Low-Income Program - GSWC requested recovery of the accumulated balance of the cost of its low-income program for Region II through a rate surcharge. The parties agreed that when the balance reaches 2% of adopted Region II revenues, the company may file an advice letter providing for recovery of the balance through a rate surcharge. (¶ 11.01.)

4. The Cost Allocation Studies Submitted by Both GSWC and DRA Contain Significant Flaws, So a Study Developed by the Commission Will Have To Be Used

A major topic in GSWC's testimony on general office issues was GSWC's cost allocation study. Such a study is used to assign costs that cannot be directly assigned between the three Commission-regulated GSWC water districts, on the one hand, and the company's unregulated and out-of-state affiliates, on the other. In view of the very large revenue increase that GSWC is seeking here for its general office operations, the cost allocation study became a major issue in this proceeding.

4.1. Overview of GSWC's Position on Cost Allocation

In his testimony in support of the allocation study, GSWC vice president Keith Switzer begins by describing the cost allocation approach the company has used in the past, and then argues that a new approach should be used in this case:

Previously, GSWC relied on the four factor methodology to allocate its general office costs. With that method, the general office costs were totaled and then the total amount was allocated based on a single set of [four] allocation factors. Those allocation factors were derived with the four factor methodology that the Commission has authorized for many years.[⁴]

In contrast to previous GRC applications, GSWC is proposing in this application to develop, whenever possible, a cost determinant factor for certain functional activities within the general office and to use that factor as the basis for assigning costs associated with that functional activity rather than allocating the total general office costs with a single cost allocation factor. As explained below, GSWC is proposing four different allocation/assignment factors for the general office costs. These four allocation factors will be applied to different cost items within the general office. Once the individual cost items have been allocated with their specific allocation factor, then the allocated costs will be summed by entity to derive an allocation factor for GSWC's regulated operations, CCWC and ASUS. (Ex. 6, Switzer Testimony, pp. 5-6.)

GSWC emphasizes that it is not proposing to abandon the four-factor approach entirely. On the contrary, Switzer states, the company intends to use the four-factor approach "for these broad-based functional activities where there is not a natural single cost determinate." (Id. at 6.) He continues, however, that many of the types of costs that are incurred in general office operations - costs that GSWC has grouped into 42 "cost centers" - lend themselves to a single natural cost determinant.

GSWC's opening brief gives the following summary of how the approach described above was used in the GSWC cost allocation study sponsored by Switzer in his testimony:

In its study, GSWC identified three ways to attribute cost-causation directly per the Commission's directive. First, for the Customer Service Center, GSWC's call center, GSWC allocated costs based on the number of phone calls received. GSWC Switzer, Ex. 6 at p. 10. Similarly, GSWC allocated costs for billing and cash processing based on the number of bills generated. Id. at p. 11. Finally, GSWC allocated Human Resources costs based on the number of employees. Id. at p. 12. These three cost allocation factors are based on cost causation for each cost center to which they are applied. The resources of the call center are spent in direct proportion to the calls received; the resources of billing and cash processing are spent in direct proportion to the bills generated; and, the resources of human resources are spent in direct proportion to the number of employees served. (Id. at p. 9-12.)

Where such natural determinants are not available, GSWC used the four factor allocation method. The costs to be allocated using the four factor include, but are not limited to, the costs for corporate executives, regulatory affairs, accounting and finance function, risk management, water quality, environmental, legal affairs, and corporate communications. These functions are very broad based in scope and are therefore appropriately assigned using the four factor methodology. (GSWC Opening Brief, pp. 12-13.)

Using the approach advocated by Switzer, GSWC concludes that 95.45% of its general office costs should be allocated to the three GSWC water districts regulated by this Commission, 2.34 % to its unregulated affiliate, ASUS, and 3.21% to 54% (CCWC), a water company in Arizona that is regulated by the Arizona Corporation Commission. Under GSWC's allocation proposal, customers in GSWC's Region II would bear 34.67% of overall general office costs, while customers in Region III would bear 33.46% of these costs. (Exhibit 6, Switzer Schedule D.)⁵

4.2. Overview of DRA's Position on Cost Allocation Issues

In its testimony and briefs, DRA has severely criticized GSWC's cost allocation study. DRA's opening brief sums up DRA's criticisms as follows:

GSWC's cost allocations basically involve determining [] two components: the allocation base and the allocation factors to apply to bases. According to GSWC, GSWC assigned the total actual [general office, or GO] costs to 42 cost centers; aggregated these individual cost centers to form various combinations of allocation bases; and applied a single allocation factor to many of these bases. GSWC claims that it applied the `traditional four-factor' to the bulk of the GO costs.

DRA finds that the allocation bases formed by GSWC are unreasonably [] limited and disregard [] including other related cost centers. As for the allocation factors, DRA recommends rejecting both GSWC's single factors and its purported `traditional four-factor,' which actually [are] not the four factors GSWC utilized in past GRCs and which [were] found acceptable.

For example, when allocating the costs associated with Customer Service Center (CSC), GSWC chose to combine only 4 out of the 42 Cost Centers . . . However, GSWC fails to justify restricting the CSC allocation base to only these 4 Cost Centers. (DRA Opening Brief, pp. 10-11; footnotes omitted.)

DRA is particularly critical of GSWC's decision to use the number of calls, number of bills generated and number of employees to allocate some, but not all, general office costs. On this issue, DRA states:

The Commission should reject GSWC's use of single cost allocation factor[s]. GSWC has filed to justify that any of its proposed single factors - number of phone calls, number of bills generated, and number of employees - is more reasonable to use than the four-factor cost allocation methodology. The Commission should adopt DRA's cost allocations . . . which are based on the four-factor methodology. (Id. at 13.)

Based on its own analysis, DRA recommends that 18.21% of the GSWC's general office expenses should be allocated to ASUS, 3.21% to CCWC, and 9.70% to Bear Valley Electric Company (BVEC). (DRA Opening Brief, p. 9.)

In addition to advocating that substantially more of GSWC's general office expenses should be allocated to GSWC's affiliates than does the company, DRA argues that GSWC's general office revenue requirement should be reduced by $2.96 million for each of the three years covered by this GRC (2007-2009). According to DRA, such a reduction is appropriate because "this is the amount of indirect costs that should have been but were not assigned to ASUS for the nonregulated contracts in effect from 1999-2003, because no cost allocation had occurred during those periods or subsequently." (DRA Reply Brief, p. 14.) Although only adverted to in DRA's testimony (Ex. 23, pp. 4-13 and 4-15), the development of this calculation is set forth in a DRA workpaper that was the subject of extensive cross-examination. (Ex. 45, p. 7; Tr. pp. 780-787.) DRA takes the position that GSWC's general office revenues should be reduced by $2.96 million for each of the years involved in this rate case (for a total of $8,872,314) because, in DRA's view, GSWC has failed to comply with D.98-06-068, the Commission decision that authorized GSWC's predecessor, Southern California Water Company (SCWC), to establish a holding company.

4.3. Discussion

For the reasons set forth below, we conclude that neither the cost allocation approach advocated by GSWC nor that advocated by DRA is satisfactory. However, based on the testimony of GSWC witness Switzer and the schedules attached to this testimony, we believe that the data in GSWC's study (along with approximations from other recent GSWC cases) can be used to produce a cost allocation approach that is acceptable for this general rate case. We do not believe the approach advocated by DRA is susceptible to such repairs.

4.3.1. DRA's Proposed Reduction of the General Office Revenue Requirement by $2.96 Million Annually to Account for "Missed Allocations" is Unjustified

We begin our discussion by observing that there is no merit in DRA's contention that GSWC's general office revenue requirement should be reduced by $2,957,438 in each of the three years covered by this GRC. As noted above, DRA seeks this reduction to make up for indirect costs that, in DRA's view, should have been but were not charged to GSWC's non-regulated affiliates during the years prior to 2003. We reject DRA's position because we think it is clear from D.04-03-039 that the Commission has already made a revenue adjustment to account for the incorrect implementation of D.98-06-068, the decision that authorized GSWC's predecessor to form a holding company.

In D.04-03-039, the Office of Ratepayer Advocates (ORA) - the predecessor of DRA - challenged SCWC's proposal to account for the costs of transactions with its affiliate, ASUS, through the Other Operating Revenue (OOR) sharing mechanism established by D.00-07-018. ORA contended that such sharing was inappropriate because the OOR was not intended to apply to a water utility's transactions with affiliates, and that the proper guidance on affiliate transactions was to be found in D.98-06-068.

The Commission agreed with ORA that the requirements of D.98-06-068 governed, and that the OOR mechanism was not intended to apply to affiliate transactions:

[D.98-06-068] adopted specific guidelines for the various transactions conducted between SCWC, the holding company and affiliates. The overriding theme is that ratepayers should not subsidize affiliate operations. The justification for allowing SCWC to establish the holding company structure and the implementation principles are directly applicable to the issue of allocating costs to unregulated operations in this GRC. SCWC should follow the policies and guidelines adopted in D.98-06-068.

Rather than following the principles and guidelines of the holding company settlement, SCWC has instead used the principles established in D.00-07-018, which established [the OOR] mechanism applicable to water utilities that intend to offer non-tariffed services. SCWC has misinterpreted the intent of that decision. The revenue sharing mechanism is intended to apply to a water utility (1) providing non-tariffed services, (2) sharing the gross revenues with ratepayers, and (3) absorbing all incremental costs. It does not apply to non-regulated affiliates of the water utility. While we regulate water utilities, we have no direct authority over non-regulated affiliates. Rather than imposing a sharing mechanism on the revenues of non-regulated affiliates, we instead attempt to ensure that utility and affiliate costs are properly separated and common costs are fairly allocated. In that way, sharing of non-regulated revenues with ratepayers is unnecessary. (D.04-03-039, mimeo. at 28-29; emphasis supplied.)

To carry out its determinations, D.04-03-039 ordered SCWC to conduct a cost allocation study "to be included in [SCWC's] next Region III GRC." (Id. at 29.) However, the Commission declined to "suspend" the affiliate revenue sharing procedure advocated by SCWC until the required cost study was completed. Instead, the Commission adopted ORA's suggestion to increase the company's general office revenues by $101,300, an adjustment that D.04-03-039 described as "a proxy for potential adjustments that might result from a cost study." (Id.) In other words, pending SCWC's completion of the cost study, the Commission decided to accept the $101,300 adjustment in lieu of the full range of adjustments that an immediate cost study might have required.

In view of this discussion of the general office revenue adjustment in D.04-03-039, there is no basis for the $2.96 million annual reduction to general office revenues that DRA proposes to make here. Although D.04-03-039 chided SCWC, GSWC's predecessor, for its failure to adhere to the requirements of D.98-06-068 in accounting for affiliate transactions, it is clear that the Commission chose to deal with the problem prospectively by (1) increasing the amount of OOR revenue sharing that SCWC had proposed, and (2) awaiting a proper cost study in the company's next GRC. Nothing in D.04-03-039 suggests that the Commission would be inclined to deal with improper cost allocations in the past by making retroactive adjustments to the company's general office revenue requirement.

In its opening comments, DRA argues that the proposed decision (PD) errs in rejecting DRA's claim that $2,957,438 should be deducted from GSWC's general office revenue requirement for each of the three years covered by this rate cycle to make up for "missed allocations" of general office expenses that were not made to ASUS for the years 2000-2003. DRA argues that the PD misreads D.04-03-039 in concluding that the Commission dealt with this issue by (1) increasing the amount of OOR revenue sharing the company had proposed by $101,300, and (2) awaiting a proper cost allocation study to be submitted in this GRC. By reading D.04-03-039 in this way, DRA argues, the PD ignores Conclusion of Law (COL) 14 in D.04-03-039, which in DRA's view makes it clear that "the Commission could not have intended to remediate ratepayers with the proxy amount of $101,300," because that amount applied only "prospectively after 2003 to the rate cycle years addressed in D.04-03-039 . . ." (DRA Opening Comments, p. 9.)

DRA's argument is unpersuasive for two reasons. First, the discussion of the $101,300 adjustment in D.04-03-039 leaves no doubt that it was intended to be a substitute for whatever adjustments might be required as a result of the cost allocation study mandated by the decision. On this issue, the Commission said:

ORA requests that the Commission suspend the affiliate revenue sharing procedure proposed by SCWC, until the cost study is implemented. ORA's adjustment amounts to an increase of $101,300 in GO revenues. We adopt ORA's adjustment but characterize it as a proxy for potential adjustments that might result from the cost study. Under the circumstances here, where the utility has apparently not been following procedures adopted in its holding company decision, the adjustment is reasonable. (Mimeo. at 30; emphasis supplied.)

The second reason DRA's argument is unpersuasive is that COL 14 in D.04-03-039 dealt with an issue entirely separate from the question of how much general office expense should be allocated to GSWC's unregulated subsidiaries. COL 14 states:

SCWC customers should be reimbursed, in some manner, for the cash that was diverted from the water operations, due to the CPP. To do otherwise would result in a windfall for shareholders.

The "CPP" referred to in COL 14 stands for "Cash Preservation Plan," a program the Commission discussed in a section of the decision that is separate from the discussion of whether the affiliate rules set forth in D.98-06-068 had been properly implemented. In its discussion of the CPP, the Commission noted that the program, which was first implemented in 2001, was designed to limit cash expenditures on capital projects and O&M for SCWC's electric and water operations during the California energy crisis, which had caused a deterioration in SCWC's cash flow due to the sharp increase in energy prices. Because of this increase, the rates collected from SCWC's Bear Valley customers were insufficient to cover the costs of providing these customers with electric service. The CPP "included measures such as a hiring freeze, reductions in operating expenses, and elimination or deferral of all capital projects except for those projects that were considered essential either to meet public safety and health requirements or to provide continued service." (D.04-03-039, mimeo. at 30-31.)

While ORA did not argue that the CPP was unreasonable in view of the energy crisis, it did urge a $3.6 million reduction in SCWC's O&M revenue requirement, and a reduction of $3.2 million for the carrying charges on capital projects that had been deferred during 2001 and 2002. ORA was concerned that as a result of the CPP, "ratepayers were denied the benefit of funds that the Commission approved for those purposes," and that ratepayers were being asked to "pay[] twice for the same O&M expenses and projects." (Id. at 30.)

In its discussion, D.04-03-039 noted that while SCWC had deferred a substantial amount of capital expenditures in 2001 and 2002, the recorded capital additions for those years were nonetheless somewhat higher than the amount authorized in the company's previous GRC decision. Thus, it appeared that ratepayers were not being asked to pay for capital projects that had been deferred from the prior GRC. (Id. at 33.)

With respect to O&M expenses, however, the Commission agreed with ORA that a reduction was appropriate, because the proposed maintenance expense amounts included items deferred from previous years. (Id. at 34.) The Commission determined that in order to avoid a windfall to shareholders (who alone would benefit from the balancing account that had been set up to keep track of electric cost undercollections), water ratepayers should receive "the difference between what should have been spent for maintenance for 2001 and 2002 and what was actually recorded for those years." (Id.) After determining that this difference was $1,056,600 for Region III of the company, the Commission ordered that this amount should be amortized over a three-year period "as a reduction to the [water] ratepayers' obligation to fund maintenance expenses." (Id. at 35.)

It is evident from this discussion that COL 14 in D.04-03-039 has nothing to do with the question of how much general office expense should be allocated to ASUS. Moreover, the "windfall" referred to in COL 14 was one that might have arisen from the special-purpose balancing account, an issue that is not present here. Accordingly, we conclude the PD's reading of D.04-03-039 is correct, and we endorse it.

4.3.2. DRA's Proposal to Allocate About 30% of GSWC's General Office Costs to the Company's Affiliates is Unreasonable

In addition to arguing that GSWC's general office revenues should be reduced by $2.96 million annually to make up for "missed" allocations to ASUS in the past, DRA takes the position that a proper application of the four-factor methodology in this case should result in 18.21% of the GSWC's general office expenses being allocated to ASUS, with 3.21% allocated to CCWC.

GSWC has savagely attacked DRA's position in its briefs:

In contrast to GSWC's strict adherence to the Commission's guidelines, DRA ignored the Commission's instructions to allocate costs based on direct causation and randomly added or subtracted factors from the four-factor allocation methodology . . . This random methodology resulted in DRA's recommendation that 21.86% of the General Office expenses be allocated to GSWC's non-regulated affiliates. DRA's recommendation has no relationship to a reasonable estimate of the relative obligations or burdens imposed on GSWC's resources by ASUS. (GSWC Opening Brief, p. 13; emphasis in original.)

GSWC also argues that DRA's position defies economic logic when one compares GSWC and ASUS operations. On this issue, GSWC witness Switzer states in his rebuttal testimony:

The first indicator of the unreasonableness of DRA's proposal is simply to look at DRA's recommendation from an overall cost perspective by comparing the cost allocation to revenues generated by the Company. DRA recommends that nearly 20 percent of the general office costs be allocated to ASUS even though ASUS operations represent only about 1.5 percent of the Company's annual revenues (based on the latest recorded year).

In terms of dollars, DRA proposes allocating $5.5 million to ASUS, whose total revenues (as reported in GSWC's latest annual report) are $3.6 million. This $5.5 million figure is based on DRA's recommended level of general office costs of $27.8 million. GSWC, however, has requested general office expenses in Test Year 2007 of $42.3 million. Based on the level of expenses requested by GSWC, DRA's proposed allocation factor of 18.21 (or corrected to 19.96) percent would allocate to ASUS between $7.7 and $8.4 million in general office costs, which is more than double ASUS's annual revenues. It defies logic to think that ASUS would continue to operate at such a loss.

The second indicator of the unreasonableness is to look at DRA's recommendation for ASUS compared to DRA's allocation to GSWC's PUC-regulated water operations. As noted previously, DRA recommends that 18.21 percent of the general office costs should be allocated to ASUS. By comparison, DRA's calculation, as shown in its workpapers, would allocate only 13.74 percent of the general office costs to GSWC's PUC-regulated operations in Region I. It is inconceivable that DRA would propose an allocation to ASUS in excess of the allocation to one of GSWC's own operating regions. GSWC's Region I is a $33 million operation that is fully supported by every department at the general office. By comparison, ASUS is a $3.6 million operation that receives limited support from some, but not all, departments in the general office. (Ex. 13, pp. 5-6; footnote omitted.)

As explained below, we have concluded that DRA's proposed allocations of general office costs to ASUS cannot be accepted because of the inconsistent and unexplained way in which DRA has applied the traditional four-factor methodology. In addition to these deficiencies, we agree with Switzer that the anomalous results produced by DRA's approach raise serious doubts about its soundness. Even if one assumes that ASUS's new contracts will generate significantly more revenue in the future, the amounts of general office expense that DRA proposes to allocate to ASUS are so large that they would require special justification. DRA has not provided such a justification here.

4.3.3. DRA's Application of the Four-Factor Methodology is Too Quixotic and Unexplained to Justify Acceptance of its Cost Allocation Proposals

In addition to arguing that DRA's proposal would result in excessive cost allocations to GSWC's affiliates in view of the revenue these affiliates generate, GSWC also argues that DRA's proposal should be rejected because "DRA ignored the Commission's instructions to allocate costs based on direct causation and randomly added or subtracted factors from the four-factor allocation methodology." (GSWC Opening Brief, p. 13.) As indicated below, we conclude that this criticism has merit.

The four factors that have traditionally been used to allocate indirect costs not capable of direct assignment are set forth in the 1956 Commission memo that was admitted as Exhibit 41. As noted in the memo, the four factors to be used in allocating indirect costs are (1) direct operating expenses, (2) gross plant, (3) number of employees, and (4) number of customers. These factors have been used with a high degree of consistency over the years, and when the Commission has approved allocation formulas based on factors other than these four, it has clearly stated its reasons for doing so.

Although DRA claims that it followed the four-factor methodology in making recommendations about how much general office expense should be allocated to GSWC's affiliates, it is apparent from the record that this is not the case.

One of the key documents about allocation issues is Exhibit 45, a DRA workpaper that was introduced during the cross-examination of DRA's witness of general office issues, Mehboob Aslam. Exhibit 45 shows the allocation factors DRA applied to each of the 15 contracts held by ASUS. In most cases, three factors were applied, but the number of factors applied to a particular contract ranged from two to five. After a factor was applied to a particular contract and a percentage developed, the percentages for all the factors deemed relevant were then summed and divided by the number of factors used to yield the ultimate allocation percentage.⁶ Exhibit 45 itself does not explain why some factors were applied to particular contracts and not others.

During cross-examination, Aslam freely admitted that he added and subtracted factors for particular contracts, depending on his assessment of the factor's importance and on the quality of the available data:

Q. If we went all the way through your Exhibit 45 and we looked under each contract at the bold headings, we would see what factors you used as your methodology, right?

