SCE's test year 2006 forecast of SONGS O&M costs includes an adjustment to provide additional staffing needed in advance of an anticipated large number of retirements caused by the aging of its SONGS 2&3 workforce. The adjustment is reflected (SCE's share) in Account 517 ($296,000), Account 520 ($319,000), Account 524 ($1,423,000), and Account 528 ($2,390,000). SCE projects that personnel retirements will increase for SONGS 2&3 in several job classifications that will require a lengthy period of training and qualification before individuals are allowed to perform their duties. In some cases, this training lasts as long as three years. According to SCE, these new hires will replace an increasing number of employees expected to retire from job classifications between 2004 and 2008.
SCE states that it developed the forecasted hiring of new employees in 2006 by comparing the age and length of service of the SONGS 2&3 existing workforce for each year from 2004 to 2008 to the historical average age and length of service of employees at retirement, for each job classification. This comparison produced an estimate of the number of employees expected to retire in each job classification each year through 2008. SCE subtracted the historical retirements from estimated expected retirements. This produced the forecast number of employees in each job.
DRA opposes SCE's aging workforce adjustment for SONGS. DRA states that SCE has been preparing for the 2004 retirement bubble since the 2001/2002 timeframe. DRA considers these preparation costs, especially for 2003, historically embedded. Further, in preparing for the 2005 and 2006 retirement bubbles, a portion of the costs should have been incurred in 2002 and 2003, and again are historically captured. Since it believes that the impacts of an aging workforce have already been embedded in SCE's base forecast, DRA recommends the Commission disallow the $5,900,000 incremental increase request for test year 2006.
In response to DRA' recommendation, SCE states DRA errs in claiming that costs associated with the retirement bubble are already embedded in base forecasts. SCE states that it removed 2003 recorded aging workforce costs of $1,500,000 (100% level) when deriving the incremental aging workforce replacement costs of $5,900,000 (100% level) for test year 2006.
It appears that SCE did remove prior aging workforce costs from the recorded data prior to estimating and including test year 2006 aging workforce costs. While there is no double counting, SCE's forecasted level of test year aging workforce costs is not totally supported.
For example, regarding Health Physics (HP) Technicians, SCE states "Retirements of HP Techs average one employee per year at age 52. Based on this historical data, an additional 10 employees will reach the age of 52 and retire by year-end 2008. HP Tech qualification and training takes three years. SONGS 2&3 will hire and begin training 10 HP Techs in October 2005. These additional employees will begin filling Retirement Bubble vacancies in September of 2008."21
SCE's explanation is incomplete. There is no mention of how many employees who were 52 or older did not retire historically. Just because, on average, one employee, at the age of 52, retires each year, it does not necessarily mean ten employees will retire if ten employees reach the age of 52. If 52 were the maximum retirement age, or if some other reason employees all retired at age 52, it would be reasonable to assume the 10 employees would retire, if they were projected to reach age 52 during the relevant timeframe. However, more than likely, employees at ages more and less than 52 retire, and on average one employee with an average age of 52 retires each year. It does not appear that SCE takes into consideration that, of the employees who reach the age of 52, some may not retire until they are older than 52. Under these circumstances, where 52 is a fairly young age for retirement, more information is needed. For example, if there were an average of four employees at age 52 or older who did not retire historically, it would indicate that only 20% of those in that age category retire. The addition of 10 employees to that age category would likely result in two additional rather than 10 additional retirements.
On the other hand, retirements for test technicians averaged 0.8 technicians per year at an average age of 61. Retirement at 61, average age or otherwise, is more likely than at 52. We would expect a higher percentage of retirements as additional employees reached that average retirement age. For the other categories, retirement ages between 56 and 60 are assumed in SCE's analyses.
In general, SCE has demonstrated that its nuclear related workforce is aging. For that reason, we believe that its proposed adjustment is reasonable. However, for the reasons discussed above, we will not include $480,000 requested for HP Techs. The adopted estimate is therefore $5,420,000 (100% level) or $4,068,000 (SCE share).
SCE recovers costs of SONGS 2&3 work scope occurring on a cyclical basis or requiring special focus in the Site Projects Functional Group. The majority of the activities are O&M projects developed in response to action requests or to external events or requirements. The number and scope of the projects and efforts vary from year to year. SCE's estimate for these projects is based on the 2003 recorded amount plus an adjustment to recognize the Used Fuel Transfer Project as being a new and unique project. On a 100% basis, SCE estimates the cost of the fuel transfer project to be $1,700,000 per year over the three-year GRC cycle.
