The current two-step licensing process (CP and OL) was instituted before the nuclear industry was fully mature. There were many first-time license applicants, designers, and constructors with unproven and incomplete design concepts; at that time, plant designs needed a final evaluation prior to operation. Now, reactor engineering may have matured to the point where final designs for most plants can be described at the CP stage. Therefore, the industry argues that a two-step licensing process no longer is necessary.

The utilities and the nuclear industry contend that the two-step procedure exacerbates construction scheduling problems because the plant design, regulatory design review, and hearings all occur during construction. They would like to change this to a one-step process that would place all three activities before construction begins. They believe this would improve the predictability and efficiency of the licensing process by making scheduling more certain. Also, an OL is perceived in many cases to be pro forma, but it still requires a full EIS and optional but usually requested hearings.* They suggest that a one-step procedure might encourage earlier identification and resolution of licensing issues while continuing to accommodate participation by interest groups and State and local governments.

There are two ways to achieve the equivalent of a one-step NRC licensing process: by combining the CP and OL, and by banking reactor designs and sites. The NRC and DOE legislative packages include proposals for both of these measures. It should be noted that neither the DOE nor the NRC bills is tied to the use of standardized designs, either in the provisions for combined CP/OLS or for design banking. However, in the following discussion of these proposals, it is assumed that plants will be much less customized, relying on only a few standardized and complete designs. An earlier OTA study, Nuclear Powerplant Standardization, found that standardization of designs and construction, operation,

*The ASLB refusal to authorize an OL for Commonwealth Edison Co.'s Bryon plants may indicate a change in approach at NRC. Even if the decision is overturned by the ALAB, it is unlikely that utilities will ever again consider the OL to be a formality.

and licensing practices could alleviate many the nuclear industry's difficulties in verifying th safety of individual plants. In addition, standard ization could facilitate the transfer of safet lessons from one reactor to another and coul help reduce the rate of cost and leadtime escal tions (10). As discussed in detail in chapter 4, is likely that any new plants would try to max mize these advantages by standardizing design to the greatest extent possible.

The NRC legislative proposal specifies that get a COL, an application must contain "sufficie information to support the issuance of both th construction permit and the operating license The NRC staff analysis of the proposal interpre this to mean that the application must includ an essentially complete design. Under the NR bill, an optional hybrid hearing could be r quested before the COL is issued and again b fore the plant goes into operation for matters th were not considered in the first hearing. The fin review before a plant goes on line would er with NRC issuing an "operation authorization that would be the regulatory equivalent of a cense for purposes of inspection and judic review (23).

This proposal would eliminate the duplicati of detailed environmental and safety reviews th are currently needed for an OL; otherwise, it the equivalent of the present two-step proce with a new name. It is likely that hearings wou still be held before construction and again befo operation. Moreover, if the plant were a uniq rather than standardized design, this procedu could take even more time than the current tw step process.

In the DOE legislative proposal, NRC wou provide an expedited procedure for COL hold to start operation by allowing the licensee to c tify safety when the plant is virtually comple NRC would publish notice of the certificati with a 30-day comment period, and the st would have 45 days from the date of that noti to review the plant for safety, consider the pu lic comments, and recommend action to NR There would then be an additional 30-day pe

d in which NRC could take action to prohibit r limit operation if the certification was found › be incorrect. If NRC did not prohibit operaon during that period, the plant could go on line. he only opportunity for public hearings would e at the issuance of the initial COL.

The COL proposal is controversial because of ncertainty about the level of design detail that ould be required to obtain a combined license, nce this is left up to NRC to specify through lemaking. In addition, neither bill directs NRC › resolve all outstanding safety issues prior to censing. Nuclear critics argue that the number design changes still being made between a CP id an OL and the critical safety issues still beg uncovered at the OL stage indicate that the dustry and NRC are not yet ready for one-step censing. Such a procedure could reduce atten›n to unresolved safety issues raised at the CP age and could be used to restrict NRC's ability order backfits. Regulators and critics especially >ject to the DOE bill because it allows the linsees themselves to certify safety, with a limited ne for the NRC staff to verify that certification, d no real opportunities for citizen participation. Some utilities are not convinced that a one-step ocess would be any more predictable than the rrent two-step process in terms of requirements r a license and backfits. Furthermore, it is possie that the proposed COL procedure, when coued with hybrid hearings, would take longer than e current CP and OL process. Using procedural anges to improve the management of the hear¿s and implementing site- and design-banking, ich together would serve as a surrogate for e-step licensing, probably would do more to crease the efficiency and predictability of licens3 than a switch to a COL.

Current NRC regulations allow for design re'w prior to the filing of the CP application, but è results of the review are not binding upon CP determination. Alternatively, reactor venrs can submit generic designs for approval ough rulemaking. Many industry analysts are that reactor engineering has matured suffintly to allow preapproval of standardized plant signs, or of major system or subsystem designs, d.both the NRC and the DOE bills include pro

visions for "design-banking." Debate continues, however, on the degree of specificity that should be required for preapproval of designs and whether such approval would act as a disincentive to the continued improvement of designs.