A. That is true, right.

Q. So sometimes there is three factors and sometimes there is five factors, right?

A. No. Five factors are only for two of the [ASUS] contracts. There are about 13 contracts listed here. So I would not adhere to [your] generalization.

* * *

Q. I'm directing your attention to Contract 12, Andrew[s] Air Force Base; Contract 13, Fort Monroe; Contract 14, Fort Lee . . . You used two factors in your allocation methodology for those specific contracts, right?

A. These are the two [factors] that were available, yes.

Q. You stated in your report that in thinking about applying the four-factor test you used your judgment and excluded one or more of the factors if such factor was likely to skew the costs. Do you recall that? [GSWC counsel then refers Aslam to Ex. 23, p. 4-8, lines 6-7.[⁷]

* * *

A. Yes. That is the general premise [] which I have used, yes.

Q. And what did you mean by `skew'?

A. `Skew'[,] that is basically making [the] allocation lopsided. Going - allocating more toward one entity and kind of deliberately not allocating a fair share to the other entities. That is what I meant by `skew.' (Transcript, pp. 799-801.)

Unfortunately, this colloquy is the clearest explanation that exists in the record for the approach Aslam used. At no point in its briefs or testimony did DRA present a clear explanation that would enable us to determine whether it is appropriate in a particular case to add to, or subtract from, the four factors traditionally used for cost allocation purposes. Without such a showing, we cannot accept DRA's recommendations. ⁸

4.3.4. GSWC Has Not Shown That the Use of Single Factors to Allocate Nearly Half of GSWC's General Office Revenue Requirement is Justified

Although GSWC is correct in asserting that (1) DRA's proposed $2.96 million reduction in the annual general office revenue requirement is unjustified, and (2) DRA's application of the four-factor methodology cannot be accepted, this does not mean that the company is entitled to prevail on its contention that under D.03-09-021, it is reasonable to allocate nearly half of GSWC's general office costs on the basis of single allocation factors. As we shall see, although there have been some exceptions, the Commission has generally frowned on the use of single allocation factors, and there is good reason to reject them here.

GSWC's claim that it is appropriate to use single allocation factors rests upon a passage in D.03-09-021, which accepted a settlement negotiated in a general rate case involving California Water Service Company (CalWater or CWS). Although the Commission accepted the settlement, it criticized CalWater for failing to have in place a methodology for allocating indirect general office costs to affiliates, and it directed the company to develop such a methodology. In so ruling, the Commission made the following statement, on which GSWC relies:

In developing an allocation methodology, we direct Cal Water, where feasible, to rely on a cost-causation based factor to allocate common expenses, costs, or plant. For example, for billing services, it would be meaningful to allocate expenses by the number of bills sent out or by the hours the employees and equipment were used for regulated and non-regulated services." (Mimeo. at 27-28; emphasis supplied.)

This statement - a dictum describing two approaches that seemed theoretically reasonable for allocating the costs of bill preparation - stops well short of being "an allocation methodology approved by the Commission," as GSWC's Switzer claims.

We acknowledge that in a handful of cases over the years, the Commission has suggested that the use of a single factor can be appropriate for allocating some kinds of indirect costs. For example, in D. 80207, 73 CPUC 597 (1972), the Commission agreed with the applicant that it was more reasonable to allocate the costs of maintaining customer accounts and preparing bills on the basis of the number of customers rather than on the traditional four-factor methodology.⁹ However, no decision we are aware of suggests that it is appropriate to allocate nearly half of a company's indirect, general office costs on the basis of individual allocation factors, as GSWC is proposing to do here.¹⁰

Contrary to GSWC's position, Commission decisions in recent years have either approved the use of the traditional four-factor methodology, or the use of less than four factors if it can be demonstrated that one or more of the traditional factors are irrelevant or would skew the allocation study results in unreasonable ways.

In D.03-05-078, for example, one of the issues was how much of the expenses of the corporate parent of Suburban Water System (Suburban) should be allocated to Suburban. The company argued that the Commission should use the traditional four-factor analysis, whereas ORA argued that only three factors should be used. In accepting ORA's position, the Commission explained that using four factors in the manner advocated by Suburban would shift costs on to ratepayers:

In allocating parent company expenses to subsidiaries, the Commission generally follows a four-factor approach, measuring each subsidiary's (1) direct operating expenses, (2) end-of-year gross plant, (3) total customers, and (4) payroll. The results are applied to determine a subsidiary's share of its parent company expenses. Suburban applied these four factors to its allocation. ORA applied a three-factor test, eliminating `customers' of each subsidiary because non-regulated subsidiaries like ECO reported that they had clients rather than customers. By entering `0' for ECO customers, ECO's share of parent company expenses was reduced, and Suburban's share was increased, despite ECO's annual revenue of $62 million or more.

ORA notes that it has used two- or three-factor analyses for other Class A water companies where appropriate . . . ORA's analysis is persuasive, and we adopt the ORA [three-factor] allocation formula in this proceeding. Suburban thus is allocated 32.6% of the parent company costs, rather than the 45.2% recommended by Suburban. (D.03-05-078, mimeo. at 21-22.)

In this case particularly, the single factor that GSWC is proposing to use to allocate nearly 40% of its general office costs - an entity's number of employees - seems likely to result in a shift of costs away from GSWC's unregulated affiliates and toward its ratepayers.

The allocation study attached to Switzer's direct testimony indicates that as of September 30, 2005, GSWC and its affiliates had 505 employees, of whom only 14 were employed by ASUS. (Ex. 6, Switzer Schedule B, p. 2.) But it is obvious from the annual reports of GSWC's corporate parent, American States Water Company (ASWC), that ASWC hopes for and anticipates substantial growth in ASUS's operations over the next few years. The 2004 annual report states that ASUS had submitted bids to operate the water systems of over 20 military bases, while the 2005 annual report indicates that ASUS has won contracts through subsidiaries to operate the water and wastewater systems of Fort Bliss, Texas; Andrews Air Force Base, Maryland; and Fort Monroe, Fort Eustis and Fort Story in Virginia. (ASWC 2005 Annual Report, Letter to Shareholders of Lloyd Ross and Floyd Wicks.)¹¹

It is difficult to believe that over the next three years, the small number of personnel employed by ASUS at the end of 2005 could operate the water and wastewater systems of these military bases, some of which are quite large. Moreover, it seems likely that as ASUS (or its subsidiaries) add employees to handle the increased workload from the bases, the new employees will need significant assistance from more-experienced GSWC operating personnel. Thus, allocating nearly 40% of GSWC's indirect costs on the basis of 2005 headcounts for GSWC and its affiliates - and just before an expected ASUS growth spurt - seems likely to produce results that do not fairly approximate the demands GSWC's unregulated affiliates place on its personnel.¹²

In D.01-06-077, which was cited with approval in D.03-09-021, the Commission noted that using traditional cost allocation formulae when substantial growth is taking place at a utility's new, unregulated affiliates can produce unreasonable results. In D.01-06-077, one of the issues was whether the Commission should use a three-factor allocation formula advocated by Roseville Telephone Company (RTC), or a general allocator based on expenses that had been approved by the FCC and was favored by ORA. ORA opposed the three-factor formula favored by RTC on the ground, among others, that it placed undue emphasis on past asset accumulation, since RTC's new affiliates had not had time to accumulate significant assets.¹³ The Commission agreed with ORA:

We are persuaded by ORA that RTC's three-factor formula does not reflect cost causation and instead over-allocates costs to RTC. ORA correctly points out that the three-factor formula over emphasizes asset accumulations, both through the gross plant factor and through depreciation expense reflected in the expense factor. As a mature company, RTC has accumulated considerable assets over a long period of time. In contrast, in a dynamic and fast changing period in the telecommunications industry, most of RTC's affiliates . . . were just coming into existence during the audit period. Even though these affiliates obviously required the expenditure of general and administrative costs, they have had little time to accumulate assets. Consequently, the use of accumulated assets as a significant factor in allocating common costs - as reflected in the gross plant factor and the depreciation component of the expense factor - does not provide a reasonable approximation of the extent to which affiliates caused common costs to be incurred. (D.01-06-077, mimeo. at 47.)

D.01-06-077 closed its discussion of the allocation issue by noting that a principal purpose of allocation rules is to "guard[] against cross-subsidy of nonregulated ventures by regulated services." (Id. at 50, n. 5, quoting In re Separation of Costs of Regulated Telephone Service from Costs of Nonregulated Activities, FCC 86-564, 62 Rad. Reg. 2d (P&F) 163, pars. 33, 37.) In this case, in view of the growth in ASUS that can reasonably be expected in the near future, GSWC has not shown that the allocation method it favors would avoid subsidizing ASUS at the expense of GSWC's ratepayers. For that reason, we cannot accept GSWC's proposal, which would allocate nearly 40% of the company's general office costs on the basis of the number of employees.

An additional reason we are rejecting GSWC's proposed approach is that, like Suburban in D.03-05-078, GSWC has skewed the operation of the traditional four-factor methodology - which it uses to allocate the remaining 60% of its general office costs - by assuming that ASUS had only 11 customers, one for each of the contracts that ASUS held on September 30, 2005. (See Ex. 47, p. 1.)

This single assumption makes a significant difference in the outcome of the four-factor methodology. Even though - according to Exhibit 46 - 91,115 customers received service through the 11 entities with which ASUS held contracts at the end of 1995, the practical effect of assuming ASUS had only 11 customers is to assign one of the four traditional allocation factors - the number of an entity's customers - a value of zero. It is clear from the discussion in D.03-05-078 that we have disapproved of this practice because it results in a serious distortion of the four-factor methodology.

However, we also do not approve of the solution that DRA's witness Aslam used to "correct" this distortion; viz., assuming that 74,270 customers served by the 11 entities with which ASUS had contracts should be considered the equivalent of full GSWC customers. We disapprove for two reasons. First, Aslam could not clearly explain how he arrived at the figure of 74,270. Second, as he was forced to concede during cross-examination, there is considerable variation in the nature of the services that ASUS provides to its contracting parties, a situation that can make it inequitable to assume that each customer served by an entity with which ASUS has a contract is equivalent to a full GSWC customer.¹⁴

In view of the variation in services that ASUS provides to the entities with which it contracts, there is clearly a need to develop a methodology for determining a "weighted" number of customers for these entities that reasonably reflects the level of service ASUS actually provides. In the next section, we suggest one such method and then apply it to the data in the record.

4.3.5. As an Interim Expedient, GSWC Will Be Required to Use a Three-Factor Formula Based on Total Expenses, Total Labor and a Weighted Number of Customers

Although we are rejecting the allocation formulas proposed by both GSWC and DRA, we are well aware of the need to develop an allocation formula that is acceptable, at least for this general office rate case. Under the rate case plan adopted in D.04-06-018, GSWC was supposed to file a rate case for its general office operations in February 2005. The company did so, but as noted in D.06-01-025, the company and DRA filed a stipulation on August 3, 2005 in which they agreed to (1) defer a decision on the general office rate case for one year, and (2) in the interim, use certain percentages and amounts as the basis for calculating the share of general office expenses to be allocated to GSWC's Region III ratepayers. The Commission accepted this stipulation. (See, D.06-01-025, mimeo. at 66-69 and Appendix B.) In view of the fact that a decision on GSWC's general office expenses has already been deferred once and it will not be filing another general office GRC until July 1, 2008, we do not wish to defer a decision again.

However, the paucity of reasonable allocation proposals offered by the parties raises the issue of how an acceptable allocation formula can be devised for those general office expenses that cannot be directly assigned by GSWC. We have concluded that the most reasonable allocation formula, in view of the likely growth of ASUS's operations and the increased demand on GSWC's general office services that can be expected as a result, is to use a variant of the three-factor allocation approach that the Commission has recently employed in GRC decisions such as D.03-05-078.

Under this approach, the three factors we will examine for GSWC and its affiliates are (1) total labor costs, (2) total expenses (including, in the case of affiliates, the affiliate's own Operations and Maintenance (O&M) costs as well as costs that are direct-billed by GSWC), and (3) a weighted average number of customers based upon (a) the number of ultimate customers, ratepayers or connections served by the entity with which the GSWC affiliate has a contract, and (b) the nature of the services provided by the affiliate.

This approach, which can be implemented using the data in the record plus reasonable approximations based on past SCWC rate cases, will help to ensure that the general office costs allocated to GSWC affiliates - especially ASUS - fairly reflect the demands that the operations of these affiliates actually place on GSWC's resources. By examining total labor costs, for example, we are examining the nature and extent of the work actually performed for the entity under consideration. In the case of an affiliate such as ASUS, while the number of employees shown on a formal organization chart may not fairly reflect the full extent of the work performed by the affiliate, measuring total labor costs without regard to whose employee is performing the work should give a more accurate picture of the size of the enterprise. Likewise, measuring total expenses - including those billed by GSWC - should help to give a more accurate measure of the total work undertaken by the affiliate, more illuminating than the affiliate's total revenue or gross plant (which, as noted in D.01-06-077, may only recently have begun to grow.)

The most challenging of the three factors - both conceptually and computationally - is the weighted percentage of customers that should be attributed to the affiliate. In the case of CCWC, the computation is easy because it is a full-service utility, and 100% is appropriate.¹⁵ In the case of ASUS, however, the computation is more difficult, because - as noted above - ASUS provides varying levels of service to those entities with which it now contracts.

Among the ASUS contracts, the recent ones with military bases are relatively easy to evaluate in terms of the number of "customers", because all of these agreements - which concern Fort Bliss, Texas; Fort Lee, Virginia; Andrews Air Force Base, Maryland; and Forts Eustis, Monroe and Story, all of which are also in Maryland - essentially call for ASUS to provide full water and wastewater services to these bases. It is therefore appropriate to use 100% of the connections at these bases to determine the appropriate weighted percentage customer count.¹⁶ Based on the data set forth on page 2 of Exhibit 46 (which is GSWC's response to a DRA data request), the combined number of connections for all of these military contracts combined is 12,614.¹⁷

In cases where ASUS is providing less-than-full utility services, determining the weighted number of customers is more complex, because the extent of the services offered to the contracting parties - most of which are medium- to small-sized municipal utilities - varies from contract to contract. However, an appropriate discount factor can be developed using the ratios that O&M expenses minus supply costs, Administrative and General (A&G) expenses, amortization and depreciation, and taxes paid by GSWC bear to GSWC's net operating revenues (minus supply costs and cost of capital) in recent rate cases.¹⁸

We have decided to examine these items because they assure a reasonable degree of comparability between GSWC and the entities with which ASUS contracts. It makes sense to exclude supply costs, for example, because all of the parties with which ASUS has contracts are responsible for supplying their own water. Similarly, even though military bases and municipal utilities may use different accounting terminology, all of them should be putting aside money to replace water-related assets as they wear out over time. Finally, even though municipal utilities and military bases do not pay taxes (or pay smaller amounts of tax than do private utilities such as GSWC), consideration of GSWC's tax burden is required to assure reasonable comparability between the proportion of its total A&G and O&M expenses (less supply costs) and those of the entity with which it is being compared.

Using this approach, if one examines Appendix D to D.06-01-025, which sets forth the summary of earnings for GSWC's Region III for 2006, net operating revenues minus supply expenses and cost of capital) equal $43,666,600. This total is comprised of the following elements: (1) total O&M expenses less supply costs ($11,383,800), (2) total A&G expenses ($13,304,900), (3) depreciation and amortization ($8,162,500), and (4) total taxes, including property, payroll and income taxes ($10,815,300). Table 3 in Attachment B to this decision sets forth comparable data for Regions I and II of GSWC, and derives appropriate A&G, O&M, amortization/depreciation and tax percentages for the entire company.

The next task is to apply the company-wide A&G and O&M percentages thus derived to particular ASUS contracts. In the case of ASUS's contract with the City of Torrance, for example, ASUS has agreed to provide a full range of A&G support services (including billing, cash processing and call handling), but it has not agreed to provide any other services. Since A&G expenses comprise 30.1% of GSWC's net operating revenues less supply expenses and cost of capital for the company's three regions, it therefore makes sense to attribute 30.1% of the 34,000 customers shown for the City of Torrance on Exhibit 46 to the ASUS contract for purposes of the allocation formula we will be using. Using this approach, the appropriate weighted number of customers attributable to the ASUS-Torrance contract amounts to 10,234 (30.1% x 34,000 = 10,234).

In the case of ASUS's contract with the City of Tustin, on the other hand, ASUS has agreed to provide meter reading, a labor-intensive O&M service, in addition to various A&G services. In this case, we think it is appropriate to attribute one-third of the percentage that O&M expenses less supply costs comprise on GSWC's system, or 8.2% (24.6% ÷ 3 = 8.2%), to this O&M service. When added to the 30.1% attributable to the A&G services that ASUS provides to Tustin, this amounts to 38.3% (30.1% + 8.2% = 38.3%). Since the total number of customers shown for the City of Tustin on Exhibit 46 is 15,000, the correct weighted number of customers to attribute to the ASUS-Tustin contract is 5,745 (38.3% x 15,000 = 5,745).¹⁹ Table 2 in Attachment B to this decision shows the derivation of the weighted number of customers we are attributing to ASUS's contracts with other non-military customers.

The final step in the process is to average the percentage allocations of customers attributable to GSWC and its affiliates, CCWC and ASUS, with the percentage allocations for the other two factors we are examining for these entities; i.e., total labor costs and total expenses. The total labor costs and total expense figures we use are taken from Exhibit 47, the first page of which was taken from Switzer's own cost allocation study. The averaging process for the three factors, which yields the overall cost allocation factors, is shown on Table 1 of Attachment B.²⁰ These percentages are as follows:

However, as Mr. Switzer points out in his direct testimony, once GSWC's overall share of general office costs has been determined, the final step in the process is to assign this share to the company's three water regions and to its small electric company, BVEC, which serves the Big Bear vacation area. (Ex. 6, Switzer, p. 16.)

In his own study, Switzer determined the amount of GSWC general office costs assigned to BVEC, 10.27%, by using the Commission's traditional four-factor allocation methodology, the results of which are shown on Schedule C of his study. (Exhibit 6, Switzer Schedule C.) We have decided this same percentage should be used here because, as we understand it, BVEC is a company GSWC has owned for some time, and it is not growing rapidly. Thus, use of the traditional four-factor methodology to determine BVEC's share of general office costs versus those of GSWC's three water districts does not raise the same questions of subsidization that has caused us to reject the four-factor methodology for determining the share of overall general office costs that should be borne by ASUS.

Similarly, the percentages of general office costs that should be assigned to GSWC's three water districts, which is shown in column (b) of Schedule D of Switzer's study, represents a reasonable application of the traditional four factor methodology and should be used here. Applying the percentages shown in column (b) of that schedule, the share of general office costs attributable to GSWC's California water operations that should be assigned to the three districts (after first making the proper allocations to ASUS, CCWC, and BVEC) are as follows: Region I, 19.60%; Region II, 40.91%, and Region III, 39.49%.²¹

5. GSWC's Request for a New Customer Information/Customer Relationship Management (CIS/CRM) System is Reasonable, But the Amount Requested is Unsupported and Will Need To Be Established by a Tier 3 Advice Letter

A substantial part of the increase in ratebase that GSWC has requested for its general office is due to the proposed purchase of a new computer system for handling customer service issues. GSWC has requested $9.1 million for this purpose (exclusive of overheads), and asserts that over three years, this amount is needed to serve customers in its regulated operations. GSWC proposes to spend $2,982,841 in 2006 for the first phase of purchasing and implementing the new system. GSWC refers to the new system as the Customer Information System/Customer Relationship Management System; we will refer to it as the CIS/CRM system. GSWC refers to the old system it currently uses as the Customer Information and Billing System; we will refer to this old system as the CIS system.

The company's witness on the CIS/CRM issue was Yvonne Andres, who has worked with the existing CIS system during her entire career. Since 1997, she has been the supervisor for the system and the staff who operate it. She is well-acquainted with its limitations and lays out a detailed case for replacing it.

While DRA's testimony does not dispute that a new CIS/CRM will eventually be needed, it opposes the company's request in its present form. In particular, DRA asserts that GSWC's current cost estimates for the system are "too generic and too preliminary," and that it appears a significant portion of the new system may be devoted to GSWC's non-regulated affiliates. In view of this uncertainty, DRA argues:

It would be more prudent to evaluate the cost estimates that will be put forth in a formal [Request for Proposal, or RFP] from the Company's CIS consultants. At that time a reasonable evaluation on the capabilities and features of the new CIS System[,] along with the Company's internal, regulated and external, Non-regulated needs could be effectively measured. (Ex. 23, pp. 3-3 to 3-4.)

As set forth below, we have concluded that on this issue, DRA has the better of the argument, and that recovery of costs for the new CIS/CRM system should await the submission of an advice letter with much more detailed cost information than is set forth in GSWC's testimony.