DRA used a three-year average and did not include the Used Fuel Transfer Project as a separate adjustment. DRA argues instead that, given the varying nature of projects, cost estimates, operation alterations, and the potential for project scheduling adjustments, to identify a single project for exclusivity distorts a fair comparison. DRA considers the use of "pure" averaging a better approach to account for these varying fluctuations.
In rebuttal testimony, SCE accepted DRA's use of a three-year average instead of its proposed 2003 recorded amount. However, SCE continues to argue that the cost of the Used Fuel Transfer Project, a new and unique project, should be recognized, in addition to the averaged amount.
We will use the three-year average as a base for forecasting the site projects. Regarding the fuel transfer project, under the justification for this project, SCE states:
"After the removal of Unit 1 used fuel from the SONGS 2&3 fuel pools in 2003 and 2005, capacity of the SONGS 2&3 used fuel pools will again be reached by 2007, which will prevent the capability to perform full core offloads during the refueling. The capacity to perform full core offloads is necessary for certain maintenance and inspections of the plant. SCE must continue to remove used fuel from the used fuel pools in order to maintain full core off load capability with a prudent amount of schedule contingency, and thus enable continued reliable generation of electricity."22
The Used Fuel Transfer Project, while necessary, does not appear to encompass new and unique activities. Such activities were performed in conjunction with SONGS 1. However, the removal of SONGS 1 used fuel was funded through SONGS 1 shutdown O&M expenses. The related activities for SONGS 2&3 are new and have not previously been reflected in expenses for SONGS 2&3. The three-year average will cover the types of projects that were incurred, or are similar and related to projects that were incurred, during that timeframe. However, the SONGS 2&3 fuel transfer project is a sufficiently new, unique, annually recurring cost such that it is not covered by a three-year average of historic site expenses. SCE's request to recover the used fuel transfer project incrementally to the three-year average of historic site projects is reasonable and will be adopted.
SCE records annual Nuclear Energy Institute (NEI) dues in Account 517, as non-labor. The forecast for 2006 is $652,000 (100% level), based on the 2003 recorded amount. SCE states that NEI is the nuclear energy industry's Washington, D.C.-based policy organization and that participation in NEI's programs, committees, and activities helps to address and resolve issues important to the nuclear energy industry.
TURN recommends disallowing 50% of the forecast dues because the organization engages in significant public relations, public advocacy and image advertising work. At present, legislative lobbying is undertaken with dues money collected in the past, but TURN maintains that NEI is a highly political organization with goals of encouraging the future development of nuclear energy and ratepayers should not be forced to subsidize political advocacy with which they do not agree.
NEI explains its mission as follows:
"Mission. The Nuclear Energy Institute is the policy organization of the nuclear energy and technologies industry and participates in both the national and global policy-making process. NEI's objective is to ensure the formation of policies that promote the beneficial uses of nuclear energy and technologies in the United States and around the world."23
The principal focus of NEI appears to be the advocacy of nuclear power, both nationally and globally. There are many aspects of such furtherance of the nuclear industry that may not be appropriate for ratepayer funding. SCE's direct and rebuttal testimonies do not provide information on specific activities and related benefits that accrue to the company and/or ratepayers. TURN however states that some of NEI's activities are related to work by nuclear owners to reduce costs or improve performance through work with the government and analysis of improving technologies. For this reason TURN recommends disallowing half, rather than all, of the NEI dues. We agree with TURN's characterization of the NEI and, absent any better analyses, adopt its recommended disallowance $326,000 (100% level).
SCE argues (1) there is no difference between lobbying and public policy advocacy, (2) all such costs are included in lobbying, and (3) all such costs have been excluded from its estimates. That this is the case is not at all clear. By its mission statement, public policy advocacy is the primary focus of the NEI. It would follow that public policy advocacy should then reflect a large portion of NEI's costs. However, SCE's estimated share of lobbying expenses was zero in 1999 and $91,000 in 2000.24 We are not convinced that all public policy advocacy costs are reflected as lobbying and excluded from SCE's forecast. For ratepayer recovery of NEI dues, in the future, SCE should provide more detailed descriptions of the activities, the associated costs, and the resulting company and ratepayer benefits. With that information, in the future, we can make a more informed decision regarding disallowances.
