorities with respect to CAG support needs and advice to NRR divisions concerning decisions to seek CAG participation and the level of CAG support to be sought.
l E.
PROCEDURES l
The following procedures will apply:
1.
Cost analysis for potential generic requirements remains an integral part of the regulatory analysis process and remains the responsibility of the lead NRR division.
(See NRR Of fice Letter 39 and Enclosure 1 to this office letter).
l l 1 I
2.
When cost analysis work is _ initiated as part of a regulatory j
analysis, the lead NRR division will notify CAG, through DSR0 l
(attention: SPEB), of the subject of the cost analysis and its schedule.
This notification is required regardless of whether CAG support is sought.
j 3.
(a) If the lead NRR division desires CAG support other than review of completed cost analyses, it will transmit a timely request for such support to CAG through DSR0 (attention: SPEB).
This request should ordinarily be combined with the notification under Item 2, above.
The request will identify the subject of the cost analysis, the nature and detail of cost analysis required, the nature of CAG support sought, schedule (including when the technical input will be available and when the cost analysis is needed), and principal lead division contact person.
A priority for the cost estimate will be included.
i Normally the priority will be the same as for the generic i
l issue.
(b) After acceptance of the work request by the CAG, the lead division should communicate "ith the CAG directly, t
keeping DSR0 posted on progres:.nd any major problems.
l (c) It should be noted that the CAG charter permits the CAG to reduce or eliminate any advisory role that it may have during the development of a specific cost analysis if l
other pending issues should have priority call on CAG's resources.
DSR0 will be responsible for representing NRR l
views as to priorities in any such determination and will 1
bring to the attention of the Director, NRR, any conflict-ing views as to priorities that cannot be resolved at i
the Division level.
j
. 4.
All completed regulatory analyses will be transmitted to the CAG, with a request for review of the cost aspects, con-currently with transmittal of the analyses to NRR Division l
Directors for review prior to submittal to the Director, NRR, i
l (Office Letter No. 39,Section III, Phase 1).
This will I
ordinarily be a copy of the same regulatory analysis package that is sent to the NRR Division Directors.
The transmittal to CAG is to include a cover note containing the following l
information l
(a) Name and telephone number of the principal lead division contact.
(b) Scheduled date for completion of internal NRR review (i.e., completion of Phase 1,Section III, Office Letter No. 39).
(c) Nature and extent of prior CAG participation, if any.
5.
The lead division will either accommodate CAG comments or identify the differences between the CAG comments and the regulatory analysis.
Such differences should be brought to the attention of the Director, NRR.
l l
l 1
l 4
j i
ENCLOSURE 5 INTERIM GUIDELINES FOR THE PREPARATION OF A BACKFIT ANALYSIS I
The new backfit rule (10 CFR 50.109) became effective October 21, 1985.
1 The rule defines backfitting, in part, as any modification or addition to a facility or procedures or organization required to design, construct, or operate a facility.
The rule states that backfitting of a facility is required only when a systematic and documented analysis shows that back-fitting would substantially. increase the ovor all protection of the public health and safety )r the common defense and security nd that the direct and indirect costs of backfitting for a facility are justified in view of tha protection.
A statement of the determination, with regard to backfitting, made by the q
regulatory analysis must now be included in the preamble of each proposed or
)
final rule.
Recent Commission guidance (Memorandum, S. Chilk to V. Stello,
)
March 5, 1986) indicates that "the backfit analysis provided with a rulemaking package for public comment should contain sufficient detail in addressing each consideration listed in the backfit rule so that anyone wishing to comment on the analysis may do so without requesting additional detail or documents.
In addition, the analysis should clearly state, based on the facts considered, how the staff decided that the proposed rule woulo provide a substantial increase in public health and safety and that the costs are justified in view l
of this increased protection." A literal interpretation of the Commission vidance would indicate that the regulatory analysis, along with all referenced documents, be published in the federal Register.
Clearly, this is not practical.
We have interpreted the backfit rule as requiring a regulatory analysis for all proposed actions and a separate "backfit analysis" for all backfit actions which will be published for public review and comment (i.e., Rules, Orders, Regulatory Guides, Standard Review Plans, Standard Technical Specifications or revisions thereto).
In the interim period, all regulatory analysis packages
. developed by NkG should reference the backfit rule and specifically address each of the nine factors identified in the backfit rule.
The "backfit analysis" car nen be developed by extracting as much information from the regulatory analysis as is needed to satisfy the Commission's objective of providing a document that, while brief, contains sufficient detail to allow anyone review-ing the document to make meaningful comments without requiring additional detail or documents.
The "backfit analysis" should clearly reference the regulatory analysis and all other documents used as a source of data, analyti-cal methodology or of historical significance.
Reference documents should be placed in the Public Document Room at the time that the rule or proposed backfit action is submitted for publication, and the Federal Register Notice should note that all referenced documents are available for examination and/or opying, for a fee, at the Public Document Room at 1717 H Street.