5.1. GSWC's Rationale for Replacement of the Existing CIS System

In her testimony, Andres lays out a detailed justification for why the proposed new CIS/CRM system is needed. She begins by noting that the present CIS system has significant limitations with respect to its age, design and system documentation, and she asserts that "the risks of remaining on the current system are substantial enough to jeopardize normal daily operation of the company." (Ex. 5, Andres Testimony, p. 9.)

With respect to the age of the CIS system and the limitations it creates, Andres states:

GSWC's current system was installed in June 1994, but the system itself was actually developed back in 1977. The system utilizes Report Program Generator (RPG) as its programming language, which originated in the 1960s as a report-building program and evolved into a procedural programming language. Like other languages of its type and age, such as COBOL, it has proven cumbersome and hence costly to modify. It is increasingly difficult for the vendor of the system to hire RPG programmers, as the RPG programming language is considered an obsolete skill. Due to the vendor's difficulty in finding and hiring RPG programmers, system modifications routinely take an excessive amount of time to deliver, sometimes later than promised to and needed by the company. (Id. at 2.)

With respect to the CIS system's design, Andres notes that when it was designed in 1977,

[T]he needs of the utilities [were] . . . very stable and static. [The CIS system] was not designed to easily accommodate the realities faced by the utilities today, such as the need:

To implement changes in business rules and processes such as electric deregulation, Sarbanes-Oxley compliance, and Department of Health Services reporting;

To access and analyze customer and billing information for effective and proactive management decision-making;

For a user-friendly interface to the customer and billing information;

To exchange data to or from other utilities such as meter reading management software, third-party payment vendors, financials software, mobile-computing software, and knowledge-management software;

For customer self-service through telephone Interactive Voice Response (IVR) systems, Internet access to account information, and Electronic Bill Presentment and Payment (EBPP). (Id. at 2-3.)

Andres continues that in order to meet modern requirements, GSWC has recently had to increase the budget for programming CIS modifications from $50,000 to $100,000, and that this latter amount "accommodate[s] only the highest priority requirements." She also asserts that some of GSWC's needs - such as mobile computing, Internet access to account information and knowledge management - "cannot be cost-effectively addressed with the current system." (Id. at 3.)

Andres also emphasizes that deficiencies in the documentation for the 1977 CIS system have made it difficult to implement modifications even when they are high-priority:

The vendor's system documentation is unreliable and, in some cases, non-existent. So, programming modifications are time-consuming because the vendor programmers must tediously determine how to program the requested modifications without impacting existing system processes. Programming modifications are also prone to errors due to unreliable and/or non-existent documentation. Erroneous programming modifications have been implemented into production, sometimes resulting in erroneous billing calculations. This is becoming a major issue with regards to internal control and Sarbanes-Oxley compliance. (Id.)

Andres concludes her discussion of the CIS system by giving examples of problems that have recently occurred because of the difficulties in modifying the system. These examples include the following:

-- As identity theft in Social Security Numbers became a major problem, our vendor did not have an enhancement ready for the Company to safeguard the SSNs collected from our customers in the database. GSWC submitted a program modification request to the vendor at the beginning of 2004 and was told that it [would] cost a significant amount of programming time and costs. Our vendor was unwilling to further develop the old system and preferred us to migrate to their `newer' system. As GSWC insisted that this modification is extremely important and should be treated as the highest priority, the vendor agreed to work on the modification request. This program modification was not delivered until December 2004.

-- Service orders are currently generated at local customer service areas every morning and distributed to Water Distribution Operators. When the Operators complete the jobs, they will manually write up the report and pass the service orders back to the office at the end of the workday. The Customer Service Representatives will then manually input the information into the system and close the service order. This business practice is proven to be inefficient and input errors happen[] all the time. Service orders are not closed in a timely manner[,] thus increasing customers' dissatisfaction . . . (Id. at 4-5.)²²

Andres believes there would be at least 10 general advantages to implementing the new CIS/CRM system she is advocating. These advantages include (1) agility in support of new business requirements, such as Sarbanes-Oxley, (2) lower training costs (such as cutting the training time for a customer service representative in half), (3) better access to and organization of information, (4) a lower incidence of errors, (5) better control of business rule changes, without the need for vendor intervention, (6) tighter user, application and field security, (7) improved customer service, including self-service through web-based services, (8) lower vendor support costs, (9) faster response to problems, due to an updated technology platform and tools, and (10) increased availability of skilled technologists who do not need to be proficient in outdated programming languages. (Id. at 6-8.)

In her direct testimony, Andres acknowledges that the $9.1 million cost given for the new CIS/CRM system is an estimate, and that precise costs will not be available for some time:

The $9.1 million dollar amount is an estimate of cost for a new CIS/CRM, based on standard high-level pricing models of two vendor-independent consultants. A more accurate dollar amount will be obtained once GSWC completes the full RFP phase for a new CIS/CRM. GSWC has not yet issued formal RFPs because, once a vendor offers a firm proposal in response to an RFP, the proposal will typically expire in a period of several months . . . GSWC has issued an RFP for a CIS consultant to assist GSWC in evaluating, selecting and implementing a new CIS . . . The consultant will be selected by end of first quarter 2006. With the consultant, GSWC expects to issue the RFP for a new CIS in the third quarter of 2006, finalize contract negotiation by year-end 2006, then begin CIS implementation during first quarter 2007. (Id. at 9.)

5.2. DRA's Opposition to Including the Costs of the New CIS/CRM System in this Rate Case

In its report on general office issues, DRA opposes approval of the amounts GSWC has requested for the new CIS/CRM system because (1) GSWC's cost estimates are unreasonably vague, and (2) it appears that a significant portion of the new system's capacity may be devoted to serving the needs of customers of GSWC's non-regulated affiliates. After noting the preliminary nature of GSWC's cost estimates, and the fact that the company has only recently begun the process of hiring a CIS consultant, DRA's report continues:

[T]he Company fully utilizes its Customer Service Center resources to serve a great number of customers in its Non-regulated businesses. For example, currently the Company is serving approximately 74,270 Non-regulated customers under Customer Service Contracts, and the numbers are growing. The Company constantly pitches its `state-of-the-art' Customer Service Center to attract more Non-regulated business . . .

* * *

What is the driving force behind the need of replacing existing CIS System? Is it the obsolete software language or the demand that the Non-regulated businesses are putting on the Company? For example, in one of its Non-regulated contracts with [the] City of Torrance, the City puts . . . stringent Customer Service Performance Standards on the Company . . .

* * *

It is therefore evident that replacing the existing CIS System must take the Non-regulated related costs into account. Currently, GSWC based its generic costs only on the number of regulated customers; however, once the new System is installed it will also be used to service the Non-regulated businesses['] needs.

The current cost estimates are too generic and too preliminary, rendering approval of this project at this stage not good sense. It would be more prudent to evaluate the cost estimates that will be put forth in a formal RFP from the Company's CIS consultants. At that time a reasonable evaluation on the capabilities and features of the new CIS System[,] along with the Company's internal, regulated and external, Non-regulated needs[,] could effectively be measured. (Ex. 23, pp. 3-2 to 3-4.)

5.3. Discussion

Although Ms. Andres has presented a good case for why GSWC needs the new CIS/CRM system, and has sought to rebut a number of the points made in DRA's testimony, we agree with DRA that GSWC's cost estimates are too vague, and that there are too many questions about how much of the new system's capacity will be used for GSWC's non-regulated affiliates, to allow us to approve the CIS/CRM funding request proposed in this application.

Rather than approve the $9.1 million (before overheads) that GSWC is seeking here for the new CIS/CRM system, we have decided to approve only the $2.983 million that the company proposed to spend on the system in 2006. (Ex. 5, Andres, Schedule 1.) In order to recover any additional amount, GSWC will have to submit a detailed Tier 3 advice letter which will be subject to protest by DRA, the City of Claremont, and any other interested party. Before it can be approved by Commission resolution pursuant to General Rule 7.6.2 of General Order (GO) 96-B, the advice letter will have to demonstrate that (1) the new CIS/CRM system is designed principally to meet the needs of GSWC's customers, and (2) any excess capacity in the system is designed to allow for the growth in the number of GSWC customers (and the applications they may need) that can reasonably be expected during the useful life of the CIS/CRM system. GSWC will also be required to demonstrate in the advice letter that it has developed an adequate methodology for charging directly to GSWC's affiliates, whether regulated or non-regulated, a share of the new CIS/CRM system's costs (including overheads) that is fully proportionate to the demand these various affiliates (and their customers) will place upon the new system while it still has excess capacity. The advice letter must clearly explain this methodology, and must demonstrate that the CIS/CRM costs directly charged to the affiliates will not be aggregated with other costs in a way that renders them less than fully transparent. As with other Tier 3 advice letters, the Water Division will be free to seek as much additional information from GSWC as it considers necessary to prepare a resolution concerning the advice letter for the Commission's consideration. We will also require, in addition to the other service requirements imposed by GO 96-B, that GSWC serve the Tier 3 advice letter upon the assigned Commissioner and the assigned ALJ for this proceeding.

We have decided upon this treatment because, among other reasons, the cost estimates given by Andres are very vague. Not only are they admittedly general estimates "based on standard high-level pricing models of two vendor-independent consultants," (Ex. 5, Andres, p. 9), but an estimated total cost for the CIS/CRM system - including those portions that would serve GSWC's affiliates - is not even presented in the company's testimony.

We have commented in Footnote 12 of this decision on the inadequacy of the discussion in SCWC's affiliate transaction reports for 2000-2003 concerning the amounts the company charged directly to affiliates. Those concerns are especially relevant here, where it seems possible that the clients or customers of ASUS, GSWC's principal unregulated affiliate, will demand even more detailed billing and other information than the residential and business customers in GSWC's three regions are accustomed to receiving. It seems likely that such information demands will place commensurately greater burdens on the resources of the new CIS/CRM system.

While Andres's rebuttal testimony addresses a number of the specific points raised by DRA, it is significant that she does not deal with the larger issues that DRA raises. Thus, for example, Andres spends a good deal of time rebutting DRA's claim that "the `driving force' behind replacing the CIS/CRM system is to serve `a great number of customers in its Non-regulated businesses.'" (Ex. 14, p. 1.) Andres's rebuttal on this point includes a table purporting to show that the number of "Non-regulated customers" - i.e., those served via the ASUS contracts with Brooke Utilities, Inc., the City of Torrance, the City of Bell Gardens, the Goleta Water District and Wellspring International, Inc. - fell from 67,892 in 2002 to 43,913 in 2006. (Id. at 2.) She also states:

Furthermore, ASUS growth activities are no longer focused on Customer Service Contracts. This is evident by the lack of new Customer Service Contracts in the past six years. Rather, ASUS activities are and have been focused on contracts that would not benefit from the Company's new CIS System." (Id. at 2-3; emphasis in original.)²³

It is noteworthy, however, that Andres does not deny that the new CIS/CRM system has apparently been sized in part to meet the needs of clients and customers served through GSWC's affiliates.²⁴ Nor does she attempt to address the obvious questions about what kind of demands, and the magnitude of those demands, that the contracts with military bases ASUS has won (and in some cases is still pursuing) are likely to place on the new CIS/CRM system. Without clear answers to these questions, we cannot approve the funding for the CIS/CRM system that GSWC is seeking in this application.

We view the Tier 3 advice letter process that we wish to use for determining how much funding GSWC should receive for the CIS/CRM system as an updated version of a process that parties in water cases have occasionally used during recent years. In D.05-07-022, for example, we approved a settlement involving the use of advice letters for capital improvements in consolidated rate cases filed by CWS. The decision explained the parties' use of advice letters as follows:

An important feature of the Settlement is the proposal to exclude many plant additions pending the completion of these additions. Parties propose that as each plant addition is completed and in service, CWS may recover the cost through an advice letter filing. Furthermore, each plant addition will be `capped,' thus establishing the maximum amount that can be included in each advice letter. Should the recorded cost exceed the cap for any plant addition, the excess cost will be reviewed for reasonableness in the next GRC for the specific district in which the plant addition is located." (Mimeo. at 17; footnote omitted.)²⁵

In this case, we think that the paucity of information that has been furnished about the new CIS/CRM system's costs makes it inappropriate to establish a cap for those costs. However, we caution GSWC that if it fails to make the detailed showing described above in its Tier 3 advice letter, then we may well conclude in the resolution concerning the advice letter that it must be rejected, and that GSWC will have to proceed by application to recover the additional costs of the CIS/CRM system. ²⁶

6. The 20 New General Office Positions GSWC is Requesting for Reasons Other Than Compliance With the Sarbanes-Oxley Act

One of the principal issues between GSWC and DRA concerns 25 new positions that the company is seeking for its general office operations. When added together, the salaries for the disputed positions total approximately $1,850,000 (precise salaries are not stated for a few of the more modestly-paid ones). The positions cover a wide range of levels and functions, ranging from a Senior Vice President for Operations (at an annual salary of $209,000) to three Customer Service Representatives (at an annual salary of $36,349 each). Several of the challenged positions relate to GSWC's information systems, including an Application Support Manager ($113,883) and an Assistant Information Technology Manager ($88,564).

In almost all cases, DRA has challenged the need for these positions on the ground that they would duplicate work other people are now performing within the company. For example, with respect to the Application Support Manager, DRA argues:

It is obvious that a duplication of Application Support functions exist[s] in each major functional area. The new Application Support Manager position will not replace the existing functional area applications support resources. The ratepayers will have to bear unnecessary rate burdens because of GSWC having functions duplicated at the centralized and decentralized levels." (Ex. 23, p. 2-12.)

We conclude below that although DRA's criticisms have merit in a few instances, they are misplaced in a large majority of cases. For example, it is clear from both the testimony on the new positions and from Ms. Andres's testimony advocating the new CIS/CRM system that one reason GSWC is seeking the new information technology positions is to reduce its dependence on outside vendors. To accept DRA's arguments that these positions should not be allowed would amount, in most cases, to being penny-wise and pound-foolish.

GSWC's principal justification for five of the new general office positions it seeks is the need to comply with the Sarbanes-Oxley Act of 2002 (SOX). Those positions raise special issues, and we discuss them separately in the next section of this decision.

Although we are allowing virtually all of the non-SOX-related positions GSWC is seeking for its general office, this does not mean we condone the manner in which the company handled the submission of its testimony. Although GSWC presented a justification for each of the disputed positions in the direct testimony it submitted in February 2006, the company presented a considerably more extensive justification for the positions - especially those related to SOX - in the rebuttal testimony that GSWC submitted on June 9, 2006. The volume of this rebuttal testimony was so large, and the time to consider it so short, that DRA moved to strike large portions of the testimony in a motion filed on June 28, 2006.

Although we affirm the assigned ALJ's decision not to strike this rebuttal testimony,²⁷ we also endorse his view that - since the company had previously been criticized in D.04-03-039 for waiting until rebuttal to offer the principal justification for important proposals - GSWC's conduct was "not . . . exemplary" and should not be condoned. Accordingly, as explained in the final part of this section, we recommend imposing a $50,000 penalty on GSWC for its conduct.

6.1. Procedural Background of Motion to Strike

The original justification for the disputed general office positions (including those related to SOX) was set forth in the direct testimony of Jenny Darney-Lane, which was included in Exhibit 5 and filed on February 5, 2006. Darney-Lane's testimony covered new labor expense for both Region II and GSWC's general office; her testimony on the new general office positions totaled 35 pages.

On May 25, 2006, DRA filed its responsive testimony in the form of several reports. The testimony on general office issues was included within Exhibit 23 and was sponsored by DRA witness Mehboob Aslam. His testimony on the disputed general office positions comprised 23 pages.

Pursuant to the procedural schedule the parties had agreed to at the May 2, 2006 PHC, GSWC filed its rebuttal testimony on June 9, 2006. On this round, the company's rebuttal on the general office positions not related to SOX was sponsored by Joel Dickson, GSWC's Senior Vice President for Operations and Administration. Dickson's rebuttal testimony comprised 87 pages plus attached exhibits and was admitted into evidence as Exhibit 11.

As noted above, DRA moved to strike all of Dickson's rebuttal testimony (as well as a portion of Robert Sprowls's testimony concerning the SOX positions) on June 28, 2006. GSWC filed a reply on July 5, 2006. Although the ALJ found that GSWC's conduct in the matter had "not been exemplary" (especially in view of the admonishment the company had received in D.04-03-039), the ALJ Ruling Denying DRA Motion to Strike also concluded that the prejudice to DRA did not appear to be so great as to justify striking the entirety of Dickson's testimony. Instead, the ruling concluded, the preferable course was to follow the Commission's usual practice of admitting the testimony, but then "afford[ing] it only so much weight as the presiding officer considers appropriate." (Ruling, p. 2.)

6.2. The Testimony of GSWC and DRA Concerning the Disputed General Office Positions Not Related to SOX

As noted above, much of the justification for the 20 disputed general office positions not related to SOX is contained in the rebuttal testimony of Joel Dickson, which comprises 87 pages plus extensive exhibits. This rebuttal testimony is divided into three parts. The first 31 pages are concerned with a discussion of 11 changes in the regulatory landscape that allegedly support the need for the new positions. The next 31 pages (pp. 31-62) set forth a justification for each of the disputed jobs. In the final portion of his testimony (pp. 63-87), Dickson offers an answer to Aslam's criticisms of GSWC's in-house training program that is known as the EDU, as well as to other DRA claims, including the contention that the company withheld information about some general office positions in the prior GRC on general office issues, A.02-11-007.

6.2.1. The 11 Factors That GSWC Claims Have Significantly Increased the General Office Workload

Because the rebuttal testimony frequently refers to the 11 factors that Dickson claims have changed the regulatory landscape and increased general office workload, even though the number of GSWC's customers has remained about the same, we begin with those 11 factors. First, Dickson argues that the need for infrastructure replacement has increased general office needs. He states that in 1996, GSWC undertook 164 capital projects to replace worn-out water supply and distribution facilities, while in 2006 it planned to undertake 276 such projects. Dickson also states that the size of the company's engineering staff has not increased during this period; instead, to handle the additional work, the company has had to hire outside engineering firms such as CH2M Hill. Dickson also notes that infrastructure replacement increases the demands on other departments (such as GSWC's purchasing department), and requires more coordination between regional management and the communities where streets are being torn up. (Ex. 11, pp. 6-10.)

Second, Dickson argues that GSWC's practice of applying for low-cost financing for its projects, especially under Proposition 50, would be undermined if DRA's staffing recommendations were to be accepted. According to Dickson, the company's 67 applications under Proposition 50 not only required over 500 hours of engineering department staff time in 2005, but also significant amounts of lobbying in the Legislature by senior executives to ensure that private water companies could be beneficiaries of Proposition 50 funding. (Id. at 10-11.)

Third, Dickson argues that the increasingly stringent water quality regulations of the past decade (such as for arsenic) have increased the need for general office staff. GSWC operates 41 water systems in California, and Dickson notes that the new water quality regulations are more complex than their predecessors and often require increased monitoring and management attention. He notes, for example, that when Unregulated Contaminant Monitoring Rule 1 (UCMR1) took effect in 1999, many companies including GSWC found that their contract laboratories had difficulty in reporting the relevant data directly to the U.S. Environmental Protection Agency, as UCMR 1 required, which led to numerous notices of violation nationwide. Dickson also points out that if a well is found to exceed Maximum Contaminant Levels (MCLs), numerous steps and permits are usually necessary before the well can be put back into service. (Id. at 12-15.)

The fourth factor Dickson cites is the increased number of water quality lawsuits and the risks associated with them. The company has been involved in over 20 such lawsuits in the past decade, in many of which parties that are potentially responsible for contamination of groundwater supplies sue water distributors such as GSWC on a variety of theories. Although Dickson asserts that GSWC has done well overall in this litigation, the lawsuits require a great deal of time from senior management and general office staff. (Id. at 16-17.)

Fifth, increased certification requirements for water system operators during the past decade have increased required training time, as well as the workload of the Human Resource Department (which must keep track of the certification process). (Id. at 17-19.) Dickson notes that the increased certification requirements have made it more difficult to attract and retain appropriately skilled employees, "especially at the most critical level of distribution system superintendent. This is an industry wide phenomenon that was not anticipated by the regulatory agencies when they adopted the [new] rules." (Id. at 19.) Dickson also notes that GSWC's Employee Development University has played a critical role in training and qualifying the company's existing employees for certification. (Id.)

The sixth factor cited by Dickson is the increased need for water company security brought about by the attacks of September 11, 2001. These include updating Emergency Preparedness and Response Plans (EPRPs) and ensuring that all affected GSWC employees participate in simulations and training related to the plans. The new requirements also require the Human Resource Department to conduct more intensive background checks, and also require company employees to be present and conduct inspections when outside vendors of chlorine and other chemicals deliver and install dispensing tanks. (Id. at 20-21.)