7.7.1. Flexible Outage Schedule Mechanism
SCE requests that the Commission establish a forecast cost for its refueling and maintenance outage costs of $61,200,000 at the 100% level ($45,900,000, SCE share) per outage per unit. SCE forecasts two outages in TY 2006. However, since it is difficult to predict with certainty for SONGS 2 & 3 whether there will be zero, one, or two outages several years into the future, SCE requests that the Commission continue to utilize a flexible outage schedule mechanism for the post-test years, like that adopted and affirmed in its decision on SCE's last GRC, D.04-07-022. The post-test year flexible outage schedule mechanism establishes a standard per unit per outage cost in the GRC and then allows determination of whether zero, one, or two outages will occur in 2007 and/or 2008 through the flexible outage schedule mechanism to most accurately predict post-test year costs.
SCE's request for a flexible outage schedule mechanism for the post-test years is reasonable, unopposed and will be adopted. The outage cost reflects core and one-time activities, both of which are in dispute. The adopted cost per outage of $56,060,000 (100% level) is discussed below.
7.7.2. Refueling Outage - Core Costs
SCE developed its Test Year 2006 refueling outage forecast, at the 100% level, in two parts: (1) core outage activities for $50,940,000 and (2) one-time outage activities for $10,250,000. In making its forecast, SCE considered costs of six refueling outages (actual costs for SONGS 2&3 Fuel Cycle 11, SONGS 2&3 Fuel Cycle 12, and SONGS 2 Fuel Cycle 13 and estimated costs for SONGS 3 Fuel Cycle 13) to develop its refueling outage costs.
For estimating the refueling outage core costs, SCE averaged typical core costs for the recorded five refueling outages. Additionally, SCE proposed three adjustments by (1) adding a non labor escalation premium of $3,300,000, (2) adding a Bechtel supplemental labor contract change of $750,000, and (3) reflecting a $3,800,000 credit due to a change in capitalization criteria.
Aglet proposes removal of the non-labor escalation premium because SCE has not provided adequate justification. SCE criticizes the adjustment because Aglet singled out the one adjustment. SCE claims it provided the same level of detail for all three adjustments.
We will adopt Aglet's recommendation to exclude the non labor escalation premium, because SCE did not meet its burden to justify the request. Other than identifying the adjustment, SCE's testimony provides no information to support the need or calculation. To ignore the issue, as SCE suggests, only because there may be other unsupported costs as well is not a viable option. As stated in our preliminary discussion, SCE has the burden to affirmatively establish the reasonableness of all aspects of its application. Based on the record, SCE's request for a non-labor escalation premium is unsupported and will be disallowed
SCE's claim that both the Bechtel supplemental labor contract change and the credit to reflect a change in capitalization criteria have support similar to the non labor escalation premium needs to be addressed also. These two adjustments are similarly unsupported and, in principle, should be disallowed. To remove the Bechtel adjustment of $750,000 is straight forward and we reflect its removal in this decision. Regarding the $3,800,000 credit to reflect a capitalization criteria accounting change, we could merely reverse the accounting, that is increase expense by $3,800,000 and decrease capital by $3,800,000 assuming that there was an increase to capital because of this adjustment. Whether this would be correct is unknown, because the adjustment is not explained. Another option, since there is no support, would be to exclude the $3,800,000 entirely, that is, remove it from expense as proposed by SCE and assume there was an increase of $3,800,000 to capital and exclude that amount also. However, this appears harsh considering that the adjustment is an accounting change and presumably is not a request for money to fund additional activities. Since it is only an accounting change, we will instead leave the adjustment as proposed by SCE, that is reduce expense by $3,800,000 and assume capital has been increased by a like amount.
Exclusion of SCE's proposed non labor escalation premium and Bechtel supplemental labor contract change results in our adopted forecast of core costs of $46,890,000 (100% level).
7.7.4. Refueling Outage - One-Time Activities
For estimating the refueling outage one-time activities, SCE averaged the one-time costs for the five recorded outages and the forecasted one-time costs associated with the SONGS 3 Fuel Cycle 13 outage.
Aglet proposed to reduce SCE's average of one-time outage activities by $1,082,000. Aglet removed the costs to repair of the Unit 3 main generator rotor, amounting to $6,490,000, from the average arguing that it is an outlier. This forecasted item was the highest cost activity considered in the averaging of one-time activities.