)
The "backfit analysis" should address the nine following factors, as appro-priate, as well as any other considerations relevant and material to the proposed backfit:
(1) Statement of the specific objectives that the proposed backfit is designed to achieve; I
(2) General description of the activity that would be required by the licensee or applicant in order to complete the backfit; (3) Potential change in the risk to the public from the accidental
)
off-site release of radioactive material; (4) Potential impact on radiological exposure of facility employees; 1
(5) Installation and continuing costs associated with the backfit, including the cost of facility downtime or the cost of j
construction delay; I
i I (6) The potential safety impact of changes in plant or operational complexity, including the relationship to proposed and existing regulatory requirements; l
l (7) The estimated resource burden on the NRC associated with the proposed backfit and the availability of such resources; j
l (8)' The potential impact.of differences in facility type, design or age on the relevancy and practicality of the' proposed backfit; (9) Whether the proposed backfit is interim or final and, if interim, the justification for imposing the proposed backfit on an interim basis.
Backfit analyses provided with recent proposed rules have not been of sufficient level of detail to allow a thorough review without the use of reference documents.
l Accordingly, the Commission has solicited public comments on whether each of the backfit analyses provided for recent proposed rules adequately implement the backfit rule.
a Through the experiences to be gained in preparation of backfit analyses for future proposed rules and plant actionb, reviews with the CRGR and the Commission, and insights gained from public comment solicited regarding the adequacy of the backfit I
analysis for recent proposed rules, we will develop final guidelines.
The guidelines will explicitly define the level of detail and specific format to be i
followed henceforth.
At that time the interim guidance provided herein will be withdrawn and be replaced by a final guidance document.
{
1 l
1 0
1
!(pq'o, UNITED STATE s 4*j NUCLEAR REGULATORY COMMISSION
,E WASHINGTON. D C. 20555
.q
\\l,,,
007 0 3 m MEMORAhDUM FOR:
All NRR Enployees FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation
SUBJECT:
NRR OFFICE LETTER N0. 16, REVISION 2.--
REGULATORY ANALYSIS GUIDELINES l
This revised office letter supersedes the March 14, 1983, version (Revision 1) of NRR Office Letter No. 16.
The Executive Director for Operations (ED0) has issued NRC-wide regulatory.
analysis guidelines. These are contained in NUREG/BR-0058, " Regulatory i
Analysis Guidelines for the U.S. Nuclear Regulatory Commission."
(The current version of NUREG/BR-0058 is Revision 1, dated May 1984.) NRR will prepare regulatory analyses in accordance with NUREG/BR-0058 in support of covered submittals for review to the Committee to Review Generic Requirements (CRGR) or to the Deputy Executive Director for Regional Operations and i
Generic Requirements (DEDR0GR) or for decision to the EDO or the Commission.
j Covered submittals include two categories, as specified in NUREG/BR-0058.
i l
1.
Major proposed and final rules, meeting threshold criteria stated in NUREG/BR-0058, Section IIA.
There the guidelines apply fully.
l 2.
Other rulemaking actions and non-rulemaking generic requirements or guidance (referenced in Section IIB of NUREG/BR-0058).
For these i
much less detail is required.
The extent of detail ~should be commensurate with the safety importance of the issue, the estimated q
magnitude of the proposed action's impact, and the complexity end i
analytical tractability of the issue.
The required content of CRGR review packages and procedures for submitting l
them are described in NRR Office Letter No. 39.
Regulatory analyses-constitute a part of such packages (Item 5.1 in Section II of Office Letter 39, Revision 2).
Preparation of the regulatory analyses is the responsibility of the lead Division within NRR for CRGR submittal in accordance with NRR Office Letter l
No. 39.
The Division of Safety Technology (DST) will, as r. quested by the i
lead Division, provide assistance in interpreting the guidel nes and l
consultation on regulatory analysis techniques and serve as focal point for l
NRR requests for assistance of the Cost Analysis Group.
The lead Division will coordinate with DST before submitting a regulatory analysis to the Director, NRR, in accordance with Office Letter 39.
J.Ly g L y 4 -
y
001 0 y 984_
2 The purpose of this letter is to supplement the hRC-wide regulatory analysis l
guidelines (NUREG/BR-0058) in some_ areas with more specific NRR guidance.
This supplementary NRR guidance is enclosed.
In accordance with the Comission's Policy and Planning Guidance for 1984 (NUREG-0885, Issue 3),-sa'ety goals and associated numerical guidance will not now.be used in regulatory analyses; they will be used only as the Comission may direct in the future.
During the two-year safety-goal evaluation period, commencing in March 1983, i
DST will prepare or will have prepared, after a decision on an issue has been j
made, a separate evaluation for each NRR-originated major rule (i.e., action meeting the Section IIA criteria of NUREG/BR-0058) and for selected other-significant NRR-originated actions.
In these evaluations the Comission's I
proposed safety goals will be used as the basis for the evaluation of the irsue. These separate evaluations are intended to contribute to' NRC's efforts to evaluete the safety goals.