Seventh, Dickson points to the 2001 legislation sponsored by Senator Kuehl that requires builders to prove there will be enough water to serve their projects. The bill, which requires local agencies such as GSWC to verify that they have enough water to serve new projects of 500 or more homes for 20 years, has increased the company's general office workload in a way not suggested by the normal rate of customer growth. Also contributing to the increased workload, according to Dickson, has been the bill by Senator Costa requiring the submission of comprehensive urban water management plans every five years. (Id. at 21.)

The eighth factor cited by Dickson is the need to protect GSWC's water supply through water basin adjudication. Dickson states that GSWC's 41 water systems have approximately 300 wells that pump out of 19 separate groundwater basins. Two of the basins are managed, five have been adjudicated, and 12 basins are still non-adjudicated. Although Dickson believes GSWC's customers have been well-served by the two adjudications the company commenced, these legal proceedings typically last for several years and require an extensive investment of time by senior management, as well as follow-up by operational personnel who serve on basin management committees. (Id. at 21-22.)

Ninth, the procurement and dispatching of electric power needed for GSWC's BVEC, which serves the Big Bear vacation area, has increased the demands on GSWC's Accounting Department since the California energy crisis of 2001. Previously, all of the power for BVEC was purchased under a full requirements contract with Southern California Edison Company. (Id. at 22-23.)

Tenth, Dickson argues that compliance with the SOX has significantly increased the time demands on GSWC's senior management and general office staff. While these burdens are described in more detail in the testimony of Robert Sprowls, GSWC's Chief Financial Officer, Senior Vice President of Finance and Secretary, Dickson notes that (1) the requirement under SOX § 302 that the Chief Executive Officer (CEO) and Chief Financial Officer (CFO) certify the company's annual financial statements, (2) the requirement under SOX § 404 that management prepare an annual "internal control report" describing the internal controls for ensuring the accuracy of financial reports, and (3) the requirement under SOX § 906 that the CEO and CFO make a quarterly certification that the financial reports comply with SEC requirements, have all required significant investments of time by senior management, as well as numerous changes in various company procedures. (Id. at 23-25.)

Finally, Dickson asserts that regulatory changes at the Commission have significantly increased the workload in GSWC's Regulatory Affairs Department. The principal cause of these changes is, of course, the Rate Case Plan adopted in D.04-06-018 and D.06-02-010. Since GSWC has three districts, the effect of the mandatory three-year GRC filing cycle set forth in D.04-06-018 is to require GSWC to file a separate rate case every year. In addition, Dickson notes, the rate case schedule set forth in D.04-06-018 and the requirement of early data responses to the Master Data Requests "front loads" the workload for the utility. GSWC also receives many more data requests now than in the past. These requirements have not only increased the workload of the Regulatory Affairs Department, but also the work of the regional offices, where much of the relevant data is located. (Id. at 25-28.)

We turn now to GSWC's and DRA's detailed justifications for their positions on the 20 non-SOX general office positions that are in dispute. Following the description of the parties' positions, we set forth our decision for each position.

6.2.2. Senior Vice President-Operations

6.2.2.1. GSWC's Position

Of the 31 pages Dickson devotes to a detailed discussion of the general office jobs in dispute, eight of them concern this position. Dickson argues in some detail that at least 10 of the major regulatory changes described above have contributed to the need for a Senior Vice President-Operations (SVP-Operations), a position that GSWC created in 2002. (Id. at 33, 40.)²⁸

Dickson takes particular issue with DRA's assertion that the position is not needed because GSWC's operations "have generally remained the same over the years." After pointing out that the increased environmental, water quality and water litigation issues described above have greatly increased the company's workload, Dickson continues that it would be impossible to do all of this work without an SVP-Operations, because "GSWC's operations are more complex than most utilities due simply to the geographical diversity and varied nature of its service areas." (Id. at 33.)

Dickson notes, for example, that the SVP-Operations has played a critical role in seven recent situations where wells had to be taken offline because they exceeded applicable MCL standards. In June 2003, for example, Goodyear Well No. 4, which serves the company's Florence-Graham system, had to be taken offline because the MCL for trichlorethylene, a carcinogenic volatile organic compound, had been exceeded. It took until January 2004 to assess the options and then file a new permit application with the California Department of Health Sciences (DHS). The permit was not granted and the well put back in service until October 2005. Dickson notes that putting this well back into service required the company to undertake all nine of the regulatory steps described in his testimony, in addition to the necessary design, engineering and construction work. (Id. at 15, 34.)²⁹

Dickson notes that another reason the new position is needed is that the SVP-Operations is frequently called upon to coordinate the work of several major company departments. In addition to coordinating the work of the Engineering, Operations and Water Quality departments in the water quality lawsuits he describes, Dickson gives the following example of the need for coordination among departments in connection with new EPA rules:

The Interim Enhanced Surface Water Rule (>10,000 population) and the Long Term 2 Surface Water Treatment Rule (<10,000 population) amended the original Surface Water Treatment Rule. The Long Term 2 Enhanced Surface Water Treatment Rule builds upon earlier rules to address higher risk public water systems for protection measures beyond those required for existing regulations. The LT2ESWTR is being promulgated simultaneously with the Stage 2 Disinfection Byproducts Rule (DBPR) to address concerns about risk tradeoffs between pathogens and DBPs. Both the LT2ESWTR and the Stage 2 DBPR contain initial requirements for extensive and complicated monitoring programs before the rules are in full effect. The initial monitoring and subsequent evaluation of data will determine the full impact of the rules for each system. Both rules will require significant effort and oversight to manage. Consequently, it is very likely that many utilities - including GSWC - will be required to either build new facilities or provide significant modification to treatment facilities located at treatment plants impacted by these rules. The role of the SVP-Operations will be critical because the Engineering, Operations and Water Quality components will need to be balanced and there will be the need for completing new facilities as part of the overall Company Capital Projects program. (Id. at 36-37; emphasis supplied.)

In addition to this coordination role, Dickson points out that the SVP-Operations (1) ensures oversight and company-wide consistency in reviewing and practicing the EPRPs and other security measures, (2) provides oversight of the company's capital improvement program, which has grown from $24.4 million in 1995 to over $60 million in 2006, (3) ensures that the new certification requirements for water system operators are adequately communicated to the Human Resource and EDU departments, (4) exercises ultimate responsibility for water supply planning through the Regional Vice Presidents, who report to him, and (5) has primary responsibility for oversight of the water basin adjudication process, which in the case of the Santa Maria Basin consumed "countless hours" of the time of the SVP-Operations. (Id. at 37-39.)

Dickson is particularly critical of DRA for failing to recognize the role of the SVP-Operations in SOX compliance. On this issue, Dickson states:

DRA claims that because there were no `material weaknesses' in GSWC's internal controls, the position of SVP-Operations is not needed . . . What DRA fails to recognize is that the SVP-Operations was in fact in place and part of the process that lead to the findings of `no material weakness' in the final audit reports. The SVP-Operations position provides a critical review point and control structure for both the regional financial accounting and capital projects accounting processes. (Id. at 40.)

Elaborating on this, Dickson notes that all of GSWC's capital spending "occur[s] not in the Accounting Department but in Operations," and that "the SVP-Operations has to ensure that all controls are followed and sign off that such is the case on a quarterly basis." The company's large capital budget requires the high-level oversight provided by the SVP-Operations, Dickson continues, because "capital construction at the current levels makes it one of the Company's most significant risk factors." (Id.)³⁰

6.2.2.2. DRA's Position

The basis for DRA's opposition to the SVP-Operations position is that the new job would duplicate functions that are already being performed adequately within the company. DRA's testimony states:

GSWC argues that the current complexity in Water Quality Compliance, Water Quality Litigation, Infrastructure Replacement & Investment, Water Supply Needs, and Sarbanes-Oxley Act, warrant this new position. Furthermore before the creation of this position in 2002, the GSWC service area regions were managed by the Vice President- Customer Service. Now, the GSWC's operations are spread among three regions, each serving between 55,000 to 100,000 customers and each having a regional vice president who report to the Senior Vice President-Operations.

DRA does not find the justifications for the position compelling. First, GSWC's operations have generally remained the same over the years. The so-called `Water Quality Compliance' functions are nothing new for a water utility operating in California. GSWC already has a Water Quality Department and a Regulatory Compliance Department, each of which is adequately staffed and has its own vice president. These facts militate against the need to add yet another management layer in the GSWC's organizational structure. (Ex. 23, p. 2-3.)

After describing the water quality staffs that already exist in GSWC's general office and regional staffs, as well as the "elaborate engineering staff" found within each region, DRA concludes:

By requesting [the SVP-Operations] position, GSWC in effect is implementing a `centralized' approach to its operations. However, GSCW does not show any savings that should result from this centralized structure. In fact, the ratepayers will be burden[ed] with both the decentralized and centralized structure working at the same time. (Id. at 2-4.)

6.2.2.3. Discussion

Although Dickson's testimony does not answer all of the questions one might have about this position,³¹ we conclude that on balance, GSWC has made an adequate showing that the growth in the general office's workload makes the SVP-Operations position necessary and appropriate to include in rates. We do not agree with DRA the job is unnecessary because "GSWC's operations have generally remained the same over the years." (Ex. 23, p. 2-3.)

We begin by pointing out, of course, that the company has had an SVP-Operations since 2002, and it is only because the job was created after the filing of GSWC's last general office rate case, A.02-11-007, that we have not been asked previously to authorize this job.

Dickson's testimony makes a strong case that the position is needed to coordinate GSWC's far-flung operations and provide oversight of its ambitious capital construction program. The energetic debate described in D.00-06-075 and D.04-03-039 about whether region-wide rates should be authorized for GSWC's Region III is strong evidence that the geographically spread-out operations of the company, especially in Region III, present special management challenges. (See D.00-06-075, mimeo. at 23-30; D.04-03-039, mimeo. at 22-25.)

It is clear from GSWC's testimony that the size and scope of its capital construction program has grown so substantially in the past decade that senior management oversight is needed. As Dickson notes, the capital projects budget grew from $24.4 million to over $60 million between 1996 and 2006, and the number of projects during this period increased from 164 to about 276. Although GSWC is also seeking authority for a Capital Projects Manager in this GRC, the need for senior management oversight of the capital program seems obvious.

Dickson also makes a persuasive case that coordination from a senior executive will be necessary to ensure that the new water quality rules he cites are properly implemented, and to ensure that the new treatment facilities (or modifications to existing facilities) needed to comply with them are constructed on a timely basis and at reasonable cost. As noted in Dickson's rebuttal testimony, these rules include the Interim Enhanced Surface Water Rule, the Long Term 2 Surface Water Treatment Rule (LT2ESWTR), and the Stage 2 Disinfection Byproducts Rule (DBPR). (Ex. 11, pp. 36-37.) The need for correct and coordinated implementation of the LT2ESWTR and the Stage 2 DBPR seems especially great, because - as Dickson notes - these rules are being promulgated simultaneously "to address concerns about risk tradeoffs between pathogens and [disinfection byproducts.]" (Id.)

Although Dickson's testimony makes a less compelling case that the SVP-Operations is needed to ensure company-wide consistency in practicing security measures and complying with new operator certification requirements, he is persuasive when he argues that the SVP-Operations is needed to oversee the company's water supply planning, supervise water basin adjudications, and help ensure that SOX requirements are met at the operational level by providing a "critical review point and control structure" for regional financial accounting and capital projects accounting.

We will authorize the position of SVP-Operations to be included in rates, and we reject DRA's view that the position be disallowed.³²

6.2.3. Capital Projects Manager-Operations

6.2.3.1. GSWC's Position

Dickson presents three principal justifications for this $124,160 per year position. The first is that GSWC's capital budget has grown so substantially in the past decade (from $24.4 million to over $60 million) that the decentralized model of construction supervision the company previously used - which relied on GSWC's three District Engineers for oversight, scheduling and inspection of construction projects - is no longer feasible. In 1996, according to Dickson, the company undertook about 160 water main replacement and supply projects, permitting for them was relatively straight-forward, and most of the projects could be completed within a year. Today, on the other hand, GSWC must handle about 275 projects per year, many more permits are required, and it is unusual for a project to be completed within 12 months. (Ex. 11, pp. 42-44.)³³

All of this, Dickson submits, shows that GSWC's existing engineering resources are inadequate to perform the work they are being asked to undertake, which is why the company had to hire an outside firm (CH2M Hill) to provide the 30 full-time equivalent staff needed to do the work on the 2005 construction program. Hiring a full-time Capital Projects Manager is the first step in expanding GSWC's internal resources, since "the position will be tasked with completing all the other steps." (Id. at 45.)³⁴

Dickson also notes that a Capital Projects Manager will be able to provide better coordination and scheduling for all of the work being performed within the three districts. He gives the following explanation:

With the expanded capital program comes a need to refine the approach to project implementation to ensure the most cost effective methods of project delivery are utilized . . . [The increase in projects from 1996 to 2006] requires different tracking mechanisms, different resource allocation methodologies, different delivery methods and an overall different approach to successful completion . . .

Another key point directing the need to add the Capital Projects Manager is that the types of projects under construction benefit from centralized oversight. GSWC's capital program consists primarily of water main replacement, well replacement, and reservoir replacement . . . With the common nature of the type of work from Region to Region, it only makes sense, then, to look at the program on a company-wide basis. For example . . . GSWC's Region 2 and Region 3 often utilize the same contractor for pipeline installation. Without centralized oversight, each Region would issue an RFP for construction, and would get responses from the same contractor with competing time frames for construction. Further, there was no method of identifying project priorities or capitalizing on reduced contractor set-up and down time in-between jobs. With centralized oversight, GSWC is better able to manage its contractors and ensure each region the most cost-effective, timely construction of its capital projects. (Id. at 45-46.)

Dickson adds that a Capital Projects Manager will enable the company to be more nimble in moving resources around in the event delays are encountered on a particular project, and that the new manager will also be in a better position to draw upon the expertise that particular Regions have acquired on particular projects. (Id.)

6.2.3.2. DRA's Position

Aslam's testimony concerning the Capital Projects Manager's position is a really a general criticism of GSWC's alleged inefficiencies in managing construction projects:

DRA finds GSWC's argument [for centralized control] unpersuasive. Instead, GSWC's proposal reflects a level of inefficiency and lack of planning on behalf of GSWC. As mentioned earlier, GSWC decentralized its Engineering Operations throughout its three Operating Regions, which resulted in an elaborate Engineering staff within each Operating Region. For example, a typical engineering staff at one of the GSWC's regions consists of Engineering and Planning Manager, Senior Civil Engineer, Civil Engineer, Engineer, and several Engineer Technicians and CAD Operators. GSWC[`s] claim that the company's engineering staff in each of its Operating Regions has to compete for the same resources of contractors and outside consultants for their respective projects hold[s] no water. (Ex. 23, pp. 2-5 to 2-6.)

Although he gives no examples, Aslam also argues that other Class A water companies doing business in Southern California must operate with similar constraints, which demonstrates to him that "better planning and self reliance are necessary in this labor competitive environment." (Id. at 2-6.)

6.2.3.3. Discussion

We conclude in this case that GSWC has carried its burden of proof on the need for a Capital Projects Manager for Operations. While managing construction projects from within each Region may have made sense in 1996 - when engineering and permitting requirements were simpler and the company's operations had just been organized into three regions - the growth in the amount of the capital projects budget, the significant increase in the number of projects, and the increasing complexity of permitting and engineering requirements, all lend support to Dickson's argument that there is a need for a senior construction manager who can provide increased coordination in soliciting construction bids, scheduling work, and so forth. In view of the increase in the amount of project design work - a situation that required GSWC to outsource a significant amount of the engineering for its capital projects in 2005 - DRA's criticisms that the company already has an "elaborate engineering staff" within each region, and that the current situation demonstrates "inefficiency and lack of planning," are not persuasive.

6.2.4. Administrative Support Analyst - Operations

6.2.4.1. Positions of the Parties

In his rebuttal testimony, Dickson argues that this $58, 208 per year position - which would report directly to the Capital Projects Manager - is needed to manage the documentation for the company's ambitious construction program:

With the growth of the capital program comes the increased need for additional analysis and oversight of the capital construction program. As discussed above, complete new delivery methods of construction are needed to improve the efficiency and cost-effectiveness of the construction program. This position is critical in analyzing the status of construction projects and construction contracts for completeness and accuracy and in providing an overall analysis of the program. With over 150 jobs in construction at one time, it is critical to ensure every contract, invoice, change-order and other construction documentation is in order. (Ex. 11, p. 47.)

Dickson notes that the new administrative support analyst would also pull together and analyze statistics necessary to make key decisions in construction resource allocation, construction scheduling, and the status of contracts and contractors. (Id.)

In its testimony, DRA opposes authorizing this position for the same reasons it opposes the Capital Projects Manager position. (Ex. 23, pp. 2-6 to 2-7.)

6.2.4.2. Discussion

In view of the growth of GSWC's capital projects program during the past decade and the need for centralized supervision over it, it is not surprising that the company is requesting an assistant to handle documentation and statistical analysis for the Capital Projects Manager. We think that the company has made an adequate showing to support authorization of this position.

6.2.5. Assistant Application Support Analyst - Operations

6.2.5.1. Positions of the Parties

Like the Administrative Support Analyst, this $50,189 per year position would also report to the Capital Project Manager. In his rebuttal testimony, Dickson argues that this position - which the company currently outsources - is needed to make efficient use of GSWC's new Project Control System (PCS) software:

GSWC has begun to utilize a [PCS] based on PrimaveraTM in conjunction with Microsoft ProjectTM. The PCS was established to track and report on the status of capital projects. PrimaveraTM is the construction industry standard software used for this purpose, and facilitates project delivery on time and on budget. The PCS allows for tracking and reporting on metrics such as project schedules, milestones, resources, budgets versus costs, cash flow, estimated completion times, project schedule estimated at completion and project cost estimated at completion. The PCS also tracks project issues and resolution of those issues, provides project descriptions and details lessons learned from projects for use on other similar projects. The PCS is also able to track and report at a program level, allowing better management of company-wide resources. Use of this needed tool can only be successful under the direction of an individual skilled in programming and updating the software and its inputs. (Ex. 11, p. 48.)

In addition to the salary savings the company expects to realize by bringing this position in-house, Dickson notes that the position "will allow GSWC to migrate its entire capital program to the PCS platform . . . GSWC currently only has a small number of projects being tracked in the PCS, which severely limits our ability to fully benefit from the value of the PCS." (Id.)

In its testimony, DRA states that, as with the Capital Project Manager position, this job should be disallowed because the company has failed to show a need for reorganizing how it handles construction projects, and has also "failed to show any cost savings that would result from such centralization." (Ex. 23, p. 2-7.)

6.2.5.2. Discussion

It is not surprising that in order to bring about the centralized control over its construction program that the office of Capital Project Manager promises, GSWC would need new software. Moreover, even though the company is apparently spending a significant sum to outsource the programming and updating of this software, it has not yet been able to place its entire capital program on the new PCS system.

We will authorize the requested position so that the promised efficiencies can be realized, but in GSWC's next general office GRC, we will expect to see a persuasive demonstration that the promised construction efficiencies have been realized.

We would also point out that the justification provided for this position in the company's direct testimony - as well as the justification for the Administrative Support Analyst for Operations - consisted of little more than a job description of the kind that might be posted on a company bulletin board or website. (Ex. 5, Darney-Lane, pp. 15-16.) Such job descriptions are not very informative, and it is not surprising, therefore, that DRA chose to oppose the position. In future GRCs, we expect to see a fully adequate justification for this and other new positions set forth in the company's direct testimony.

6.2.6. General Clerk - Information Technology

6.2.6.1. Positions of the Parties

In his rebuttal testimony, Dickson argues that GSWC needs this $30,000 per year position because GSWC now receives payments in many more varied forms than in the past, a situation that has proven quite labor-intensive to deal with:

Over the past few years more and more customers are now paying their water bills through payment agencies, banks and financial institutions. This type of payment makes it easier for the customer, but often requires much more manual work for GSWC. For example, when payments are received through CheckFree they are entered electronically into our system. However, for most other institutions, including EPrinceton.ecom, the data arrives in a file that we cannot use electronically or on printed paper forms. We also receive multiple checks from financial institutions with an accompanying printed listing with the customer's name, account number and amount paid. All of this data must be manually entered into our system, the account numbers verified and control balanced along with other payments. There are 500-700 of these manual entries keyed in and verified each day. This requires 5-8 hours of real time, employee activity per day to complete this task. (Ex. 11, p. 49.)

DRA devotes a surprisingly large amount of discussion to its opposition to this position. After noting that the justification given for the position seems to entail more than "processing electronic bill payments from banks and internet service providers," Aslam states:

Currently a staff of 19 is employed within the GSWC's Information System Department in General Office. Five of them are General Clerks. In addition, GSWC regularly hires temporary workers as needed. GSWC did not present any analyses that explained the reasons behind the increased level of activities in [the] mail room. (Ex. 23, p. 2-7.)