SCE argues that other high cost activities will likely occur during 2006-2008. SCE states that it normalizes peaks and valleys that routinely occur between refueling outages by averaging both core and one-time activities, and Aglet's proposal to remove the peak cost skews the average.
We note that in the average to determine one-time activities, there are already a number of significant one-time costs in the recorded data for the five completed refueling outages. Typical one-time activities will be reflected in the forecast of outages performed during this GRC cycle. The main generator rotor repair is however almost twice as large as any of the other adjustments. It is also a forecast for 2005, not a recorded amount. Since it is forecasted, there is some uncertainty as to whether the activity will actually be done in the timeframe suggested by SCE or whether SCE's cost forecast is accurate. For these reasons, we will not reflect the main generator repair in the average to determine one-time activities for this GRC cycle. However, if the activity is actually performed, the recorded amount can be used in the averages to determine one-time activities in future GRCs. The adopted forecast for one-time activities is therefore $9,170,000 (100% level).
SDG&E owns a 20% interest in SONGS. Under the operating agreement between SCE and SDG&E, as the operator, SCE bills SDG&E for SDG&E's proportionate share of the actual total costs incurred by SCE in operating the plant, plus contractual overheads. The Commission has consistently used SCE general rate cases to determine the revenue requirement that SDG&E may charge its customers related to its share of SONGS billed by SCE (exclusive of fuel costs).25
In this proceeding, SDG&E has provided testimony that addresses the calculations and methodology for deriving its allocated share of SONGS costs. SDG&E points out that some of the costs that are allocable to SDG&E are found outside the SONGS portion of SCE's Results of Operations model. There are three principal groups of costs in this category. First, there are SONGS-related SCE shared services billed to SDG&E. Second, there are Results Sharing costs, which are SONGS-related incentive compensation (for SCE employees) billed to SDG&E. Third, there are contractual overheads, which are SCE-applied loaders for SONGS-related A&G, Pension & Benefits and Payroll Tax allocated to SDG&E.
Assuming SCE's requested overall revenue requirements, SDG&E calculates its 2006 SONGS revenue requirement (exclusive of refueling outage O&M, directly incurred SDG&E costs, and fuel) to be $94,000,000 (2006 dollars), its revenue requirement per SONGS refueling outage to be $15,400,000 (2006), and its share of SONGS capital expenditures to be $25,600,000 for 2006, $21,000,000 for 2007, and $16,700,000 for 2008. SDG&E acknowledges its SONGS revenue requirement will differ from these numbers to the extent that the Commission adopts related costs other that those requested by SCE.
No party has challenged SDG&E's methodology for calculating its SONGS related revenue requirement based on costs allocated by SCE. It is reasonable, and we will use it to calculate SDG&E's share of SONGS related costs in this proceeding.
Based on the costs adopted by this decision, SDG&E's share of SONGS related costs are as follows:
2006 SONGS Revenue Requirement (2006 dollars)
O&M $86,003,000
Depreciation 2,842,000
Taxes other than on Income 208,000
Income Taxes 1,511,000
Return 3,773,000
Revenue Requirement 94,337,000
Rate Base 45,843,000
Rate of Return 8.23%26
D.04-12-015 (the decision in the 2004 base margin phase of SDG&E's last cost of service application) authorized recovery by SDG&E of Nuclear Regulatory Commission (NRC) Design Basis Threat (DBT) costs of up to $14,469,000 of 2004 capital expenditures and $760,000 of O&M expenses, subject to a one-way balancing account (the SDG&E Security Costs Balancing Account) until there was a review of the costs in a future Commission proceeding. D.04-12-015 stated that this review could occur in an SCE proceeding.27 SDG&E has provided information on DBT costs in this proceeding and requests that the Commission include a finding that SDG&E has made the showing required by D.04-12-015, and conclude the amounts authorized by D.04-12-015 should no longer be subject to refund.