The evaluations will indicate how, if at al t, the Comission's proposed safety goals and numerical guidelines might have affected the decisions, and what the effect on the decisions might have been if goals or guidelines other than ' hose issued by the Comission were used. DST has developed, and will update as necessary, guidance for the l
nature, scope, and fermat of such evaluations for NRR coordination with the DEDR0GR and other NRC offices.
These evaluations will be coordinated with the originating Division and prepared for the signature of the Pirector, NRR, for transmittal to the DEDR0GR. Absent special difficulties, each evaluation should be ordinarily completed within two months of the decision
- involved, f
n l
Harold R. Denton,. Director Office of Nuclear Reactor Regulation
Enclosures:
l 1.
Supplementary Guidance for NRR 2.
Decision Making Using Value-Impact Analysis L
3.
Decision Factors Supplementing L
Value-Impact Ratio l
4.
Interim Procedures for Requests l
for Cost Analysis Group-Participation in Regulatory Impact Analysis Process cc:
V. Stello J. Sniezek R. Minogue R. DeYoung J. Davis
ENCLOSURE 1 SUPPLEMENTARY GUIDANCE FOR NRR NCTE:
Section numbers refer to correspondingly numbered sections of -
NUREG/BR-0058, Revision 1, " Regulatory Analysis Guidelines for the' O.S. Nuclear Regulatory Comission," May 1984.
.131.B.1.
Statement of problem This section will include a clear definition of the scope and bounds of, j
.the issue to which the proposed action is addressed.
This is essential ir.-
arriving at-a sound and applicable analysis since issues'are~often complex and interrelated with other issues.
i I
III.B.4.a.
Costs and Benefits of Alternatives This will always include comparison of the proposed action with the no-action alternative. Other alternatives will be covered only where significant other alternatives exist.
The no-action alternative should reflect implementation of all applicable requirements or guidance.to date, even if implementation 'is not completed at the time of analysis..
l Where the effect on different classes of plants varies substantially separate estimates.will be made and sepsrate conclusions drawn for each l
class. The analysis should show that the requirement or guidance is i
directed and focused as narrowly as practical.
Where additional information must be obtained from industry'to help bound an affected class of plants, obtaining that information should be justifie'd on the basis of its potential safety significance and cost.
The regulatory analysis should be' reopened for reconsideration should the new-information bring into questionL the t
soundness of the original conclusion.
, I l
The required analysis of-costs and benefits constitutes a "value-inpact" 4
assessment.
NUREG/CR-3568, "A Handbook for Value-Impact Assessment," may be used as a refsrence on methods for performing such assessments.
Professional judgment will be required in choosing amonc the options presented in the l
Handbook, in adopting methods to the issue at hand, and in applying methods not specificeily covered when special aspects' of the subject matter make that appropriate.
A general discussion of decision making using value-impact assessment is j
enclosed (Enclosure 2 to Office Letter No. 16, Revision 2).
1 The Costs and Benefits section should ordinarily be organized as outlined below. This outline is done primarily for safety issues, but for issues involving primarily other subjects (environmental protection, licensing and regulatory-process improvement, etc.), following a similar (or analogous) out ne is suggested. The outline follows:
1 (i) Value -- Risk Reduction Estimates The value is the safety importance of.an issue and is usually represented by the change in expected risk that resolution could effect.
Risk is ordinarily expressed here in terms of the product j
of the frequency of an. occurrence and the public dose (in man-rem) that would result in the event of the accident.
If more than one i
l accident scenario is important within the necessarily' rough risk i
estimates, the risks are summed.
Value is attributed to reductions
)
l in the average, mean, or expected risk. The dose is calculated for-the 50-mile-radius area around the plant.
I
]
the public man-rem-based estimate may not be the only 6ppropriate-measure of an' issue's safety importance in all cases. Alternative
{
measures of safety importance should be used when appropriate. 'For 1
1 i
l l,
1 I
exerrplc, when a possible core melt or other accident is involvec but releest cuisioe containr.ent would be minor or highly improbable,
or hichly uncertain in probability or magr.itde, cor,tribution tc q
l the cere-mcit probability should be estimated in additicn to (or
]
even in lieu of) the probebilistic public dose effect.
There is l
controversy as to whether such " contained accidents" are a sufficient basis for regulatory action.
Sincc core melts usually have a i
significant probability of failing ontainment, this is net ar issue for most core melts.
Howeve r.
other accider.ts car, be contained. The information on signif. cant cor.tained accidents should be displayed in the analysis, but the analyst must weigh the significance of this factor in deciding to what extent, if ary, it should influer.ce the conclusion.
It should be noted that safeguarding against an excessive core-melt frequency may justify impcsition of a safety requirement even without demonstration of offsite consequences.
Where significant occupational exposure is incurred in implementing resolution of a safety issue, such exposure is taken into account, but stated separately.
Such exposure is vicwed as a negative l
component in the net risk-reduction value.
l Where er accider,t may result in significant occupational exposure j
due to the accident directly or in post-accident plant cleanup, I
such exposure is taken into account and stated separately.
Fcr plant-wide cleanup after a contained severe core-damage accident causing substantial wide-spread plant contar.ination, 40,000 man-rem may be used as a rule-of-thumb occupational exposure figure when no better issue-specific figure is available.