Aslam goes on to suggest that customer growth in GSWC's regulated operations cannot be the reason for the new position, since there have been only about 8,000 new customers during the past five years. Instead, Aslam speculates, the new position is needed to provide service to customers of the entities served by GSWC's non-regulated affiliates, customers who total 74,270 by Aslam's count. (Id. at 2-7 to 2-8.)

6.2.6.2. Discussion

As with several of the other new positions GSWC is requesting and DRA is opposing, opposition to this job might have been avoided if GSWC had initially provided a straight-forward explanation of the need for the position along the lines set forth in Dickson's rebuttal testimony. Instead, the company's direct testimony consists of another job description that suggests the key task - data entry of payments in non-check form - only briefly. (Ex. 6, Darney-Lane, pp. 16-17.)

Even though GSWC did not do a good job of justifying this position in its direct testimony, the need for the position (given the limitations of GSWC's current computer system) seems clear. We also think that our resolution of the general office cost allocation issue elsewhere in this decision is adequate to address the cross-subsidy concerns raised by DRA, to the extent they have merit.

Although we are approving the General Clerk-Information Technology position, we emphasize that we are doing so only for this GRC cycle. It seems to us that if the new CIS/CRM system delivers all of the benefits that Ms. Andres describes in her testimony, the position will become unnecessary once the CIS/CRM system is on-line.

6.2.7. Assistant Information Technology Manager - Information Technology

6.2.7.1. Positions of the Parties

In his rebuttal testimony, Dickson argues that this $88,564 per year position is needed to ensure the security of GSWC's hardware, software, and data bases, and that "there currently isn't an individual within the Company with the expertise" to do this. (Ex. 11, pp. 48-49.) He notes that each of the 11 factors he identifies as having changed the regulatory landscape has contributed to the need for an Information Technology (IT) security officer, as has the increase in the size of GSWC's infrastructure replacement program. (Id. at 49.) Dickson also notes that SOX makes having an IT security officer essentially mandatory:

SOX has necessitated that the IT Department develop and review change control systems for all applications software and operating systems as well as Internet security throughout GSWC in order to comply with Section 404. Approximately 33% of this position's job functions are related to Sarbanes-Oxley compliance. Controls and security have become of paramount importance since the passing of the Sarbanes-Oxley legislation. Conducting reviews and daily monitoring of IT controls, and financial application security records is a time consuming function. These areas require constant monitoring and frequent review and auditing of report and system log records, which is currently putting a strain on internal personnel resources. (Id. at 50-51.)

DRA opposes this position mainly on the ground that GSWC's Information System Department already has 19 people, 14 of whom are IT-related staff. (Ex. 23, p. 2-8.) In addition, DRA notes that "GSWC obtains IT[-]related help on [a] regular basis from outside consultants and vendors," and claims to have found no evidence to support the Company's assertion that 33% of the new position's duties are related to SOX compliance. (Id. at 2-9.)

6.2.7.2. Discussion

It is virtually common knowledge that security concerns in IT Departments the size of GSWC's are rapidly increasing, and that people with the skills necessary to deal with these issues can command a premium. It is also not surprising that GSWC would want to have the necessary expertise in-house, rather than having to rely on outside contractors. Thus, we find DRA's general criticisms of the rationale offered by Dickson for an Assistant IT Manager to be unpersuasive.

We also think it is not unreasonable to assume that one-third of the Assistant IT Manager's time would be devoted to dealing with SOX compliance. As Dickson states in his description of the 11 major factors that have changed the regulatory landscape, SOX § 404(a) requires companies such as GSWC to prepare an annual "internal control report" that "state[s] the responsibility of management for establishing and maintaining an adequate internal control structure and procedures for financial reporting." Further, SOX § 404(b) requires each "registered public accounting firm" that prepares or issues an audit report for a company like GSWC to "attest to, and report on, the assessment made by the management of the issuer." We do not doubt Dickson when he states that these requirements have increased the amount of time the company's senior managers must spend interfacing with its auditors. We are also persuaded when Dickson says of SOX § 404(b):

It also requires the Company to continually assess 16 mega accounting processes and document and test about 250 key controls (more than 400 key controls in 2004) to ensure compliance. This requires a continuous monitoring and updating of accounting policies and procedures. (Id. at 25.)

In short, we think that GSWC has met its burden of proving that the position of Assistant IT Manager is necessary, and we will authorize this position to be included in rates.

6.2.8. New System Administrator-Developer - Customer Service

6.2.8.1. Positions of the Parties

Dickson's arguments in favor of this $68,307 per year position complement those set forth in the testimony of Yvonne Andres, the company's principal witness on the need for a new customer service computer system. Dickson states:

The Company is in desperate need of a new CIS/CRM System. This position is needed to assist in report writing, customization and modification of programs for the new CIS/CRM System that is being requested as part of this application. This position is also in charge of documentation of change management and maintains the integrity of program code. This position will also assist in system administration and upgrade processes. Having a developer in-house will significantly decrease the programming time and cost related to hiring an outside consultant. This position will ensure consistency of implementation without having to pay for outside vendor support. (Ex. 11, p. 51.)

In its report, DRA does not question the tasks proposed for this new position, but points out that "the Commission has not yet approved and authorized the CIS/CRM System projects. This requested new position is therefore unnecessary until the CIS/CRM System project[] is authorized by the Commission." (Ex. 23, p. 2-9.)

6.2.8.2. Discussion

As noted in our discussion of Ms. Andres's testimony, we believe she has made a good case for the need for the new CIS/CRM system, and for having an in-house capability to customize and modify the software for it. However, because the company's cost estimates for the CIS/CRM system are so preliminary, we are declining at this time to authorize more than the $2,982,841 (before overheads) that the company has requested for calendar year 2006 to pay for the new system. As stated in section 5.3 of this decision, in order to recover any greater amount, GSWC will be required to use the new Tier 3 Advice Letter process under General Order 96-B, a process that requires Commission approval of the advice letter by resolution before it can take effect, and also allows affected parties such as DRA to file protests.

In light of this, there is considerable appeal to DRA's argument that the Commission should not approve the New System Administrator-Developer position at this time. However, we also recognize that in order to begin deployment of the new CIS/CRM system, the services of the Administrator-Developer are likely to be necessary. Since we expect that a substantial sum for the new CIS/CRM system will ultimately be included in rates (although perhaps not as much as GSWC is requesting in this application), we will authorize the new position.

6.2.9. Three New Customer Service Representatives

6.2.9.1. Positions of the Parties

In both its direct and rebuttal testimony, GSWC requests that it be authorized to increase the number of full-time customer service representatives (CSRs) from 21 to 24, at an annual cost of $109,047 (without overheads). GSWC argues that although the number of its retail customers has not grown a great deal in recent years, the increase is justified because (1) the average time devoted to each customer service call has increased, (2) the turnover rate among temporary CSRs (of whom the company has three) is high, and (3) it is less expensive to hire permanent CSRs rather than temporaries, due to the high training costs. (Ex. 5, Darney-Lane, pp. 21-22.)

In his report, DRA's Aslam opposes the request because he thinks the real reason GSWC is seeking more CSRs is to deal with calls from the 74,270 retail customers Aslam believes are served through contracts with ASUS, GSWC's non-regulated affiliate. Aslam states:

GSWC historically did not request new CSRs when there were no Non-regulated contracts. For example, in year 1998, GSWC had 16 CSRs that served a total of 241,491 regulated customers. This represented a ratio of one CSR to 15,093 customers. However, in that year, GSWC did not request additional CSRs in it[s] GRC application, thus implying that the ratio of 1:15,093 was working well.

In year 2002 when GSWC was serving 248,776 regulated customers, it requested 5 additional CSR positions in General Office, raising the total CSR positions to 21, which results in a ratio of one CSR to 11,846 regulated customers when at that time GSWC began serving Non-regulated customers. Therefore, applying a ratio of 1:15,093 for CSRs staffing to the present number of regulated customers, only a total of 18 CSRs would be necessary. (Ex. 23, p. 2-10.)

In his rebuttal, GSWC's Dickson argues that the three additional positions are necessary to meet GSWC's internal standards for call response time:

DRA states the Customer Service Center had 16 representatives to address the customers' needs. In reality, GSWC had had no less than 20 CSRs since 1998. The request for 24 CSRs is not due to an increase in the amount of non-regulated calls; rather it is to address the service level needs of our regulated customers and the increasing call volume.

Eighty percent (80%) of calls are to be answered in forty . . . (40) seconds or less, this is the established service level for GSWC. The industry standard is eighty percent (80%) of calls in thirty seconds or less. The Customer Service Center (CSC) requires an average of 24 representatives to support the 80/40 standard service level and scheduling needs. (Ex. 11, p. 52.)

6.2.9.2. Discussion

As noted in Section 4.3.5 of this decision, GSWC's non-regulated affiliate, ASUS, does not provide customer call service to all of the retail customers of the entities with which it has contracts. Moreover, we think the equivalent number of full retail customers that can be attributed to ASUS is about 33,370, approximately 45% of the number that Aslam assumes. Thus, we do not find Aslam's analysis of the reasons that GSWC has requested three more CSRs to be persuasive. Instead, we are persuaded by Mr. Dickson that the increase (which is really designed to bring the temporary CSR positions in-house) is needed to maintain the current standard of call response time. We also think Ms. Darney-Lane is correct in asserting that bringing these positions in-house will serve to reduce turn-over and hence training costs.

Accordingly, we will authorize GSWC to include the three new CSR positions in rates.

6.2.10. Call Center Support Analyst

6.2.10.1. Positions of the Parties

GSWC's testimony notes that this position was created in 2003, in large part to free up the time of the Customer Service Supervisor so that he or she can focus on training and coaching GSWC's 24 CSRs. In his rebuttal testimony, Dickson states:

The support position of Call Center Support Analyst allows the supervisor to focus effectively on the important tasks of coaching, developing, and training, thus improving service levels. As discussed above, GSWC has a standard of answering 80% of calls within 40 seconds, a goal much lower than the industry standard of 80% of calls within 30 seconds. GSWC has achieved this mark of 80/40 only 112 out of the past 60 months.

The full scope of responsibility for this position includes: payroll entry, attendance/punctuality tracking, scheduling, escalations [i.e., requests to speak to a CSR's supervisor], and informal PUC complaints. By providing support for these tasks, the supervisor's attention can be dedicated to the development of each CSR. This has allowed GSWC to meet its service level goal for seven months in a row starting fourth quarter 2005 into 2006. (Ex. 11, p. 53.)

DRA opposes authorization for the Call Center Support Analyst because it believes the position was "hidden" from DRA in GSWC's last general office GRC, A.02-11-007. On this issue, Aslam states:

[In A.02-11-007] GSWC did not justify the need for the position. The salary expense for the position was hidden as part of the overall labor expense. DRA protests this sort of evasiveness. GSWC must present and justify all additional expenses clearly and specifically.

The Commission's approval of an overall labor expense should not be interpreted as Commission approval for new positions, especially when the new positions are not specifically requested. This elusiveness deprives DRA of fair notice and due process and obstructs the Commission's ratemaking responsibilities. (Ex. 23, pp. 2-11 to 2-12.)

6.2.10.2. Discussion

On the question of whether a Call Center Support Analyst should be authorized, we conclude that GSWC has the better of the issue.

Although DRA vaguely suggests that the GSWC's Call Center is overstaffed, the real source of its opposition to the position seems to be the perception that is was somehow misled about the position in A.02-11-007. In his rebuttal testimony, GSWC's Dickson emphatically denies this, and insists that A.02-11-007 was handled like the company's prior rate cases:

Mr. Aslam claims the positions were hidden in the last [GRC] and that DRA had no opportunity to review them or rebut the need for them. This is not true. The DRA had every opportunity to examine all costs requested by GSWC and make recommendations. Labor costs in total were examined and cost increases and upward trends in labor expense were closely examined by DRA. The DRA staff assigned to that part of the case chose which costs to challenge and which costs not to challenge . . . There have been many GRCs filed by GSWC over the years where the DRA chose not to challenge various positions. In instances where DRA did challenge the positions[,] detailed justification was always provided upon request. (Ex. 11, p. 76.)

We find this defense of GSWC's conduct in A.02-11-007 persuasive; Aslam has not presented any evidence that in A.02-11-007, DRA asked about the Call Center Support Analyst and received an inadequate or misleading response. Moreover, GSWC has made a convincing case that the job is needed to free up the time of the Call Center Supervisor to train and coach the staff of 24 CSRs. Accordingly, we will allow this position to be included in rates.³⁵

6.2.11. Applications Support Manager - Applications Support

6.2.11.1. Positions of the Parties

In her direct testimony, Ms. Darney-Lane states that GSWC's IT department "offers efficiency primarily in the hardware side of the technology." (Ex. 5, p. 24.) The choice of software, on the other hand, has been left up to now in the hands of the company's various "functional areas":

Major application software selections and upgrading are located in [the] respective functional area[s]. For example, the customer service application software was selected and has been maintained by the Customer Service Center; operations select and maintain software such as SCADA for enhancing the data gathering and operating efficiency; [the] Accounting and Finance department provides application supports for accounting/finance, job costs and payroll/human resources related enterprise software." (Id.)

Darney-Lane continues that this new $113,883 per year position will offer the following advantages to GSWC:

Provide consistency and documentation for all application implementations and upgrades.

Direct and lead business process analysis for efficiency improvements among all GSWC's divisions.

Integrate all application systems to enhance overall performance of the system.

Ensure integrity of system transactions among all applications for internal control purposes.

Direct the development of applications to meet company-wide business process requirements and to eliminate localized developments.

Oversee and review security architecture standards, database integrity and testing procedures for new implementations. (Ex. 5, pp. 24-25.)

In his report for DRA, Aslam recommends this position be disallowed because it will result in a duplication of functions:

It is obvious that a duplication of Application Support functions exist in each major functional area. The new Application Support Manager position will not replace the existing functional area application support resources. The ratepayers will have to bear unnecessary rate burdens because of GSWC having functions duplicated at the centralized and decentralized levels. (Ex. 23, p. 2-12.)

In his rebuttal testimony, Dickson denies that the new position will result in any duplication,³⁶ and he recites verbatim the list of benefits for the position set forth in Darney-Lane's testimony. (Ex 11, p. 54.)

6.2.11.2. Discussion

Although GSWC has not made as strong a case for this new position as for several of the others involving information technology, we have decided to approve it nonetheless. It seems reasonable that GSWC may realize benefits and efficiencies from having overall direction of its choice of software applications, and - according to both Darney-Lane and Dickson - that is what this position is intended to provide.

In another portion of Dickson's rebuttal testimony - where he defends positions that GSWC contends were approved in A.02-11-007, but which DRA is now challenging on the ground that no detailed justification was provided - Dickson notes that "currently, GSWC does not have an Applications Support Manager," and that the only position currently dealing directly with this function in the company is the Senior Applications Support Analyst. (Id. at 81.) Dickson describes that person's duties as follows:

The Senior Support Applications Support Analyst is responsible for assisting in the analysis, design, development, test and/or implementation of new or revised programs in conjunction with application vendors and department users to meet and support the needs of a segment of the company. (Id.; emphasis added.)

As noted in testimony of both Dickson and Darney-Lane on this position, one of the principal functions of the new manager will be to provide consistency in software selection among GSWC's various departments and functions. We think it is likely enough that efficiencies will result from this consistency that we are willing to approve the position. However, in the company's next general office GRC, we will expect GSWC to present credible evidence that such efficiencies have, in fact, been realized.

6.2.12. Corporate Communications Manager and Communications, Media and Technical Generalist

GSWC is seeking authorization for two positions related to corporate communications. The first is a Corporate Communications Manager, who would receive a salary of $103,417 per year. The second is a Communications, Media and Technical Generalist, who would receive an annual salary of $65,000 per year.

6.2.12.1. Positions of the Parties

The descriptions offered by GSWC of the two positions are quite similar, and the company has not explained very clearly how the duties of the Manager and the Generalist would differ. It appears, however, that the Generalist's emphasis would be on communicating with customers, whereas the Manager would be more responsible for formulating strategies to communicate better with all of GSWC's constituencies, including regulatory agencies and shareholders.

In their respective prepared testimony, Darney-Lane and Dickson offer the following identical descriptions³⁷ of the Generalist's duties:

Informing customers on a regular basis about the water they consume is a very important part of earning and building a customer's trust. Educating customers on an ongoing basis about their water supply, rules, regulations, and Company operations that may affect the cost they might pay is even more important. Effectively delivering this information to the customer helps them better understand the quality and cost of providing such a service. Conversely, the Company gains a better understanding of the diversity of the customer and their values. This important exchange of information will help develop and establish which means of communication best fit our customers. (Ex. 5, Darney-Lane, p. 27; Ex. 11, p. 58.)

Darney-Lane and Dickson also both note that up to now GSWC has not had a Corporate Communications Manager, and then emphasize that the company needs one so it can explore new methods of communicating more effectively with customers and other constituencies. They suggest, for example, that bill inserts are of doubtful effectiveness because of the negative connotations inserts carry when they arrive with a bill. (Ex. 5, Darney-Lane at 39; Ex. 11 at 61.) After noting that the most appropriate forms of communication may differ depending on whether GSWC is reporting financial results or telling customers about planned repairs or water conservation, Darney-Lane and Dickson continue:

How this [varying] information is communicated is very important. The Company believes in utilizing all methods of communication that are both effective and efficient that also benefit the customers. The Company is interested in methods of communication that help build, strengthen and maintain effective communication. In the past, the Company used traditional communication sources such as individual mailers, newsprint, and spots on radio and television. The Company could not be sure if these traditional methods were effectively reaching, let alone educating or benefiting[,] our customers. Newer, more specialized methods of communication should be assessed. (Ex. 5, Darney-Lane at 40; Ex. 11 at 61-62.)

In his report for DRA, Aslam argues that neither the Generalist's nor the Manager's position is needed. With respect to the Manager position, Aslam notes that the company already has a CEO, two senior vice presidents, a chief financial officer and two vice presidents, whose total salaries approach $1.4 million.³⁸ He continues:

In addition, GSWC makes use of every possible method of communication, from simple mail inserts to hi-tech, web-based broadcasts. It is difficult to understand how despite these levels of management and communications capabilities, the GSWC is failing to communicate its objectives, goals, and visions to employees, customers and shareholders. (Ex. 23, p. 2-26.)

With respect to the Generalist, Aslam argues that the company is already so well-staffed with managers who know how to deal with customers that it does not need one:

Presently, GSWC is adequately staffed in the areas of Water Quality and Customer Service, the two areas that bear directly on both GSWC and its customers. The existing Customer Service Manager could easily perform the functions of the new position with the occasional help of GSWC's Water Quality resource [including the Water Quality Vice President.] These executives should get involved with corporate communications and conduct public outreach with their customers as a requirement [of] their job function." (Id. at 2-18.)

6.2.12.2. Discussion

Of all the new positions GSWC is seeking for its general office, it has done the worst job of justifying these two. As noted above, the direct and rebuttal testimony in favor of the two positions is essentially identical, and the company's justifications as quoted above are almost a parody of a corporate communications manual.

However, we think GSWC has made one valid point, which appears in virtually identical form in the direct and rebuttal testimony for both positions. Dickson makes that point as follows with respect to the Generalist position:

Currently, GSWC does not have an employee that is dedicated to this particular job. There are daily situations where a specialized individual in this position would benefit both the Company and the customers. In the past, GSWC has encountered high-profile media situations where an experienced employee in media communications would have greatly helped communications with both the customers and the Company. Informing customers and community leaders about water conservation, low-income programs, and the benefits of proposed capital improvements within their customer service area would be beneficial to all. (Ex. 11, p. 58.)

We agree that having a "Generalist" with broad media experience is likely to pay benefits for both GSWC and its customers, and we do not agree with DRA that adding media responsibilities to the job duties of the customer service and water quality staffs will be sufficient. Thus, we will authorize the Generalist position.

However, GSWC has clearly not met its burden of proof with respect to the proposed Corporate Communications Manager. It is hard to disagree with DRA's Aslam when he states that in view of the number of GSWC's officers and the many different modes of communications it uses, "it is difficult to understand how . . . GSWC is failing to communicate its objectives, goals, and visions to employees, customers and shareholders." (Ex. 23, p. 2-26.) We suspect the main reason GSWC is seeking the Manager position in addition to the Generalist is that it fears no one below the rank of manager will be taken seriously when he or she offers media advice.

6.2.13. DRA's Attack on the Employee Development University (EDU) and GSWC's Request for Three More EDU Positions

6.2.13.1. Background

Since the mid-1990s, GSWC has maintained an operation within its Human Resources Department that was originally known as the Employee Development Program (EDP) and is now called the EDU. The program grew out of an audit of GSWC's predecessor, SCWC, that was conducted by the Barrington-Wellesley Group, Inc. (BWG) in late 1992.³⁹ The audit identified employee training as one of the weaknesses in SCWC's management, and noted that the company had recently hired a Manager of Employee Development and Training (who reported to the Vice President - Administration) to begin rectifying the situation. (Ex. 11, pp. 63-64.)