In D.04-12-015, it was stated that before final recovery of DBT O&M and capital expenditures would be authorized, "SDG&E must make a clear and complete showing that (1) the recorded costs are attributable solely to new security activities and investments that are required by the April 29, 2003 NRC orders; (2) the recorded costs are truly incremental, i.e., they are not included in this Phase 1 decision; (3) if any current (i.e., included in this proceeding) security activities or planned investments are supplanted by compliance with the new NRC requirements, so that costs for those activities and investments are reduced, such cost reduction are properly accounted for; (4) the costs must be incurred by SDG&E and the other plant owners, and not by taxpayers; and (5) the recorded costs are otherwise reasonable."28 SDG&E provided the following information and facts, which were not disputed by any of the parties to this proceeding:
· In this proceeding, SCE provided testimony on both O&M and capital expenditures related to the new security requirements. In Exhibit 13, SCE describes Work Order No. 1809-0509 as the capital project that satisfies the requirements of the April 29, 2003, NRC DBT order and subsequent guidance provided by the NRC. SDG&E's share was $15.0 million for 2004 and $1.1 million for 2005 (year-to-date through March). In Exhibit 8, SCE describes Adjustment #41 related to the increased O&M expense necessary to comply with the NRC security requirements. The related costs at 100% (excluding overheads) for 2004, 2005, and 2006 are $4.5 million, $9.8 million and $9.8 million in 2003. SDG&E's share of O&M expenditures (including overheads), are $2.5 million for 2004 and $1.1 million for 2005 (year-to-date through March).
· The incremental costs are not included in the SDG&E Cost of Service Phase I decision as SONGS O&M or capital expenditures. The incremental costs were separately identified and tracked in the SDG&E SONGS Security Costs Balancing Account.
· SCE in its Adjustment #41 "Increased Security to Comply with NRC Requirements," identified the changes to the forecast years 2004-2006 from 2003 to reflect the O&M expenditures of the SONGS Security Division necessary to comply with all existing NRC security orders. In its forecast, SCE identified costs that would decrease from the 2003 recorded level and included that decrease in its 2006 estimate.
· SCE's response to Data Request DR-ORA-180 supports the assertion that the costs have been incurred by SDG&E and the other plant owners, and not by taxpayers. In that Data Request, SCE was requested to answer the following questions: "a) advise whether recovery of a portion of security costs, via outside entities (e.g., Homeland Security), were available to SCE. Did SCE receive or expect to receive any recovery for security costs from outside sources? b) In response to a) above, if no recovery was available, advise why no recovery was unable given SONG[S]'s potential as a target." SCE responded as follows: "(a) There are no available sources of funding from the federal government or other outside entities for SCE to recover all or a portion of the increased security costs to comply with NRC security requirements resulting from the September 11, 2001 terrorist attack. SCE has not received any funding or notice of proposed funding from the federal government or other outside entities. (b) The Department of Homeland Security has distributed some first responder grants to state and local governments to supplement local funds, but to-date no such funding has been made available to public utilities....To date, no sources of funding have been available for the nuclear industry, including SONGS 2&3, to recover all or portion of the increased security costs to comply with NRC security requirements."
· SDG&E's share of DBT O&M and capital costs have exceeded the amounts initially estimated in A.02-12-028 and authorized in D.04-12-015. SDG&E's capital expenditures and O&M expenses exceeded the amounts authorized for 2004 and 2005 and thus no portion of these costs need to be refunded to SDG&E's ratepayers.
SDG&E's showing on DBT costs conforms to the specifications of D.04-12-015 and is reasonable. SDG&E's request that the amounts authorized by D.04-12-015 should no longer be subject to refund is unopposed and is granted.
21 SCE, Exhibit 6, p. 22.
22 SCE, Exhibit 91, p. B-2
23 TURN, Exhibit 356, p. 32.
24 SCE, Exhibit 91, p. D-2.
25 SDG&E also incurs some costs associated with SONGS directly, rather than having them billed to SDG&E by SCE. These directly-incurred costs are considerably less than the SCE-billed costs. SDG&E's directly-incurred costs are litigated in SDG&E's own general rate cases, because they do not overlap the subject matter in SCE general rate cases in the same way that costs billed by SCE to SDG&E for SONGS overlap the issues in SCE general rate cases. SDG&E's directly-incurred SONGS costs, therefore, are beyond the scope of this application. They were last forecast by the Commission in SDG&E's last "cost of service" case (A.02-12-028 decided in D.04-12-015) and will continue to be reflected in SDG&E's rates per decisions in that application until SDG&E's next GRC is decided.
26 Rate of Return authorized for SDG&E in D.05-12-043.
27 D.04-12-015, mimeo., pp. 28-30.
28 Id.