I J
s 1
'f
)
l 1 1 l
Usually public dese is an adequate surrcgate for cff-site contamination.
However, the analyst should be alert to circumstances in which radiolccical effects not adequately reflectec ir, the public dose as surrogate may be present.
Examples include on-site danage, occupational exposure, liquid pathway effects, and supportive medical treatment, among others.
Long-term cor.tamination of the environment including water bodies and ether agricultural cr industrial facilities should be taken inte accour.t where that is a principal effect.
(ii) Impacts -- Cost Esticates j
l l
In January 1984, the EDO established a charter for the Cost Analysis 1
Group (CAG) formed in the Office of Resource Management.
NRR staff is encouraged to seek CAG advice and support, as needed and j
available, to help develop sound cost analyses for use in l
regulatory analyses.
Requests to CAG should be coordinated through l
the Division of Safety Technology.
The Safety Program Evaluation Branch is the contact point.
Procedures are described in the enclesed memorandum, " Interim Procedures for Requests for Cost Analysis Grcup Participation in Regulatory impact Analysis Process."
(Enclosure 4 to Office Letter No.16, Revision 2.)
Dependability, in terms of guarding against omission of important or even dominant cost elements, or against inclusion of costs that are not in fact entailed, is more important than high precision of the estimates, it is particularly important to oetermine correctly whether plant downtime (or prolongation of downtime incurred for unrelated reasons, such as refueling) will be involved, and if so, to
1 l
l l
?
l I l
estimate well the duration of the outage.
Dowr. tine costs, when irvolved, are of ten larger thcn all other costs. When no more specific basis exists for estimating osily cost of plant outace fcr f
the issue at hand, a figure of $500,000 per day may be used.
i l
It shculd be borne ir. mind that procedural changes and prcfessior.cl I
work are not cost-free, and may have costs that are substantial.
l I
Both irdustry and NRC costs (as well as costs to others when involved) should be estimated. Transfer costs (e.g., irsurance)
.need not be calculated in NRC regulatory anclyses since cur perspective is a net national impact.
In some cases plant-damage costs averted by the proposed action can substantially affect the cost-benefit evaluation.
Estimates for such averted costs are developed and used in separately stated calculations, so that the results both with and without adjustment for averted plant-damage costs are readily apparent. The averted I
costs may include those of averted equipment failures, limited-time plant outage, or limited plant-contamination cleanup.
In the extreme, they can also include averted permanent loss of use of the plant, and plant-wide cleanup, multiplied in each case by the reduction in frequency of such events that would be brought about by resolution of the generic safety issue.
In the absence of better issue-specific estimates, the estimated cost of plant-wide cleanup, before discounting to present worth, may be taken as $1.2 billion (1984 dollars; based on TMI estimates). While again a controversial issue, plant damage is to be considered an impact and not a value in NRR analyces. Should a favorable conclusion about a proposed generic requirement be contingent on taking plant damage into account, the analyst should state that conclusion together with a statement of the sensitivity of the conclusior to this factor.
q
[
m
, Wher, the discount rate is an impcrtsrt facter in the evaluatior cf the wcrthwhilencss cf a proposed action, the sensitivity.of the results to the discount rate should be tested, as suggested in-NUREG/BR-0058 (at III.B.4.a). Alternative real discount rates should include a 5 percert rate; in addition.tc the= 10 percent rate specified in NUREG/BR-0058.
High. discount rates reduce the impact (present worth) of fbture accidents but also re' duce the impact of the continuing costs associated with implementing new requirements or guidance.
(iii) Vclue/ Impact Ratio l
The total net safety value of the proposed action, typically in man-rem of public dose avoided,.is related to total net costs (NRC, i
industry, plus any other) in terms' of a ratio,-typically l
l dollars / man-rem. This ratio, along with safety importance, can be I
used as a supplementary basis for comparing alternatives, including evaluation against the no-action alternative, and ranking for implementation priority in relation to other issues.
(iv) Uncertainty Bounds Major sources of uncertainty in the benefits and costs should be identified and judgments as to their quantitative significance indicated as information warrants.
However, particular attention should be paid to those factors that have associated uncertainties of a substantially different nature or magnitude than thos'e encountered in typical risk or cost' analyses.
1 1
I I
I l l 1
(v) Special Ccrsideretions l
The value-impact anclyses should be es quantitative as the situation reasonably permits.
Items i to iv, above, emphasi:'e the quantitative aspects.
However, while the calculated risks and value-irrpact ratios are often valuable aids to judgment, other l
considerations nct adcquately reflected, or nct reflected at all, in the basic numerical formula employed are often helped in j
ccrroboratirg or adjustirg tnc results.
Decision making is helped l
l by explicit identification of such other considerations and l
explanation of how they bear on the conclusions.
A partial list l
and discussion of decision factors which might supplement the calculated ris6 and value-impact ratio is enclosed (Enclosure 3 I
to Office Letter No. 16, Revision 2).