In its first GRC following the audit, A.94-06-015, SCWC requested funding for a "comprehensive [EDP] to include personnel salaries, capital costs, and operating expenses needed for effective employee training and development." (Id. at 64.) DRA supported the request, finding that the EDP proposal - which relied on in-house training supplemented by some outside training - was a reasonable training program for the utility to undertake. (Id. at 65.)

Today, the EDU is accredited and - according to the executive summary from a 2001 follow-up to the BWG audit included in Dickson's rebuttal testimony - "is well-regarded both within the Company and within the industry." (Ex. 5 to Ex. 11, p. I-2.)

6.2.13.2. DRA's Position on the Value of the EDU

Although it has supported funding for EDU in the past, DRA is not doing so in this application. Not only is DRA opposing GSWC's request for three new EDU positions (an issue we discuss below); DRA is also urging that EDU should be "dissolved" and that two of its employees (the Dean and the Senior Employee Development Specialist) should be moved to the company's Human Resources Department. Noting that GSWC is "the only Class-A water utility in the State of California that has an in-house university," DRA witness Aslam continues:

After carefully analyzing the functionality and claimed benefits of the EDU, DRA finds the EDU in-house training functions are not a core competency of the utility. It is more economical and more efficient to leave such employee training to professional organizations whose core competency is to educate and train a workforce. (Ex. 23, p. 2-13.)

The first reason for DRA's opposition is that, according to Aslam, a cost-benefit analysis of the EDU demonstrates it is not cost-effective. Aslam summarizes his assessment of the cost-benefit analysis as follows:

In this proceeding, DRA requested GSWC to provide a cost/benefit analysis for its in-house [EDU.] The company responded with a study that considered the last ten years of EDU expenses and capital expenditures but which only showed a savings of merely $94,550 over the past ten years.

However, once DRA analyzed certain cost estimations that GSWC used, it became evident that there were no savings at all. For example, GSWC estimated that for its Customer Service Related training the cost will be $53.06 per hour, whereas DRA believes that after an adjustment of traveling cost the more appropriate cost will be $23.70 per hour. Similarly, GSWC estimated its Management Development and Safety related training costs at $124.69 and $33.78 per hour respectively. However, DRA believe[s] that by becoming a long term partner with the training provider GSWC could make use of membership discounts that would reduce the training costs to $111.88 and $24.75 respectively. These minor changes in the cost estimations resulted in an actual loss over the last ten years for GSWC's in-house EDU operations. (Id. at 2-15.)

Aslam continues that the size of the loss arising from the EDU can be calculated at nearly $4.5 million over the past decade if one takes into account "the value of other existing training programs that run parallel and in addition to" the EDU programs. (Id.) Aslam provides the following list of what he considers these other training programs:

Management Initiatives, Succession and Training Cost⁴⁰

Corporate membership for the American Water Works Association (AWWA)

Employees membership for AWWA

Corporate membership in the AWWA Research Foundation

Outside consulting

Instead of maintaining its own costly EDU, Aslam argues that GSWC should rely on the numerous training resources available through the AWWA. Aslam summarizes these resources as follows:

The AWWA, an international non-profit scientific and educational society, is the authoritative resource of training, information, and advocacy to improve the quality and supply of water in North America and beyond. The largest organization of water professionals in the world[,] the AWWA also advances public health, safety, and welfare by coordinating the efforts of the entire water community. This organization also offers a wide range of training on distribution systems, water production and treatment. (Id. at 2-16.)

After noting that GSWC spends $22,817 for its corporate membership in the AWWA and another $45,000 for its membership in the AWWA Research Foundation (which provides its members with numerous peer-reviewed papers and reports), Aslam concludes:

It is quite evident that no water utility on its own can develop the extensive water expertise that is available from AWWA. GSWC should focus its limited resources on its core competency, water production and distribution. The task of training should be left to such professional organizations as the AWWA, which can provide the needed water training more efficiently and cost effectively. (Id. at 2-17.)

6.2.13.3. GSWC's Position on the Value of the EDU

In his rebuttal testimony, Dickson strongly disagrees with Aslam's assessment and devotes 12 pages to refuting it. He asserts that Aslam has misunderstood or misstated the facts on virtually every important point concerning the EDU.

As to the argument that GSWC is somehow deficient because it is the only Class A water company in California to operate its own in-house training program, Dickson responds:

Mr. Aslam suggests that GSWC . . . do as other Class A water companies do, but has provided no analysis on what the other companies are doing. He does not show they are more cost effective or offer better training. For all he knows, they may be spending much more than GSWC and he highly inefficient in comparison. (Ex. 11, p. 67.)

Dickson also notes that GSWC decided to establish its own in-house training program rather than rely on outside vendors because in-house training could be more rapid and better-tailored to the needs of the company:

Our Company's management supported a centralized process for employee development and learning and further believed in staffing the department with skilled personnel who are certificated in water operations, customer service, instructional design, information technology and training background. In so doing, staff can design training programs that are closely linked to our business to benefit both the customers and the Company.

In general terms, the main benefit of centralized learning is the cost savings that result from standardization, central reporting and record keeping and quality control. The advantages of such a system include:

Reduction in number of individual systems required to handle corporate learning.

Immediate population of the central database with course completions and certifications.

Promotion of standardization via reduction in the number of duplicate courses.

Significant cost savings through reduction in number of administrators.

Standardization of content, certifications and competencies.

Ability to easily align employee objectives with corporate objectives.

Simplified reporting (from one system versus many).

Accuracy in reporting (from one system versus many)." (Id. at 67.)

Dickson particularly attacks Aslam's conclusion that the per-hour training costs reported by GSWC could be reduced significantly by using outside vendors. On this question, Dickson states that EDU's total training costs are comprised of four elements, of which Aslam considered only the second:

Mr. Aslam simply made bad assumptions from the data provided to him. The cost of training of $53.06, $124.69, and $33.78 per hour is the total rolled in cost of all training functions performed by EDU[,] not just the in class portion of the training. The four training functions performed by EDU are as follows:

1. Training and Development Needs Assessment: EDU works with management, strategic and business plans, employee development plans and individual intake with employees to assess business and/or career training needs. EDU also uses tools such as surveys, post training evaluations, and others to assess training needs.

2. Training and Development activities: After assessing business needs, EDU then develops (if needed) and provides training courses, workshops, and other activities to meet the business training needs.

3. Application of Training (follow up): Following training events, EDU staff work with employees to help ensure that the training is being applied in the workplace and that the business needs identified during the assessment stage are being satisfied. EDU staff work with employees and travel to the various work locations in the Company, meet with employees and supervisors and provide individualized coaching where needed.

4. Evaluation of Training: EDU performs evaluation through all stages of the training process to focus on what is working and what is effective in meeting the business training needs of the Company. Evaluation is performed on various levels including:

· Employee reaction - how did they feel about the training [based upon an evaluation form]? . . .

· Learning - did they get it? Measured by pre and post training assessments/exams . . .

· Application - can they do it, use it in their work? . . .

· Business Results - was it worth the effort? - look at cost savings, improved performance in the workplace, etc." (Id. at 71-72.)

After faulting Aslam for considering only the second of the four cost elements in EDU training, Dickson also points out that very few outside vendors offer more than one or two courses that would meet GSWC's needs, and that the cost of membership in organizations offering such courses would often exceed any savings that might be realized. In addition, Dickson asserts that greater use of outside vendors would significantly increase travel costs. (Id. at 73.)

Dickson then takes aim at Aslam's assertion that when other training programs "that run parallel and in addition to in-house EDU training" are taken into account, the loss during the past decade from maintaining the EDU program approaches $4.5 million. On this issue, Dickson states:

This [assertion] reflects a great deal of misunderstanding by Mr. Aslam. The DRA [claims] that there were duplications of training costs due to requesting data in multiple formats. DRA incorrectly assumed that the $4,481,456 for the items below was a duplicate of EDU training costs:

Management Initiatives, Succession and Training Cost - $3,187,356; DRA incorrectly assumed that this training is a duplication of other programs offered by EDU. This cost is for strategic management consulting and assessments provided directly to the executive leadership of the organization.

Corporate Memberships for AWWA $149,248

Employees Memberships for AWWA $88,920

Corporate Membership in the AWWA Research Foundation $450,000

Outside Consulting $49,230

Not only is [it] inappropriate and incorrect to apply these costs to the EDU costs over ten years - these costs would be incurred regardless of having EDU and it is significantly misleading to characterize them as training costs in a comparison to the training costs of outside programs.

All the costs included in the $4,481,456 figure above are outside of the function and services provided by EDU over the ten year historical experience of EDU with the exception of approximately $121,000. (Id. at 74.)

Finally, Dickson takes issue with Aslam's assertion that GSWC incurs an average annual expense of $318,723 under its Management Initiatives, Succession and Training (MIS&T) programs, and that these programs duplicate similar courses offered by the EDU:

This [assertion] is untrue and a mischaracterization of the information provided to Mr. Aslam. EDU does succession training programs for rank and file employees in order to have qualified entry level supervision within the Company. The Company conducts separate succession planning and training for executive management independent of EDU's activities; customers are not charged twice for the same function. (Id. at 75.)

6.2.13.4. Discussion of the Value of the EDU

Although supplementary testimony by DRA might have cleared up some of its differences with GSWC, we are sufficiently convinced of the value of the EDU - and of its cost-effectiveness - that we will reject Mr. Aslam's proposal to dissolve the program as it currently exists and transfer EDU's Dean and Senior Employee Development Specialist to GSWC's Human Resources Department.

Based on the record before us, we agree with Dickson that DRA has failed to show EDU is not cost-effective.⁴¹ Aslam concedes that using the data provided by GSWC for EDU's expenses and capital expenditures over the past 10 years, there was a savings during that period of $94,550 over the costs of outside training. (Ex. 23, p. 2-15.) The basis for Aslam's claim of a $4.5 million loss over the 10-year period is his contention that GSWC is not making efficient use of the money it spends for outside consulting and various memberships in AWWA, which - according to Aslam - offers training that frequently duplicates what is available through the EDU. (Id. at 2-15 to 2-16.)

Dickson has convincingly argued that the $4.5 million figure is incorrect, and that Aslam has mixed apples and oranges in order to reach it. First, Dickson is credible when he argues that the MIS&T costs of approximately $319,000 per year that GSWC incurs to train its senior management are separate and distinct from the Management Development costs EDU incurs to train qualified entry-level supervisors for the company. (Ex. 11, p. 75.) These MIS&T costs make up over 70% of the alleged $4.5 million loss. (Ex. 23, p. 2-15.)

Second, Dickson is credible when he states that because of GSWC's size and geographic diversity, it would have to incur substantial AWWA costs whether the EDU program existed or not.

Third, we find credible Dickson's argument that Aslam has taken into account only the dollar costs of developing and presenting EDU classes, and has left out of his analysis the presumably significant costs that must be incurred in making needs assessments, following up to be sure that training is properly applied, and evaluating whether particular training is effective. (Ex. 11, pp. 71-72.) Aslam does not deny that these other steps are part of an adequate training program, yet his analysis apparently made no attempt to quantify them.

Fourth, we are persuaded by Dickson's argument that turning to outside vendors for GSWC's training needs would not be a very cost-effective option. As Dickson points out, while certain vendors offer some of the courses the company needs, there is no single vendor (or even, apparently, a small group of vendors) which offers all of them. Moreover, if more outside vendors were used for training, it does seem likely that travel costs would increase significantly.

Finally, we agree with GSWC that it is doubtful any outside vendor would be flexible enough to deal with some of the special personnel issues the company faces. Dickson provided one example of such an issue when, in referring to the non-quantifiable benefits of having an EDU, he described the program's special efforts to help a large number of GSWC operators obtain up-to-date certification:

EDU provided a narrative explanation of cost avoidance by its efforts in helping 68 . . . GSWC operators pass the required DHS exams to obtain certification required to keep their jobs and keep the Company from having to replace a large portion of its work force by importing certified personnel from elsewhere . . . More specifically, much of the Company's service area in Region 2 (with the highest concentration in customers throughout the entire Company) is located in economically disadvantaged inner city neighborhoods. It is the Company's policy to hire from the residents in these areas if at all possible. This policy is consistent with and complimentary to the various diversity initiatives championed by the CPUC. We have found that many . . . employees hired from these areas are readily trainable in the physical water system operations needed to professionally operate a potable water system. However, we also find that many of them are deficient in certain academic skills such as technical writing and basic mathematics needed to pass classroom testing required by the certification requirements. It is difficult for GSWC to find outside training in these basic academic areas. EDU has developed in house programs to meet these basic needs [that are] unique . . . to GSWC. (Ex.11 at 70; emphasis supplied.)

In a June 2006 letter to Dickson from the lead consultant on the audit that BWG conducted in 1992, the consultant wrote:

Achieving a culture shift, which we strongly believed [in the audit] was essential for the future of the Company, required a strong approach. We believed that, at least for a period of time, in-house training was an essential competency necessary within the Company. (Ex. 5 to Ex. 11, p. 2.)

It seems clear from the longevity and growth of the EDU that GSWC continues to find benefits from having a significant in-house training program. DRA has failed to demonstrate that the program is not cost-effective, and it seems clear from the discussion above that the program pays benefits that are significant but sometimes difficult to quantify. Accordingly, we reject DRA's recommendation to dissolve the EDU and transfer its Dean and Senior Employee Development Specialist to the company's Human Resources Department.

6.2.14. New EDU Positions Requested by GSWC

As noted above, DRA's broad attack on justification for having an EDU was originally occasioned by the company's request for three new EDU positions. Dickson does not discuss these positions in his rebuttal testimony, but Darney-Lane set forth a description of the positions and the rationale for them in her direct testimony.

The first position is EDU Facilitator-Instructor, at an annual salary of $80,001. After noting that few young people at job fairs express any interest in going into utility work, Darney-Lane states:

[T]o attract new workers to water utility operations in order to meet the looming crisis of the water operators worker shortage, it is incumbent upon water operations management personnel to come up with alternatives. [The EDU] is taking a proactive approach by adding to the current EDU team a technical instructor and administrator position equipped with skills in engineering, management, teaching, curriculum design and development, water and waste water, environmental, health and safety. This position will focus on continuing education and training in the technical areas of water operations and management to prepare next-generation upgrades of skilled water operations personnel. (Ex. 5, p. 28.)

The second new EDU position is that of Support Analyst, at an annual salary of $54,241. Darney Lane argues that this position is needed to keep up with the significantly-increased workload of maintaining EDU's data base:

Since the [EDU's] inception in 1992, the department has expanded its administrative activities, to include, managing a comprehensive database with employee information for mandated safety, annual training activities, tuition reimbursement program, outside vendor training, and most recently, employees' operations certification records for the California [DHS]. In addition, the database is expanded to include training information for [the] Sarbanes-Oxley Act, Security, and Standard Emergency Management Systems. Currently, one staff is assigned 25-percent time to manage the database[,] and the rest of this individual's time is dedicated to other training activities. With the increased work-load on the database activities for this staff, much time is spent to maintain the integrity of employees' records[,] resulting in other work duties being deferred to other staff members, thereby stretching staff resources. The person in the Support Analyst position will help to address the workload and allow the rest of the EDU team to spend more time training and coaching frontline employees. Approximately 50% of this position's job function is related to compliance with Sarbanes-Oxley requirements. (Id. at 25.)

The third EDU position is that of Senior Employee Development Specialist. This is already a half-time position, with the other half of the employee's time being devoted to duties in GSWC's Region I. Darney-Lane presents the following justification for making this job a full-time EDU position:

The current responsibilities of [the] American Council on Education (ACE) and the International Association for Continuing Education and Training (IACET) are shared by two staff members who also have responsibilities for mandated safety, emergency management and security, Sarbanes-Oxley, customer service, personal computer, diversity and other compliance training. The expansion of this position from 50-percent to 100-percent will ease the extra hours of current staff and allow one full-time staff to organize, plan and administrate all IACET and ACE records and launch more training authorized through these agencies. All of this position's job functions relate to compliance with the Sarbanes-Oxley act. This will benefit our water operations employees since the [DHS] approved IACET courses for continuing education credits. Also, the fulltime position will allow the staff to allocate more time to develop our current water operators with the looming operators' shortage in the industry. (Id. at 26.)

In his report for DRA, Aslam does not directly take issue with the justifications offered for these positions. He simply notes that if DRA's recommendation to dissolve the EDU is accepted, the requests for a Facilitator-Instructor and Support Analyst will be rendered moot, and that the Senior Employee Development Specialist should be accounted for in Region I headquarters expenses. (Ex. 23 at 2-17 to 2-18.)

6.2.14.1. Discussion of the Requested New EDU Positions

Although we think GSWC has presented an adequate justification for the Facilitator-Instructor job and the most of the duties of the Support Analyst, we do not think the company has demonstrated how its need to comply with SOX can justify half of the Support Analyst position and all of the Senior Employee Development Specialist's position, as Darney-Lane contends. Although the changes to accounting practices that GSWC has had to make as a result of SOX presumably include some tightening up of how it reports dealings with outside vendors and other training expenses, the description in Darney-Lane's testimony of how these jobs are tied to SOX requirements is too vague to be persuasive.

Accordingly, even though the Senior Employee Development Specialist position already exists within the company, we will not authorize GSWC to recover half of that job's costs as a general office expense. Instead, the company should present a full justification for its contention that this job is needed due to SOX requirements in GSWC's next general office GRC.

6.2.15. Associate Rate Analyst

In her direct testimony, Darney-Lane notes that due to a retirement and other changes in 2004, GSWC reassessed the needs of and reorganized its Regulatory Affairs Department. Whereas the department had previously had a vice president, three managers and five regulatory analysts, after the 2004 reorganization the department has a vice president, one manager, a senior regulatory supervisor, a senior regulatory specialist, and two associate regulatory analysts (the last being an entry-level position). In the reorganization, duties performed by some of the managers were reassigned to more junior employees. (Ex. 5, Darney-Lane, pp. 36-38.)

Darney-Lane notes that while the 2004 reorganization increased the size of the Regulatory Affairs Department by one, "the overall labor expense of the new organization is less than the expense of the currently approved organization." (Id. at 38.)

6.2.15.1. Positions of the Parties

The position for which approval is sought here is one of the two associate regulatory analyst positions. Darney-Lane argues that this position is needed principally because of the new filing requirements under the Rate Case Plan:

It was felt that the department was short staffed at the analyst level. Under the new rate case plan, GSWC is required to file a GRC application every January. In addition, under the new rate case plan, water utilities are required to prepare their rate case filings in a shorter period of time and are also required to provide much more information at the time of the filing than was previously required. An additional analyst is required by the Company to work on the rate case team. (Id. at 37-38.)

In his report for DRA, Aslam opposes this position on the ground that it cannot be justified merely because the 2004 reorganization slightly decreased the payroll for the Regulatory Affairs Department. Noting that companies like GSWC typically hire outside consultants to help them prepare rate case filings, and that many people within the company besides those in the Regulatory Affairs Department help to prepare testimony, Aslam argues that GSWC's "current Regulatory Affairs Department is adequately staffed to handle a typical GRC and other regulatory workloads."

In his rebuttal testimony, Dickson points out that the Regulatory Affairs Department today is no larger than it was in 1996, although the workload certainly is. (Ex. 11, p. 60.)

6.2.15.2. Discussion

We will approve the regulatory analyst position that GSWC has requested. Although we recently amended the Rate Case Plan in D.07-05-062 to eliminate the annual GRC filings described by Darney-Lane, there is no doubt that with the increased requirements for information brought about by the Master Data Requests, future rate cases will be more labor intensive for GSWC and other Class A water companies than in the recent past. We also think it is clear from the deficiencies of proof in the company's showing in this rate case that GSWC needs to strengthen its Regulatory Affairs Department overall.

6.2.16. EPRP Coordinator

6.2.16.1. Positions of the Parties

GSWC argues that the new position of EPRP Coordinator at an annual salary of $79,986 is necessary because of the passage of the Public Health and Bioterrorism Response (PHBR) Act signed by the President in 2002. In her direct testimony, Darney-Lane states:

The requirement for maintaining current Vulnerability Assessments from USEPA and the ongoing requirements to maintain Emergency Response Plans . . . make it critical that this position be maintained. As a Utility serving over one million people in over 40 separate water systems, each of which requires a separate plan, the task of maintaining these plans is an enormous responsibility. With the increasing requirements by state and federal agencies for water utilities to plan and prepare to respond to various emergencies, including natural disasters and as well as potential security or terrorist events in the United States, the Company has identified the need for a position to ensure we are able to do the following minimum activities:

Planning development and coordination of table top exercises as part of the Company's ongoing Emergency Preparedness and Response Plan.