Under the NRC legislative proposal, a binding design approval valid for 10 years could be granted without reference to a particular site and could be renewed for 5 to 10 years unless NRC found that significant new safety information relevant to the design had become available. The public would have an opportunity to request hybrid hearings on the design before NRC granted approval. Issues related to the design could not be raised in a subsequent CP, OL, or COL hearings unless the combination of a design with a particular site resulted in new issues that had not been addressed in the design approval or there was convincing evidence that reconsideration of design issues was necessary.

The DOE bill also would allow utilities to choose a preapproved plant or major subsystem design as an alternative to selecting a unique plant design. Design approvals would be subject to hybrid hearings. Once approved, a design would be valid for 10 years and then could be renewed for 10 years but would be subject to the same backfitting requirements as normal plants under the DOE bill. Preapproved designs would be incorporated into a CP or COL application, and the review of design issues in the hearings would be strictly limited. The DOE bill would require NRC to define the level of detail necessary for design approvals through the normal rulemaking process.

Preapproval of standard designs might make a substantial contribution to a more efficient and predictable licensing process by removing most design questions from the licensing of a particular plant, but it is likely to be as controversial as the proposal for a COL. Issues include the degree of specificity required for design approval, the conditions and procedures under which the utility or its contractors could deviate from a preapproved design once construction has begun, and the ability of NRC to order backfits on approved designs.

Nuclear critics are concerned that discussion of new or previously unresolved safety issues would be foreclosed in the CP or COL hearings on preapproved standardized designs, especially in light of the provisions that prohibit the raising of generic safety issues in the licensing of particular plants and of the provisions that shift the burden of proof to the public to show that a preapproved design does not meet current safety standards. Proponents of this change argue that preapproval of designs could improve the effectiveness of public participation in that it would allow earlier and more detailed discussion of design issues in hearings without the time constraints imposed by the licensing of a particular plant.

The critics also object to the length of time for which a design approval would be valid, given the frequency with which design changes have been instituted in the past, and to the subsidy granted by deferral of the application fee until the design is used. Furthermore, there is concern that once a design has been approved, the vested interest in it would remove any incentives to improve it. However, as discussed in chapter 7, the industry argues that its need to remain competitive with foreign countries should be incentive enough.

In the present system, NRC approval of site suitability is not initiated until the CP application is docketed, which places site review on the "critical path" for reactor licensing. The existing NRC regulations permit review of site suitability prior to filing of the CP application, but the outcome of this review is not binding in the final CP decision unless a special ASLB decision is obtained. Both the NRC and the DOE legislative packages recommend a procedure for binding early site approval that would be independent of a CP application.

In the NRC legislative proposal, a site approval that does not reference a particular nuclear plant could be granted for up to 10 years, with renewal possible for 5 to 10 years. Federal, State, regional, and local agencies, as well as utilities could ap

ply for site approvals, thus encouraging broade planning. In the NRC bill, a site approval woul not preclude the use of the site for an alternativ or modified type of energy facility or for any othe purpose. However, other uses not considered i the original approval may invalidate the site pe mit, as determined by NRC. The public woul have an opportunity to request hybrid hearing on the site approval, but issues related to the sit would be excluded from further licensing pr ceedings unless matching the site with a particul plant design raised issues that were not consi ered at the time of the site approval.

The DOE proposal is similar to NRC's, exce the site-approval procedure in the DOE bill wou not allow alternative uses and would allow C applicants to perform limited construction acti ities before issuance of a permit. A site approv would be valid for 10 years, with 10-year rene als. Under the DOE legislative proposal, the pu lic could request hybrid hearings prior to NF approval of a site.

As with design approvals, OTA concludes th site-banking could improve the efficiency ar predictability of the licensing process by taki siting out of the critical path entirely. As lo as the site-approval process allows adequate o portunity for public participation and ensur consideration of issues related to the combin tion of a particular site and design prior to iss ance of a CP, binding early site approval shou not be a controversial change. In fact, severi site approval from the CP could facilitate earli and more substantive public participation. Ti principal objections nuclear critics have to the bills are the length of time for which approvi are valid (including renewals, 20 years in the NI bill and an indefinite period in the DOE) and t subsidy introduced by deferring the applicati fee until the site actually is used or the approv expires. Furthermore, the selection of particul sites-whether they are matched with a plant not-will remain controversial, as discussed chapter 8.

In licensing a nuclear powerplant, NRC is required to make several determinations that are not related directly to safety. These include certification of the need for power from the plant (required under NEPA) and of compliance with antitrust laws.

There is general agreement that NRC is poorly equipped to judge need for power on a local or regional basis, and therefore that it is a waste of staff resources to make such a determination. Moreover, at least 45 States already require other agencies to determine the need for power either in the certification or licensing of powerplants, in rate cases, in the approval of financing, or in an independent planning process (20). Furthermore, evaluations of the need for power and the choice of alternative types of generating technologies can take up hearing time and staff time that could be better spent in the analysis of safety and design issues.