III.B.S.
Decision Rationale This section should include a recapitulation of the train points of the rationale underlying the value-impact conclusion concerning the proposed actier, including critical information on importance to the safety (or other) values expected to be obtained, costs, value-impact relation, uncertainties, and any special considerations, j
i l
Rules and other existing requirements are not, in themselves, justification for I
a decision; they can be changed (by the Comission or authorized staff) on jLst the sort of regulatory analysis basis as is involved here.
For requirements established by law, the regulatory analysis should, when appropriate, weigh alternative means of implementing the law.
l
ENCLOSURE 2 I
DECISION-MAKING USING VALUE-IMPACT ASSESSMENT 1.
General Principles of Cost-to-Benefit Comparisons The decision process and decision considerations are similar when cost-benefit comparisons are close and when they are not.
The value-impact approach, as a' disciplined process, employed in conjunction with good engineering 1
analysis, helps guard against errors regardless of the relative magnitudes-of costs and benefit.
The calculated cost-benefit -- or value-impact -- ratio, whether nearly balanced or unbalanced, is a factor in the decision, but should not be regarded as the lone or even the generally determinative' factor. Other factors such as special risks or costs, subjective perceptions, time-related fectors, regulatory stability or implementation feasibility can be important.
These factors, when present to some particular extent, are more likely to be controlling when the~value-impact comparison is close than when values and impacts are unbalanced.
l All decisions based on risk or cost _ analyses must recognize that the uncertainty of the estimates of the factors can be -- and in most cases is -- large.
Uncertainty includes not only the variab'ility of data, but also incompleteness whether from ignorance or error.
When there'is a large imbalance between safety values and impacts, uncertainty is-less important.
For close balances, uncertainty is likely to be an important consideration.
The values considered should be strictly limited to_those results affecting the protection of public health, safety and property which the NRC is authorized to regulate. All other consequences of a proposed action should be included only as impacts.
Even if the net impact is favorable, this, by.
itself, is not a justification.for imposing an action.
The absolute safety-i I
[
f !
l l
value needs to be taken inte account.
Issues of small or rc safety value i
usually are not worthwhile considering, since there is er irreducible minimum cost associated with any issue and trivial safety values are therefore not cost effective.
In addition, while issues cf rmall value may
-l h6ve small overall costs (and therefore appear to be justified) these costs are not uniformly distributed.
For example, in most cases of small issues, I
these costs fall primarily on the licensing staff of the utility and the j
l NRC. These staffs are limited and the proliferation of issues with small l
safety values would divert them from significant issues which may have much greater overall costs, but not much greater use of a limiting resource such as licensing staffs.
Such reasons as these for dismissing issues of small value should be explicitly stated in the analysis.
2.
No Set Formula The significance of various potential decision factors and their interplay vary according to the issue.
All relevant and significant factors should be recognized and taken into account according to the facts and the issue.
Some general guidance can be developed, but a prescribed and ostensibly exhaustive checklist or weighting formula is likely to constrain decision making in counterproductive ways.
The analyses involved, especially when the decisions are close, are often complex and difficult.
They require exercise of judgment, for which there is no effective simpler substitute. But value-impact analyses, or at least one case for all analyses, should usually be done with a standard method and set of assumptions to allow meaningful comparisons between issues; work is needed to develop appropriate assumption sets.
Other cases with assumptions or method most appropriate to the issue under study can then be done.
3.
Development and Display of Facts No decision involving value-impact considerations can be dependable unless f
it rests on a sufficiently comprehensive factual analysis of values and f
impacts. This is, of course, especially true for close decisions, i
i i
~l
l 3
' Sufficiently comprehensive" cannot be defined beforehand but must be
- sdged for each issue. Methods and procedures for verifyirg the accuracy j
ar.d completeness of'the factual analysis must be used.
These include internal and public review and corrnent.
This assurance of accuracy and completeness is most important when the decision is close.
However, a lack f
cf data or ' uncertainty of the data used to determine values and irpacts j) cannot be used as a basis to defer a decision since both action and inaction are decisions. One course or another must be taken on the basis of the l
i information currently available no matter how good or poor it may be.
If
')
l 1
the information base is weak, the likelihood that inaction (as well as' action) may be wrong needs to be carefully considered.
1 l
.j The analysis should include:
(a)
Identification of value and impact elements that may be significant
.I for the issue.
(b) Realistic quantitative estimates, wherever estimates are reasonably possible.
l (c) Estimates of uncertainties, where practical, and comments on any unusual.
I aspects in the nature and structure of uncertainties.
(d) Qualitative description of unquantified value and impact elements.
The bases and assumptions for all calculations and qualitative statements should be made clear since decision outcomes may be sensitive to them.
1 Where important alternative bases and assumptions exist, sensitivity to them should be explored.
4.
Conclusions The basis for the decision should be stated.
Trade-off factors, what factors and analytical results were considered, and how the various considerations entering a decision were weighed should be stated explicitly.
l
l Cecisions to take oc action are usually more difficult. One, or a few
]
icentified pessible sequences that contribute to.a significant risk, may be.
sufficient to justify a decision to require action.