Planning, development, and coordination for the implementation of water system security programs and related initiatives, in accordance with federal security requirements.

Preparation and maintenance of a comprehensive database on all existing and proposed Federal and State Security Laws, Programs and initiatives that could impact ASWC.

Working with both state and local agencies to ensure the Company's EPRP is in conformance with state standards." (Ex. 5, Darney-Lane, pp. 38-39.)

In his report for DRA, Aslam opposes the EPRP Coordinator position because he believes the job can be performed by the company's existing personnel. Aslam argues that "given the fact that GSWC already has completed the initial vulnerability assessment, the existing Safety Specialist with the help from the Regional Managers, who have first hand knowledge of their respective water systems, can perform the requirements imposed by the [PHBR] Act." (Ex. 23, pp. 2-24 to 2-25.)

6.2.16.2. Discussion

We will approve this new position. We find plausible Dickson's point that DRA's interpretation of the PHBR Act "does not take into account the need to ensure that the Emergency Response Plans . . . remain updated," which includes the need "to provide routine table top training sessions." (Ex. 11, pp. 59-60.) We are also persuaded by Dickson's points that (1) the Safety Specialist and Regional Managers on whom Aslam would rely "do not possess the time, expertise or capacity" to keep the EPRPs up-to-date, and (2) adopting DRA's position would "place[] the Company at regulatory risk from both the USEPA and DHS." (Id.)

6.3. A $50,000 Penalty is Appropriate for GSWC's Failure to Disclose Until Rebuttal Testimony the Rationale for Requesting at Least Half of the20 New General Office Positions

As noted in the introduction to this section of the decision, a significant controversy developed between GSWC and DRA on the propriety of the scope of the company's rebuttal testimony. DRA claimed it was "sandbagged" by GSWC's decision to serve the Dickson and Sprowls rebuttal testimony - which together totaled over 200 pages - barely two weeks before hearings were scheduled to begin on June 26, 2006. DRA also claimed it was prejudiced when GSWC served a very large volume of responses to DRA's data requests concerning the rebuttal testimony on June 24, just two days before hearings were to begin.

On June 28, 2006, DRA put these objections into concrete form and filed a motion to strike all of Dickson's rebuttal testimony and portions of the Sprowls rebuttal testimony. In addition to its claims of unfairness, DRA relied on language in D.04-03-039, where the Commission declined to include a $5.4 million software expenditure in the rates for SCWC, GSWC's predecessor, because the company had not provided the basic justification for the expenditure in its direct testimony, and even after DRA noted the deficiencies in SCWC's direct testimony, the company failed to justify the expenditure.⁴² In its motion to strike, DRA notes that because it did not receive GSWC's rebuttal testimony until June 9, 2006, DRA was forced to scramble to prepare for cross-examination. On June 16, DRA propounded approximately 92 data requests in connection with the rebuttal testimony. On June 23, GSWC provided partial responses by e-mail. The bulk of the responses came on Saturday, June 24, just two days before hearings were to begin, when GSWC delivered to Aslam's home over 1,000 pages and four CD-ROMs filled with data responses. (DRA Motion to Strike, p. 2.) Had GSWC met its burden by including the justifications for the 20 new positions in its direct testimony, DRA would have had ample time to review GSWC's data responses in advance of the hearing on these issues.

GSWC filed a response to DRA's motion to strike at the end of the day on July 5, 2006. In its response, the company argued that the motion to strike should be denied because all of the Dickson and Sprowls testimony was proper rebuttal, in that it responded to contentions made by DRA's Aslam.

Normally, we would have tried to deal with GSWC's conduct by allowing DRA to submit surrebuttal testimony, if requested. However, the fact that GSWC's data responses were received only two days before hearings were scheduled to begin made surrebuttal testimony an impracticable option for DRA and the assigned ALJ, since both needed to prepare for the hearings, and since the submission of surrebuttal testimony would have significantly delayed the hearings beyond the starting date the parties had agreed upon at the May 2, 2006 PHC. The harm caused by GSWC was partially mitigated by deferring the cross examination of the rebuttal witnesses.

On July 7, 2006, assigned ALJ McKenzie issued his ruling on DRA's motion to strike. Although finding that GSWC's conduct had "not been exemplary," the ALJ also concluded that the company's actions did not appear to justify striking the large volume of rebuttal testimony that DRA had challenged. (ALJ Ruling Denying DRA Motion to Strike, mimeo. at 4.) First, the ALJ noted that while the justifications for the new general office positions set forth in Darney-Lane's direct testimony were "thin" in comparison with those offered by Dickson and Sprowls, they were "nonetheless sufficient to apprise Mr. Aslam of the basis for the company's request." (Id.) Second, the ALJ observed that the rebuttal testimony had to be prepared very hastily, which likely contributed to its length. (Id.) Third, since Aslam acknowledged some familiarity with SOX, the ALJ concluded that DRA's claims of prejudice in connection with the Sprowls rebuttal testimony - which dealt largely with SOX and its implications for GSWC - were exaggerated. (Id. at 5.) Finally, the ALJ noted that the cross-examination of Dickson and Sprowls had been deferred for several days, which appeared to lessen the prejudice to DRA. (Id.)

In view of all these factors, the ALJ concluded that he should deny the motion to strike and follow the Commission's "preferred practice" of "admit[ting] the testimony into the record, but then . . . afford[ing] it only so much weight as the presiding officer considers appropriate." (Id. at 2.) While we do not quarrel with the decision of the ALJ not to strike the rebuttal testimony of Dickson and Sprowls, a decision the ALJ had to make quickly and appears to have reached with some reluctance, we find it necessary to examine whether we should take further action against GSWC for its conduct on this issue.

In reviewing the record, we find several facts concerning GSWC's conduct troubling. For example, with hearings set to begin on a wide range of issues in less than 48 hours, no attorney or staff member could reasonably be expected to digest the volume of data responses described in DRA's motion, or to work what was learned from these responses into the proposed cross-examination with any comfort. GSWC's conduct is particularly problematic, because at the very least, D.04-03-039 put it on notice that this type of deficiency is unacceptable. We warned SCWC in D.04-03-039 that it had the burden of justifying its case in direct testimony:

Providing the basic justification in rebuttal is unfair, since parties are not generally given the opportunity to respond to rebuttal with testimony of their own . . . When the utility has the evidentiary burden, we caution against the use of rebuttal testimony to provide the basic justification. As a matter of fairness, we must seriously consider either striking such testimony or extending the proceeding, at the utility's risk, to allow for responsive testimony from the other parties." (D.04-03-039, mimeo. at 84-85; footnote omitted.)

Moreover, GSWC should have been aware of the Commission's concern about DRA's claim that the company's ratepayers were being asked to subsidize the activities of ASUS, GSWC's non-regulated affiliate. (D.04-03-039, mimeo. at 26-30.) In view of this interest, statements such as Mr. Dickson's that GSWC was seeking to include in rates only 69% of the salary for the proposed new Call Center Support Analyst, and that the rest of this position's salary was allocated to "new business" for ASUS, should have been given more prominence in the company's direct testimony.⁴³ (Ex. 11, p. 53.) If it had, DRA might have been able to conduct discovery that would have enabled it to arrive at a better method for allocating call center costs between GSWC and ASUS. (See Ex. 23, pp. 4-2 to 4-3; DRA Opening Brief, pp. 10-11, 13; Ex. 45.)

GSWC's conduct is revealing when one bears in mind that in a general rate case, the burden of proof is on the utility to justify any rate increase. As this Commission stated in denying rate relief less than a decade ago to SCWC for the costs of participating in a project:

A fundamental principle involving public utilities and their regulation by governmental authority is that the burden rests heavily upon a utility to prove that it is entitled to rate relief and not upon the Commission, the Commission staff, or any interested party, or protestant to prove the contrary." (D.99-04-060, 86 CPUC2d 54, 62, quoting Suburban Water Co., 60 CPUC 183, 200 (1962) (; emphasis added).)

Other decisions from this Commission and the Federal Energy Regulatory Commission (FERC) recognize that a corollary of this rule is that a party must place the full justification for a proposal in its written direct testimony, and may not wait until rebuttal to do so.⁴⁴ For example, in a ruling striking rebuttal testimony on vertical competition issues offered by Southern California Edison Company (Edison) in the FERC proceeding that considered the proposed merger between Edison and San Diego Gas & Electric Company, the FERC ALJ said:

"[A]pplicants are required to present all the proof which they intend to offer in support of the issues on which they have the burden of proof and the initial burden of going forward . . . Applicants are not at liberty to hold back affirmative proof at this stage in order to introduce it at a later stage of the trial, and the applicant indulging in such practice must suffer the consequence of this action." (Southern California Edison Company and San Diego Gas and Electric Company, FERC Docket No. EC89-5-000, 50 FERC ¶ 63,012, p. 65,065.)⁴⁵

In his rebuttal testimony here, Mr. Dickson suggests that he views the well-established rule that a utility seeking a rate increase bears the burden of proof as a change in how the Commission has traditionally conducted water company GRCs. Dickson notes that as recently as 2002, when GSWC filed its last general office rate case, the company did not offer a detailed breakdown of the new positions it sought. Instead, "labor costs in total were examined and cost increases and upward trends in labor expense were closely examined by DRA." (Ex. 11, p. 76.) Dickson also remarks that "there have been many GRCs filed by GSWC over the years where the DRA chose not to challenge various positions," but that when DRA did challenge the need for a particular position, "detailed justification was always provided upon request." (Id.)

We find GSWC's explanation unpersuasive and unacceptable, particularly in view of the warning GSWC received against such conduct in D.04-03-039. We think it is clear that the manner in which GSWC presented its justification for the new positions here, by withholding much of the detailed rationale for them until rebuttal testimony, unfairly handicapped DRA in the preparation of its report and in its cross-examination of GSWC's witnesses. GSWC's repeated act of providing the principal justification for new general office positions in rebuttal testimony should be addressed beyond giving a stern warning and lecture. Accordingly, because of the prejudice to DRA (and hence to GSWC's ratepayers), our duty to protect our regulatory process, and the need to deter such conduct by GSWC and other utilities in the future, we intend to impose a penalty on GSWC for this conduct.

Our authority to levy a fine against GSWC for its conduct in the proceeding stems from Public Utilities Code Section 2107 (Section 2107):

Any public utility which violates or fails to comply with any provision of the Constitution of this state or of this part, or which fails or neglects to comply with any part or provision of any order, decision, decree, rule, direction, demand, or requirement of the commission, in a case in which a penalty has not otherwise been provided, is subject to a penalty of not less than five hundred dollars ($500), nor more than twenty thousand dollars ($20,000) for each offense. (Pub. Util. Code, § 2107.)

Under Public Utilities Code Section 2108, each date on which a continuing violation remains in effect constitutes a separate violation.

We believe that GSWC should be fined for the previously described violations pursuant to our authority under section 2107 because any violation of statutes, Commission decisions, and directives, regardless of the circumstances, is a serious offense that should be subject to fines. Furthermore, as the Commission has previously recognized, "[t]he primary purpose of imposing fines is to prevent future violations by the wrongdoer and to deter others from engaging in similar violations. (D.01-08-058, mimeo. at 80, and D.04-09-062, mimeo. at 62.) We find that GSWC failed to disclose until rebuttal testimony its justification with respect to at least half of the general office positions at issue. Pursuant to Sections 2107 and 2108, each of these ten positions is considered a separate offense, for a total of ten offenses. Therefore, the range of the fine may be from $5,000 to $200,000.

The question we now turn to is what is the appropriate penalty in this case? The Commission's general criteria for determining the amount of a fine are set forth in D.98-12-075. (84 CPUC2d 155, 188-90.) As stated in that decision, in cases where there has been no physical harm to the public, the relevant criteria in determining the appropriate amount of a fine are as follows:

- Economic harm: The severity of a violation increases with (i) the level of costs imposed on the victims of the violation, and (ii) the unlawful benefits gained by the public utility. Generally, the greater of these two amounts will be used in setting the fine. The fact that economic harm may be hard to quantify does not diminish the severity of the offense or the need for sanctions.

- Harm to the Regulatory Process: A high level of severity will be accorded to violations of statutes or Commission directives.

- Number and Scope of Violations: A single violation is less severe than multiple offenses. A violation that affects many consumers is worse than one that is limited in scope.

- Utility's Actions to Prevent a Violation: Utilities are expected to take reasonable steps to comply with applicable laws and regulations. The utility's past record of compliance may be considered in assessing a penalty.

- Utility's Actions to Detect a Violation: Utilities are expected to diligently monitor their activities. Deliberate, as opposed to inadvertent wrongdoing, is an aggravating factor.

- Utility's Actions to Disclose and Rectify a Violation: Steps taken by a utility to promptly report and correct violations may be considered in assessing a penalty.

- Need for Deterrence: Fines should be set at a level that deters future violations. Effective deterrence requires that the size of a fine reflect the financial resources of the utility.

- Degree of Wrongdoing: The Commission will review facts that tend to mitigate the degree of wrongdoing as well as facts that tend to exacerbate the wrongdoing.

- Consistency with Precedent: Any decision that levies a fine should address previous decisions that involve reasonably comparable circumstances and explain any substantial differences in outcome.

- Public Interest: In all cases, the harm will be evaluated from the perspective of the public interest.

Some of the above criteria suggest that only a modest fine is warranted. GSWC is a relatively small water company, and the amount of the fine must reflect its financial resources. Also, it is unclear whether there has been any economic harm to ratepayers because we do not know whether DRA would have succeeded in securing a different result had GSWC provided the detailed analysis on the new positions in its direct testimony. Of course, D.98-12-075 also states that the fact that economic harm may be hard to quantify does not diminish the severity of the offense or the need for sanctions.

On the other hand, several criteria weigh in favor of a larger fine. GSWC's action harmed our regulatory process and is the type of act that we would want to prevent in future rate cases. In its rebuttal, GSWC made it clear that it does not take its burden to justify its case in direct testimony seriously. In 2004, this Commission found that GSWC conducted similar conduct. Clearly, the message we sent in D.04-03-039 did not have an impact on GSWC. GSWC's failure to comply with D.04-03-039 and its repeated conduct weighs in favor of a higher fine. Moreover, GSWC has not taken any responsibility for its conduct in this case, nor did it take any steps to rectify the harm it caused.

As previously mentioned, in D.04-07-022, a decision concerning Edison's revenue requirement, the Commission declined to penalize Edison for its failure to provide justification for a non-controversial capital addition in its direct showing. DRA did not dispute the reasonableness of the capital additions, conceded that Edison's omission of its justification was inadvertent, and did not make any claim or showing that it submitted data requests on the justification for capital additions. (D.04-07-022, pp. 156-157.) We concluded, in relevant part, that "SCE obviously made a simple mistake. Its failure to include the justification with the application was not part of a litigation strategy whereby SCE would wait until rebuttal to spring this information on unsuspecting parties." (Id., p. 157.) We further declared, "[n]otwithstanding today's decision, we reserve the right to deny consideration of any "rebuttal" evidence that could have and should have been included with the utility's direct showing, even where, as here, a simple mistake of omission has been made by the utility." (Id., p. 158.)

In contrast, in this case, DRA disputed the reasonableness of the justifications of the 20 new positions set forth in GSWC's rebuttal testimony. Also, we do not find, nor does GSWC claim, that it simply made a mistake. While we declined to disallow GSWC's rebuttal testimony for the reasons previously stated, given the totality of the circumstances, including GSWC's repeated conduct, its failure to take responsibility for its actions, and GSWC's financial resources, we believe that a fine of $50,000 is appropriate. By levying a fine against GSWC, we send a strong message to GSWC and other utilities that direct testimony is the time to address and justify its case. In particular, when there is a proposed rate change, new policy proposals or ideas, business changes that could or should influence the treatment of historic data, dramatic regulatory or environmental events and/or significant additions to the employee base or the capital budget, the burden is particularly obvious. Furthermore, as general office expenses are routinely contentious in water cases, it is not unreasonable to expect utilities to be forthcoming in their justifications of these expenses. The integrity of our regulatory process is best served when a utility justifies and addresses the issues in its application in direct testimony.

Therefore, we direct Water Division to prepare an order to show cause for Commission consideration as to why GSWC should not be fined $50,000 for its conduct in this proceeding. We also direct Water Division to prosecute this Order to Show Cause. Issues considered in the Order to Show Cause shall be considered adjudicatory and thus subject to a ban on ex parte communications.

⁴ As set forth in a 1956 memorandum introduced by GSWC, the four factors are (1) direct operating expenses (excluding uncollectibles, general expenses, depreciation and taxes), (2) gross plant, (3) number of employees (using direct operating payroll and excluding general office payroll), and (4) number of customers. (Ex. 41.)

⁵ As noted in the application, Region II is made up of the Central Basin West, Central Basin East, Culver City and Southwest CSAs mentioned on Schedule D to Switzer's testimony.

⁶ For example, the factors used for the Fort Bliss, Texas contract were gross plant, expenses and the number of wells, which were assigned the following respective percentages when compared with these same factors for GSWC: 7.28%, 1.19% and 5.78%. These percentages were then summed (7.28 + 1.19 + 5.78 = 14.25) and divided by 3 to yield the overall percentage of 4.75%, which in turn was applied to that portion of the $30,924,483 in GSWC general office expenses for the 12 months ending September 30, 2005 that Aslam considered relevant.

⁷ The DRA paragraph referred to by GSWC's counsel provides in full:

As far as the Allocation Factors are concerned, DRA tried to use the basic Four Factors Allocation Factors. However, if the use of any one of the four factors was likely to skew the costs to either the Regulated or the Non-regulated entity, it was simply excluded. (Ex. 23, p. 4-8, lines 5-8.)

⁸ DRA's best attempt to outline its general approach appears in its August 13 opening comments, where it states that its approach takes into account the variation in services that ASUS provides to its contracting parties "by assigning a specific `cost allocation base' for each non-regulated contract. For example, . . . regarding the Rowland Water District contract, DRA reduced the cost allocation base for this contract from $30,924,483 to $21,287,614 by excluding costs such as `Customer Service-Day shift,' because the contract only requires `after-hour' call center activities. This adjustment eliminates the need for [the] `weighted' number of customers [provided for in the PD.]" (GSWC Opening Comments, pp. 5-6; footnote omitted.)

While an approach like this might be a reasonable alternative to the weighted-number-of-customers approach we are using in our interim cost allocation methodology, the central problem remains: DRA never provided a clear explanation in its testimony or briefs of the "cost allocation base" concept, or of how DRA chose to apply it to each particular ASUS contract.

⁹ In D.80207, the Commission said the following about the allocation of billing and record maintenance costs:

Staff Exhibit No. 19 states that the difference between applicant's original estimate and the staff's estimate of customer records and collection expense was predominantly due to differences in allocation percentages for payroll. The staff developed four-factor allocation percentages, whereas applicant's general manager testified that applicant spreads these expenses in proportion to the number of customers. For the rendering of bills and maintaining of customers' accounts there appears to be no justification for considering (1) direct operating expenses, (2) number of division employees and (3) division gross plant, the three additional factors used by staff. Applicant's allocation method more properly relates customer records and collection expense to the numbers of customer accounts and bills rendered. (73 CPUC at 603.)

In addition to D.80207, Exhibit 41 - the 1956 memo describing the four-factor methodology - notes that "indirect expenses which have a significant relationship to a particular factor, such as pension expense to payroll, should be segregated and prorated on the basis of an appropriate single factor.

¹⁰ In Switzer's direct testimony, he advocates that $14.9 million of the $30.9 million in general office costs that GSWC booked for the year ending September 30, 2005 should be allocated using individual allocation factors. Specifically, he advocates allocating $2.1 million in customer service costs based on the number of calls, $0.5 million in billing and cash processing costs based on the number of bills prepared, and $12.3 million in human resources costs based on the number of employees. (Ex. 6, Switzer, pp. 11-13.)

¹¹ In addition, the Ross-Weeks letter in the 2005 annual report indicates that ASUS has been awarded a contract to operate the wastewater system of Fort Lee, Virginia.

¹² We recognize that ASWC's 2005 annual report states that ASUS has formed several subsidiaries to operate the water and wastewater systems at the military bases for which ASUS has won contracts. We also recognize that it is possible the growth in personnel needed to operate these systems at the bases will take place through hiring by the new subsidiaries, and that the new personnel hired may not need significant assistance from GSWC.

However, the record in this case does not indicate whether ASUS's increased personnel needs are being met through hiring by the new subsidiaries, or the amount of GSWC personnel expense that is being directly charged to ASUS and its subsidiaries. Indeed, for the years 2000-2003, the relevant excerpts of the annual reports concerning affiliate transactions that D.98-06-068 required SCWC, GSWC's predecessor, to file, state only that "shared employees charge their time on affiliate projects directly. Timesheets are prepared and the payroll expenses and associated labor burden expenses are charged to the various contracts." (Ex. 57, pp. 2, 5, 8, and 10.) The amounts of these charges, even in aggregate form, are not set forth in these reports.