Both the NRC and the DOE legislative packages provide for binding NRC acceptance of a need for power determination made by a Federal, State or other agency authorized to do so. The NRC bill also provides for acceptance of other agencies' rulings on alternative sources of generating capacity. Only where no other agency is required to make such a determination would NRC perform a de novo review of the need for power. In both bills, these provisions are embedded within the section on a one-step licensing process, but they could be separated out. Because each agency is required under NEPA to make these determinations, legislative action would be required to delegate that authority to the States or other Federal agencies. It is possible that this provision would result in expanded opportunities for >ublic participation in the discussions of the need or power and choice of technology. However, neither bill sets minimum standards for public participation in delegating this authority to the States, nor do the bills mandate consideration of he full range of alternatives, as required in NEPA. Under current practice the Department of Jusice performs a comprehensive review of license pplications for compliance with antitrust laws.

Although NRC weighs the opinion of the Justice Department heavily in its determination, the Commission remains responsible for the final antitrust decision. As in need for power, it may not be appropriate for NRC to devote staff resources to antitrust law. One option is for NRC to adopt the Justice Department's decision on antitrust unless an affected party objects within a specified time after notice of the decision. If the objection is found to have merit, then NRC could remand to Justice for further consideration or do an independent review. Legislative action would be required to delegate this authority to the Justice Department.

The U.S. General Accounting Office (GAO) also has recommended that the NRC provide better coordination with State and local governments in NEPA reviews. At least 23 States have statutes requiring preconstruction environmental reviews similar to those required under NEPA, but NRC's NEPA regulations make no provision for coordination with the States or for eliminating duplication of efforts. GAO recommends NRC work jointly with all the States to identify common legal and procedural requirements as a first step in coordinating environmental reviews (2).

Finally, it has been suggested that introducing a little flexibility into the concept of exclusive Federal jurisdiction over reactor regulation would go a long way toward alleviating State and local concerns and improving public acceptance. For example, Oregon has a memorandum of understanding with NRC that sets forth "mutually agreeable principles of cooperation between the State and NRC in areas subject to the jurisdiction of the State or the NRC or both." This memorandum is intended to minimize duplication of effort, avoid delays in decisionmaking, and ensure the exchange of information that is needed to make the most effective use of the resources of the State and NRC. To accomplish these ends, the memorandum provides for potential future subagreements in areas of mutual concern, including siting of nuclear facilities, water quality, nuclear plant operation, radiological and environmental monitoring, decommissioning of nuclear plants, emergency preparedness, personnel train

ing and exchange, radioactive material transportation, and other areas. Subagreements adopted to date include a protective agreement for the

exchange of information, and an agreement on resident inspectors at the Trojan plant, the only nuclear powerplant in Oregon (11).

One concept that has attracted much attention in discussions of backfitting and other changes in the NRC technical regulations is the use of safety goals* to establish safety requirements and gage the need for changes in those requirements.

NRC currently is developing a safety goal policy, and the DOE legislative proposal emphasizes the importance of this effort by endorsing the Commission's efforts. The DOE bill would require NRC to report to Congress within 1 year on its progress in developing and implementing a safety goal policy. The NRC proposal is described below.

NRC Safety Goal Proposal

NRC has issued a policy statement on safety goals for nuclear powerplants that is being used on an experimental basis (25). It currently plays no part in licensing decisions, and license applicants do not have to demonstrate compliance with it. If the proposed policy receives sufficiently favorable response, NRC will consider amending its regulations to include safety goals in licensing decisions.

In developing a safety goal policy, NRC considered qualitative goals that would interpret the Atomic Energy Act's standard of adequate protection of public health and safety, as well as quantitative goals that could provide a more exact standard against which risks could be measured. Qualitative goals were adopted to lend NRC safety decisions "a greater coherence and predictability than they presently appear to have,” supported by numerical guidelines as goals or benchmarks (25). The NRC report notes that this

*NRC defines a safety goal as "an explicit policy statement on safety philosophy and the role of safety-cost tradeoffs in the NRC safety decisions" (25).

approach allows it to capture the benefits of qualitative goals and quantitative guidelines in measuring performance while avoiding the vagueness of qualitative goals without numerical guidance. It does not lock NRC into quantitative goals that may not be able to yield technically supportable results given the uncertainties inherent in quantitative risk assessment.

The qualitative safety goals established in the NRC policy statement are:

Individual members of the public should be provided a level of protection from the consequences of nuclear powerplant accidents such that no individual bears a significant additional risk to life and health

Societal risks to life and health from nuclear powerplant accidents should be as low as reasonably achievable and should be comparable to or less than the risks of generating electricity by viable competing technologies (25).

The intent of the first safety goal is to require a level of safety such that individuals living or working near nuclear powerplants should be able to go about their daily lives without special concern by virtue of their proximity to such plants. The second safety goal limits the societal risks posed by reactor accidents and includes an implicit benefit-cost test for safety improvements to reduce such risks.

These goals focus on nuclear powerplant accidents that may release radioactive materials tc the environment. They do not address risks from routine emissions, from other parts of the nuclear fuel cycle, from sabotage, or from diversion o nuclear material. The policy statement notes that the risks from routine emissions are addressed ir current NRC practice through environmental im pact assessments that include an evaluation of the radiological impacts of routine operation of the plant on the population around the plant site. Fo