A no-action decision requires as much care as an action decision, in order to ensure that no significant f actors have been omitted..This is particularly true of close-i l
no-action decisions.
5.
Reverse Decision Analysis j
n
'l Reverse analysis may be used to determine whether a value-impact -imbalance j
could be upset by countervailing considerations., (A " reverse calculation" concerns-the question, "What facts would justify the action?", rather than
.]
"What action do the facts. justify?") The advantage of such an approach is that it diminishes pressures for quantifying factors that are problematical-to quantify. One need not determine what quantitative value to assign to an unquantified factor:
it suffices to determine whether the quantitative value should be above or below the value.that would establish value-impact -
i equipoise.
l l
1 1
,l l
l
[.
1 L
1 l
b l
1 I
ENCLOSURE 3 j
J DECISION FACTORS SUPPLEMENTING CALCULATED VALUE-IMPACT RATIO A calculated value-impact comparison generally reflects a necessarily-simplified evaluation, focusing on public dose and direct cost impact. The 1
possible importance of other factors and the limitations of an often I
incomplete and imprecise data base point to a need to corroborate or adjust s
the formula results by other considerations.
Some such effects --
'f occupational exposure, averted plant-damage costs, uncertainty bounds --
require careful consideration for all issues. Others thought to be i
significant should be identified and considered--quantitatively when practical, but at least qualitatively.
Some special considerations may be
{
c;uite specific to an issue.
This list is not complete and the analyst should assure that all significant ar.d relevant factors are considered.
1.
Special Risk and Cost Aspects l
1 i
Special risk and cost aspects potentially affecting net value or. ne't impact i
i but not routinely included in numerical formula:
l Special Risks and Value Aspects
]
I I
(a) A significant net change in occupational doses.
[ Sea discussion of I
occupational exposure in a preceding enclosure (Enclosure 1-to Office.
I 1
LetterNo.16, Revision 2)].
l l
(b) Loss or severe degradation of a layer in the defense-in-depth concept (e.g., all but one mode of_ core cooling, or all modes of' containment cooling).
(c) Circumstances imparting unusual significance _ to radiological or other accident consequences (such as ingestion-pathway effects or. great j
psychological stress) or mitigating measures (such as evacuation-4 sheltering, or supportive medical attention) that are not directly included in the public dose calculations.
Q
,.g d~
[
g';
. 1' (c) Enhancement or' impairment of the velue. of the proposed action by u
l' environr, ental or safeguards benefits or impairments ~ in addition to its J
safety value.
l
.l (e) Potential for substantial unusual off-site damage aspects not-adequately
~
reflected in the public radiation dose as surrogate.
l.
a Special Costs and Other Impacts (a) Any significant non-radiation related occupational r.isk.
]
(b) Averted cost 'of plant damage and outage not only from the postulated
)
accident but also its precursors.
l (c) Potential for substantial secondary cost impacts not reflected in the.
t cost estimates.
q
.1 (d) Significant impact on small entities.(protected by the Regulatory Flexibility Act).
! -J
\\
s 2.
Uncertainties a
i Factors related 'to uncertainties stensning from an incomplete.or imprecise l
r
'I data base for the value-impact formula:
a L
(a) Uncertainty bounds, imbalance in-uncertainty factors, certajoty of cost a
r J
l to fix versus uncertainty that safety is really improved and the true' l
-extent of"such improvement.
.).
~$
r
\\
'\\'
s'
.t s
- \\
[
j f,
j
[
r s,
h;
- i. !
l a
J J
i 1 '
~(b Situation's where uncertainty is extraordinarily large -(in accident -
t probability or consequences' or in cost, or. any or all of-these), for example, where unusually great plant-specific variability of. risks and costs defeats. dependability-of generic estimates,-or shere dataf about a risk are. lacking.
(c) The potential for a' proposed change'to affect more than one accident or:
transient sequence, thus affecting risk to a. greater or lesser ' degree' than assessed in the current description of' the issue; notably, the
' j potential for a new safety decrement, or increase in risk, due to suspec'c4 unidentified effects of a proposed changs, or added complexity, or for other reasons.
l 4
(d) The value inherent in reduction of uncertainty.
(e) The potential for human intervention,.using available equipment.
3.
Subjective Perceptions Perceptions and judgments that cannot (or cannot readily)'be quantified:
L (a) Public concern about a particular issue, or special. Commission or-I Congressional concern. The effect that 'a proposed reguYatory action (or inaction) may have on such subjective ~ perceptions'can be
[
quirksome: additional safety measures may be perceived as confirmatory of a concern and heighten it rather than queil it.
Usually, NRR decisions should be based.on objective factors, but a discussion of such subjective factors can be provided as supplemental information.
w 3 -
i>
.4 l:
-l
(:) Acute knowledgeable Drcfessional controversy concerning the importance of an issue or modes of dealing with it, or about the implementation costs of the proposed action.
4 Time-Related Factors I
~
i (a) How rapidly the action can be implemented.
l (b) The potential for a better resolution becoming aveilable later, or for j
anissuedisappearinginlightoffutureimprovedknowledge(balanced i
against (c), below, when applicable).
1 (c) Potential substantial deterioration of the value/ impact ratio while awaiting a better regulatory resolution-(e.g., a potential design fix' j
that is inexpensive to apply before construction, much more expensive after the plant is largely built, and extremely expensive and problematical to apply to an operatina plant).
1 5.
Regulatory Stability (a) Consistency with the principle of keeping the number of alterations to existing requirements to a reasonable minimum.
(b) The extent to which the proposed action afford better prospects for subsequent regulatory stability and predictability.
1 i
(c) The extent to which licensee latitude is restricted without good reason.
6.
Implementation Feasibility (a) The ease of implementat,on from technical, humar,-factors, and.
regulatory standpoints.
j
..( r ) How readily compliance is verifiable.
l.
(c) Vulnerability of the proposed change to lapses in implementation.
f i
-l l
i I
i l
February 8,1984 ENCLOSURE 4 l
i l
l l
FEv RANDUM FOR:
Darrell G. Eisenhut, Director
]
Division of Licensing Roger J. Mattson, Director l
Division of Systems Integration l
i Themis P. Speis, Director Division of Safety Technology j
1 Hugh L. Thompson, Director j
Division of Human Factors Safety i
?
l Richard H. Vollmer, Director Division of Engineering l
i l
FRDM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation j
1 SUEJECT:
INTERIM PROCEDURES FOR REQUESTS FOR COST ANALYSIS GROUP
-)
PARTICIPATION IN REGULATORY IMPACT ANALYSIS PROCESS i
l 1
A.
BACKGROUND By a memorandum dated January 4, 1984 (enclosed), the EDO established a charter for the Cost Analysis Group (CAG) formed in the Office of Resource Management and defined the CAG's role in the regulatory impact analysis process. The CAG will serve as the center of cost-estimate development expertise for the NRC.
The CAG is now functioning, has work under way to develop certain generic cost estimater, and is prepared to offer saidance and support to NRC offices, including NRR, in development and review of cost estimates on a selective basis.
B.
POLICY NRR staff is encouraged to seek CAG advice and support, as needed and available, to help develop accurate and dependable cost analyses for use in regulatory analyses and other pertinent contexts, such as prioritizations of generic issues or multi-plent actions.
f q
.s I.....................................
c.,,c,....................
ounuur>................................................................
un0.....................................
- \\
C.
LEVELS OF SUPPORT I
I The following levels of support may be available from the CAG, l
depending on resource availability:
l 1
(a) CAG provides general guidance on cost estimating methods and/or cost information on a generic basis.
(b)
CAG is available on call to confer and advise.
(c)
CAG has advisory responsibility in development of the ecst analysis.
(d) CAG provides the cost analysis.
CAG also has the following review and oversight function:
(e)
CAG reviews the completed cost analysis of others.
D.
NRR C0 ORDINATION 1
I have designated the Division of. Safety Technology as the focal point
{
e for coordination of NRR requests for CAG services.
The Director, DST, j
has identified the Safety Program Evaluation Branch as the contact
- point, i
DST's responsibilities include coordination,f NRR priorities with respect to CAG support needs and ' advice to NRR divisions concerning decisions to seek CAG participation and the level of CAG support to be sought.
E.
PROCEDURES The following interim procedures will apply, pending formalization of l
procedures (by NRR Office Letter or otherwise) in the light of i
experience with these interim procedures-1.
Cost analysis for potential generic requirements remains an integral part of the regulatory analysis process and remains the responsibility of the lead NRR division.
(See NRR Office Letters 16 and 39.)
2.
When cost analysis work is initiated as part-of a regulatory analysis the lead NRR division will notify CAG through DST (attention:
SPEB) of the subject of the cost analysis and its schedule.
This notification is required regardless of whether CAG support is sought.
cmer >
at:wwe >
D **TE )....................I
-a n ra n n.n meamemw
.m
=
- l l
l 3.
(a) If the leac NRR division desires CAG support other than review cf completed cost analyses, it will transmit a l
timely recuest for such sepport to CAG through DST (attenticn: SPEE}.
Th'is request should ordinarily be L
combinee with the notification under Item 2, above.
The j
request will identify the subj ect of the cost analysis, the~ nature and detail of cost analysis required, the j
nature of CAG support sought, schedule (including when the technical input will be available and when the cost l
analysis is needed), and principal lead division contact j
person.
A priority for the cost estimate will be l
included.
Normally the priori?.y will be the same as for I
the generic issue.
(b) After acceptarce of the work request by the CAG, the lean division should communicate with the CAG directly, I
keeping DST posted on progress and any major problems.
(c) It should be noted that the CAG charter (see Enclosure, l
- p. 4) permits the CAG to reduce or eliminate any advisory role that it may have during the development of a specific cost analysis if other pending issues should have priority call on CAG's resources.
DST will be 4
responsible for representing NRR views as to priorities l
in any such determinations and will bring to the i
attention of the Director, NRR, any conflicting views as l
to prior' Lies that cannot be resolved at the Division j
level.
)
4.
f.11 completed regulatory analyses will be transmitted i
to the CAG, with a request for review of the cost aspects, concurrently with transmittal of the analyses to NRR Division Directors for review prior to submittal to the Director, NRR.
(Office Letter No. 39, Rev. 2,Section III, Phase 1).
This will ordinarily be a copy of the same regulatory analysis package as the one that is sent to the NRR Division.,irectors.
The transmittal to CAG is to include a cover note containing the following information:
(a) Name and telephone number of the principal lead division contact.
(b) Scheduled date for completion of internal NRR review (i.e., completion of Phase 1,Section III, Office Letter No. 39, Rev. 2).
(c) Nature and extent of prior CAG participation, if any.
cmce >
CUQ84A8dt>
cart >
,=w maann a w,a,a wm a rum
a
. I 5.
The lead' division will either accommodate CAG comments or i
identify the differences between the CAG comments and the i
regt,1 story analysis.. Such differences should be brought to-the attention.of'the Director, NRR.
h 'prip:e! 5;W IT
}
h E. G. 00 3 f
Harolo R. Denton, Director i
Office'.of Nuclear Reactor Regulation
Enclosure:
Memorandu for Office Directors and l-(
- Regional Administrators from W. Dircks, i
Executive Director fer Operations, " Cost Analysis Group's Participation in the.
Regulatory Impset ~ Analysis Process,"
January 4, 1984-cc w/ enclosure:
E. Case-J. Funches F. Rowsome W. Minners S. Stern i
Distribution
': Central File SPEB R/F AD/T R/F DST C/F G. Sege W. Minners F. Rowsome T. Speis S. Stern J. Funches D. Eisenhut l'
R. Mattson I
H. Thompson 1
R. Vollmer
.l E. Case i
H. Denton MGarver /l849010 s
CTwigg
- ConcurrencebyphonetoG.Segeg SMcKelvin GSege:sfm;;f SPEB: DST-T
, AD/TfDST PPAS:NRR D
.D:DL D:DSI.
i
..y
,nners
.FRowsome SStern 1Funches
- DEisenhut *RMattson' I/4 84 I /zW84
../0/84 f
/ /84
- 7. // /.84_
//3//84
//2s784 q
ement.D...:.D...H.F.. S...........
.D...:.D..E...'........ Z..
...D... D..S..M.......
."..D../.S...:.N.R..FP........
D:NRR
"................. '.....RVollmer NJS eis.
ECaseV HDenton.
ecow >.HThompson
........s.............
.3.....,....
.././.2..,7/.8. 4..........,
.../../. 3...r./. 8. 4
...........'./.../. 8. 4......
.'. I/.. 5/84..... '...../. 3 /. 8. 4.J...
r m.
.~
- - - ~
f[t>* **Gk UNITED STATES
- e g
NUCLEAR REGULATORY COMMISSION y
5 f.
p WASHINGTON. D. C. 20555
\\,...>+ /
/
sas j
MEMORANDUM FOR:
All NRR Employees l
FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation i
l
SUBJECT:
COST ESTIMATING METHODS AND REPLACEMENT POWER COSTS-l SUPF EMENT NO. 1 TO NRR OFFICE LETTER NO. 16, l
REVISION 2 -- REGULATORY ANALYSIS GUIDELINES I
The purpose of this Supplement is to ncte the recent issuance of two l
documents that may be useful in the preparation of cost estimates for use in i
regulatory analyses.
[ Item III.B.4.a (ii) of Enclosure 1 to Office Letter l
No. 16, Revision 2.] These documents are as follows:
I 1.
NUREG/CR-3971, "A Handbook for Cost Estimating:
A Method for l
Developing Estimates of Cost for Generic Actions for Nuclear Power I
Plants."
l This Handbook may be used as a methodological aid to the performance of l
the required cost analyses and a guide to some widely useful data i
sources.
In applying the estimating approach described, professional judgment will be required in choosinc a degree of detail of cost analysis that is appropriate to the total costs involved, the complexity of the issue, and closeness of the value-impact trade-off.
In addition, professional judgment will be required in properly identifying and i
i characterizing the functions to be estimated, in critical application j
of the data sources, and in seeking cost information from sources not
{
identified in tne Handbook where nec?ssary.
j l
2.
NUREG/CR-4012, " Replacement Energy Costs for Nuclear Electricity
)
Generating Units in the United States."
This report presents plant specific replacement energy cost estimates applicable to short-term outages in the 1985-86 time frame.
These estimates should be relied upon when the identification of specific reactors is warranted, and when a high degree of precision in cost estimation is required.
The information may also be generally used as input to estimates prepared for other periods and for longer outages, subject to required professional judgment in adjusting the estimates according to the circumstances.
(The $500,000/ day approximate outage cost figure may continue to be used where more detailed estimates are not warranted.)
/
n Harold R. Denton, Director Office of Nuclear Reactor Regulation j s.4k.gryMA~
1..
b i