As we recently said with respect to GSWC's predecessor, SCWC, "the burden rests heavily upon a utility to prove that it is entitled to rate relief and not upon the Commission, the Commission staff, or any interested party, or protestant to prove the contrary." (D.99-04-060, 86 CPUC2d 54, 62.) Here, the burden is on GSWC to show that it is, in fact, properly charging ASUS and GSWC's other non-regulated affiliates for employee time devoted to affiliate business. While the affiliate transaction reports for 2004 and later summarize the expenses charged directly to the affiliates, there is no way to tell for the prior years whether the company has satisfied its burden of properly charging expenses.

¹³ Another ground for ORA's opposition was that RTC's formula automatically "classifies employees with administrative and general functions as RTC employees." (Mimeo. at 45.)

¹⁴ For example, ASUS provides only after-hours call center service to the Rowland Water District. During cross-examination, Aslam acknowledged that Rowland's customers had placed only 1,800 after-hours calls during the entire 12-month period covered by the cost allocation study. Despite this limited number of calls for the single service ASUS provides, Aslam assumed a customer count of 15,000 for the Rowland Water District in his cost allocation computations. (Tr. 868-872.)

¹⁵ In CCWC's case, the affiliate is also subject to regulation by the Arizona Corporations Commission.

¹⁶ Using these military contracts also avoids the problems in determining the proper weighted number of customers for the ASUS's contracts with WellSpring International, Inc., the City of Chino Hills, and the Goleta water District, all of which expired on various dates in 2005. Rather than try to develop a weighted customer count for these contracts - none of which will be in effect during the three-year period covered by this rate case - it makes sense to use the military contracts, all of which were in effect in 2006 and are expected to be in effect for many more years. This is true even though the time period covered by GSWC's cost allocation study, which is the source of much of the data we use here, is the twelve-month period ending September 30, 2005.

¹⁷ At pages 4-7 of its August 13, 2007 opening comments, GSWC argues that the PD errs in its discussion of the interim cost allocation methodology insofar as it relates to the number of customers that should be imputed to the ASUS contracts with military bases. With respect to the contract with Fort Bliss, for example, GSWC argues that "the services in the military base are provided by ASUS's own employees, not GSWC employees," and that the Fort Bliss contract "does not require any meter reading or billing support, and does not use any GSWC employees to operate the water and wastewater systems." (GSWC Opening Comments, p. 6.) Based on these assertions, GSWC argues that the allocation factor for the contract "should be reduced from 100 percent to 17.9 percent," the percentage applied in the PD to A&G services only. (Id.)

Although the record on this issue is thin, it does appear from GSWC's cost allocation study that at least some of the work at Fort Bliss is being performed by ASUS employees, because GSWC's cost allocation study (Ex. 6, Switzer Schedule B, p. 2) states that seven of ASUS's 14 employees "are under a separate benefit program of Ft. Bliss Water Company, a subsidiary of ASUS."

However, while the work performed by these employees may not consist of all the same services that would be provided to retail water customers, the work is clearly very substantial, based on the contract that GSWC provided with its February 2006 General Office workpapers. That contract states that ASUS is acquiring the Fort Bliss water and wastewater systems pursuant to 10 U.S.C. § 2688, which empowers the Secretary of a military department to convey "a utility system, or part of a utility system" subject to the Secretary's jurisdiction, to a "municipal, private, regional, district or cooperative utility company or other entity." ASUS is acquiring the Fort Bliss water and wastewater systems over a 50-year period, and many pages of the contract are devoted to its obligations to undertake various kinds of capital improvements. The preamble to the Ft. Bliss contract (at page II) makes clear that ASUS's obligations under it are very broad:

ASUS shall assume ownership, operation and maintenance of the utility infrastructure water and wastewater distribution systems at Fort Bliss, Texas. ASUS shall furnish all necessary labor, management, supervision, permits, equipment, supplies, materials, transportation, and any other incidental services for the complete ownership, operation, maintenance, repair, upgrades, and improvements to the utility system.

In view of the fact that ASUS is acquiring the water and wastewater systems of this very large military base and has the obligation to run them, it is not unreasonable to assume for purposes of the interim cost allocation methodology that the number of connections at Fort Bliss should be treated as equivalent to retail customers. The same is true for the other ASUS contracts with military bases, which Table 1 to GSWC's opening comments (which table was inadvertently omitted from the comments and belatedly served on August 16, 2007) also asserts should have a smaller number of retail customers attributed to them than the number of connections for these contracts.

¹⁸ The PD and alternate decisions that were issued on July 24, 2007 stated that the appropriate factors for comparison were total O&M expenses, total A&G expenses, and supply expenses. Even though neither GSWC nor DRA pointed it out in their comments, when we examined the figures in Table 3 of Attachment B again, it became apparent that supply expenses (which are comprised of purchased water, purchased power and pump taxes) had been counted twice, because they appeared in both the total O&M expense and the supply expense lines. Upon finding the double-counting error, we considered the best way to deal with it. Upon further reflection, we have concluded that it makes sense to eliminate supply costs entirely, because - as noted in the text - all of the entities with which ASUS has contracts are responsible for supplying their own water. For example, paragraph C.3.5 of the ASUS contract concerning Fort Bliss makes it quite clear that the government will be supplying its own water:

Electric, natural has, and water commodity supply is not included in this contract. The Government retains the right to procure or supply electricity, and/or natural gas, and/or water, that will be transported on the system(s) covered by this contract from any source.

¹⁹ In its August 13, 2007 opening comments, GSWC argues that the PD errs in assuming that the ASUS contract with Tustin provides for some A&G services, because "as shown on page 3 of GSWC [witness] Switzer's testimony, Exhibit 6, the contract . . . is for meter reading only." (GSWC Opening Comments, p. 5; emphasis in original.)

While Switzer's testimony does indeed make this assertion, it is contradicted by the terms of the contract itself, which GSWC submitted as part of the workpapers supporting its application. The contract with Tustin (which was apparently entered into by ASWC, GSWC's parent) is included in Volume 4 of the company's February 2006 General Office workpapers. While Paragraph 4.1 of the contract relates to meter reading services, Paragraph 4.2 sets forth other customer services to be provided to the City of Tustin, including meter re-reads, turning meters on and off, hang door tags, shut-off of service to delinquent accounts, and "such other customer services duties as may be reasonably requested by City." Thus, contrary to GSWC's assertion, the PD did not err in assuming that the Tustin contract provides for some A&G services in addition to meter reading.

We also wish to point out that during the hearings in this case, the GSWC workpapers containing the ASUS contracts were not marked as exhibits. So that the record is complete concerning these contracts, we are hereby designating Volume 4 of GSWC's February 2006 General Office Workpapers as Exhibit 63 in this proceeding, and Volume 5 of the February 2006 General Office Workpapers as Exhibit 64.

Both of these volumes of workpapers were submitted under seal pursuant to General Order 66-B and Pub. Util. Code § 583. We have added an ordering a paragraph to this decision to grant GSWC's motion to place these workpapers under seal. In keeping with our usual practice, Exhibits 63 and 64 will remain under seal for a period of two years, at which time GSWC will be free to file a motion arguing that the two exhibits should continue to be kept under seal. Except where necessary, we have avoided discussing the most sensitive terms of these contracts (such as the price for services) in the text.

²⁰ In its August 13 opening comments, GSWC has also made other criticisms of the interim cost allocation methodology and urges us to make certain "corrections" that would bring the resulting allocation percentages closer to those advocated in GSWC's own study, which was sponsored by Mr. Switzer.

What GSWC's comments ignore is that the reason the PD found it necessary to develop an interim cost allocation methodology was because of the evident deficiencies in GSWC's own cost allocation study. Not only is the use of single allocation factors (which GSWC says it employed to allocate 48% of general office costs) a method the Commission has generally avoided over the years, but there can be no doubt that in view of D.03-05-078, GSWC distorted the traditional four-factor methodology when it assumed that ASUS had only 11 customers, thereby effectively assigning the customer-count factor a value of zero. While GSWC continues to argue that the use of single allocation factors is appropriate here, it cites no support for this argument beyond the handful of cases that were discussed in the PD itself, and as to D.03-05-078, GSWC's opening comments are completely silent.

While the interim methodology described in the text may not be perfect, it has been applied consistently, and it is preferable to the selective and often confusing approach used in GSWC's own study, which clearly ignored important Commission precedents.

²¹ As noted in Section 3 of this decision, ¶ 5.10 of the August 4, 2006 stipulation between GSWC and DRA sets forth an agreement between these parties to allocate certain offices' expenses to the Metropolitan CSA. Because we are concerned that these stipulated allocations may be inconsistent with the cost allocation approach we are using in this decision, we are rejecting the terms of ¶ 5.10.

²² Andres also gives examples of problems involving GSWC's efforts to comply with the Sarbanes-Oxley Act, including (1) the difficulty in assigning security access at the function level, (2) the inability of the CIS system to generate lists of all changes performed by the vendor, and (3) the inability to prevent by electronic means, dollar adjustments from being posted to customer accounts until all required approvals have been obtained. To deal with this last shortcoming, GSWC requires all dollar changes of $500 or more to be manually checked and approvals obtained. (Id. at 5-6.)

²³ In her rebuttal, Andres also addresses DRA's claim that GSWC is seeking a new CIS/CRM system partly to meet more stringent customer service performance standards contained in the contracts entered into by ASUS. Comparing the standards for the City of Torrance cited in DRA's testimony with the standards used by GSWC and within the water industry, she concludes that "the standards of the Non-regulated businesses are less stringent than the Company's standards." (Id. at 3, lines 18-19.)

²⁴ Indeed, Andres's rebuttal seems to concede that the customer needs of GSWC's affiliates have been taken into account in the design of the new system when she states that "Non-Regulated customers consist of less than 20% of the total number of customers serviced through GSWC's CIS System." (Ex. 14, p. 2, lines 7-9; emphasis added.)

²⁵ Unlike the situation here, the parties in D.05-07-022 ultimately agreed that since work in certain CWS districts was performed under contract by CWS employees for unregulated enterprises, it was appropriate for CWS's shareholders to pay not only for the expenses connected with this work, but also for a share of the common plant used by the CWS employees in their work for the unregulated companies. (Mimeo. at 18.)

²⁶ In its August 13 opening comments, DRA argues that allowing GSWC to use a Tier 3 advice letter to recover the balance of the costs for the CIS/CRM system "would diminish the due process rights of the ratepayers and other interested parties," because, inter alia, the Water Division might decline to grant discovery to DRA, the discovery period would be shorter than in the application process, and DRA would not have a clear right to an evidentiary hearing. (DRA Opening Comments, pp. 6-7.)

While we acknowledge that the Tier 3 advice letter process is not the full equivalent of an application, it gives to parties who protest the advice letter many of the same procedural protections that the formal application process provides. In view of the concerns that we have set forth in the text about the issues that GSWC's testimony on the CIS/CRM system fails to address, we fully expect that in the event of a protest, the Water Division will provide protesting parties with ample opportunity for discovery and comment. We also reiterate that if GSWC fails to make the necessary showing, then we expect that the resolution the Water Division will present to the Commission in connection with the advice letter will recommend its rejection, and require that GSWC proceed by application to recovery any additional CIS/CRM costs beyond those allowed in this decision.

²⁷ Administrative Law Judge's Ruling Denying Motion of Division of Ratepayer Advocates to Strike Rebuttal Testimony, filed July 12, 2006 (ALJ Ruling Denying DRA Motion to Strike).

²⁸ In her direct testimony, Ms. Darney-Lane notes that prior to the creation of the SVP-Operations position in 2002, GSWC's three regions were given general oversight from a Vice President-Customer Service. Today, the three regional vice presidents report to the SVP-Operations, as does the Vice President-Water Quality. (Ex. 6, Darney-Lane, p. 7.)

²⁹ Other situations where the SVP-Operations has had to supervise wells being removed from service and new permits being applied for include Converse Well No. 1 (carbon tetrachloride, 12 months to resolve); Hawaiian Well No. 1 (arsenic, 12 months to resolve); Massinger Well No. 1 (arsenic, 33 months to resolve); Centralia Nos. 3 and 4 (arsenic, 5 months to resolve); Century No. 1 (arsenic, 8 months to resolve). (Id. at 34-36.)

In cases where new permits were needed to address the updated arsenic standards, GSWC often filed the permit applications months before the new standards took effect. As a rule, however, the wells were left in service until just before the new arsenic standards took effect on January 23, 2006.

³⁰ In her direct testimony, Darney-Lane states that "approximately 10% of the SVP-Operations job is related to compliance with Sarbanes-Oxley." (Ex. 6, Darney-Lane, p. 11.)

³¹ Although the record is not entirely clear on the point, it appears that Dickson is currently serving as GSWC's SVP-Operations. Although Dickson described himself on the stand as GSWC's "senior vice president with operations and administration," he also stated that he is responsible for "all the functions within the company but the financial functions that Mr. Sprowls oversees." (Tr., p. 975.) In his testimony, Sprowls states that his job title is "Chief Financial Officer, Senior Vice President of Finance and Secretary of GSWC." (Ex. 17, p. 1.)

On the other hand, the 2005 and 2006 Annual Reports for GSWC's corporate parent, American States Water Company, lists the following four Senior Vice Presidents and their titles for GSWC: Dickson (Senior Vice President), Sprowls (Chief Financial Officer, Senior Vice President of Finance and Secretary), Denise L. Kruger (Senior Vice President of Operations), and Susan L. Conway (Senior Vice President of Administrative Services). Neither GSWC's briefs nor testimony explain the apparent overlap between the roles of Mr. Dickson and Ms. Kruger, or the duties of Ms. Conway.

³² On page 2-5 of his report for DRA, Aslam states that if the Commission authorizes the new position of SVP-Operations, DRA is not opposed to GSWC's proposal to split the salary for the existing position of Administrative Secretary-Operations between the new SVP-Operations and Vice President-Customer Service for Region I.

³³ Dickson points out that in 1996, it was unusual for GSWC to have to obtain a Conditional Use Permit (CUP) before beginning construction, and there was no requirement that the company obtain an NPDES discharge permit, submit geotechnical studies or Traffic Control Plans, or (usually) undergo a full CEQA review before beginning construction. Today, all of these things are required before construction can commence. (Id. at 43.)

³⁴ In her direct testimony, Darney-Lane notes that 2004 was the first year in which GSWC tasked someone with the assignment of acting as a capital projects manager. As a result of this trial run, "his help alone contributed to our timely closing of over 300 [General Work Orders] in 2004." (Ex. 6, Darney-Lane, p. 13.)

³⁵ Dickson also points out in his rebuttal testimony that GSWC ratepayers are being asked to pay only 69% of the cost of the Call Center Support Analyst position; the rest is charged to the "new business" accounts of GSWC's non-regulated affiliate, ASUS. (Ex. 11, p. 53.) Dickson does not comment on what this may signify about how many of the calls that come into the Call Center are from customers of GSWC versus customers of the entities with which ASUS contracts.

³⁶ On this issue, Dickson states:

The DRA assumes that this position will be separated from the application support resources, which reside in the functional areas. This is not the case. The Application Support Manager will manage all of the existing technical support personnel. While the technical support personnel of the functional areas' applications will be moved under the centralized Applications Support Department, the operations and administration personnel of the functional areas' applications will remain in their current department. So, functional resources will not be duplicated, as the DRA suggests. (Ex. 11, p. 54.)

³⁷ It appears that apart from introductory sentences, Dickson's "rebuttal" testimony on both the Manager's and the Generalist's position is identical to the direct testimony of Darney-Lane on these positions. (Compare Ex. 5, Darney-Lane, pp. 27-28, 39-41 with Ex. 11, pp. 57-59, 60-63.)

³⁸ Aslam's list of corporate officers is not consistent with the one set forth in the 2006 annual report of American States Water Company, the corporate parent of GSWC. That annual report lists the following officer positions for GSWC: President and CEO; Chief Financial Officer, Senior Vice President, Corporate Secretary and Treasurer; Senior Vice President; Senior Vice President of Operations; Senior Vice President of Administrative Services; Vice Presidents of Customer Service for Regions I, II and III; Vice President, Treasurer and Assistant Secretary; Vice President of Water Quality; and Vice President of Regulatory Affairs.

³⁹ The Commission has also used the services of BWG from time to time. See Administrative Law Judge's Ruling Denying Motion of Pacific Gas and Electric Company to Compel Discovery Responses from the California Public Utilities Commission and the Barrington Wellesley Group, Inc., issued March 20, 2001 in A.00-11-038 et al.

⁴⁰ With respect to this item, Aslam asserts that GSWC "currently incurs on average an expense of $318,723 per year under [this program.] This training is above and beyond the Management Development training that EDU provides in-house. Therefore, ratepayers have to bear the burdens of this duplication of efforts." (Id. at 2-16.)

⁴¹ In its August 13, 2007 opening comments on the PD, DRA asserts that this language reflects legal error because it reverses the burden of proof in water rate cases. According to DRA, the Rate Case Plan appended to D.04-06-018 makes clear that water utilities like GSWC have the burden of proof when requesting a rate increase, and that if this requirement is to be given any meaning here, we must deny GSWC's requests for EDU funding "because their support rests for the most part on vague and unsupported generalizations in GSWC's rebuttal." (DRA Opening Comments, p. 13.)

DRA's argument is without merit, because it confuses the burden of proof with the burden of going forward with evidence, and also ignores the procedural context of this case. As our discussion in the text makes clear at several points, we are well aware that water utility companies bear the burden of proof when requesting a rate increase. However, our decisions also recognize that there is an important distinction between which party in a rate case has the ultimate burden of proof and which party has the burden of producing evidence once a prima facie showing on an issue has been made. Re Pacific Bell, D.87-12-067, explained the distinction as follows:

[W]here other parties propose a result different from that asserted by the utility, they have the burden of going forward to produce evidence, distinct from the ultimate burden of proof. The burden of going forward to produce evidence relates to raising a reasonable doubt as to the utility's position and presenting evidence explaining the counterpoint position. Where this counterpoint causes the Commission to entertain a reasonable doubt regarding the utility's position, and the utility does not overcome this doubt, the utility has not met its ultimate burden of proof. (27 CPUC2d 1, 22.)

See also Universal Studios, Inc. v. Southern California Edison Company, D.04-04-074, fn. 13, mimeo. at 31-32 (where defendant raises a new issue in its direct testimony, complainant has the burden of going forward with evidence concerning the new issue either in rebuttal testimony or on cross-examination.)

In this case, although DRA apparently propounded data requests to GSWC about the EDU expenses and positions the utility had requested, it did not become clear that DRA would be challenging the efficacy of the entire EDU program until DRA submitted its general office report on May 25, 2006. In that report, five pages were devoted to attacking the cost-effectiveness of the EDU. (Ex. 23, pages 2-13 to 2-18.) In view of this substantial attack on a program DRA had previously supported, GSWC had little choice but to devote a substantial portion of its rebuttal testimony to countering DRA's arguments. In his rebuttal testimony, Mr. Dickson did so. (Ex. 11, pp. 63-75.)

Given this procedural posture, the burden of going forward with evidence to rebut Mr. Dickson's arguments clearly rested on DRA. DRA had to choose between meeting this burden by requesting an opportunity to submit surrebuttal testimony, or by cross-examining Mr. Dickson on his rebuttal testimony. Although DRA apparently chose the latter course, little of its cross-examination of Dickson was devoted to the EDU.

The cited pages of Exhibit 23, Dickson's rebuttal, and the parties' briefs comprise the entire record on which the PD had to decide the cost-effectiveness issue. The PD found the arguments presented by Mr. Dickson in support of the EDU more persuasive than those presented by DRA attacking the EDU's cost-effectiveness. Under these circumstances, GSWC met its burden of proof concerning the EDU's cost-effectiveness, and there was no legal error in how the PD handled the matter.

⁴² D.04-03-039 provides in relevant part: "This issue has also raised a concern regarding SCWC's burden in justifying its request. With the application, SCWC submitted testimony, which included a very brief description of the need for this particular project. After ORA recommended the project be rejected for lack of justification, SCWC provided a more detailed justification in rebuttal testimony. A project of this magnitude, which is in excess of $5 million, requires more attention than what was given by the utility in initially justifying its proposed budgets. Providing the basic justification in rebuttal is unfair, since parties are not generally given the opportunity to respond to rebuttal with testimony of their own. In this case, rebuttal was issued on May 1, 2003 and hearings began on May 12, 2003. The timeframe to conduct discovery on rebuttal, even for the purpose of cross-examination, was limited. When the utility has the evidentiary burden, we caution against the use of rebuttal testimony to provide the basic justification. As a matter of fairness, we must seriously consider either striking such testimony or extending the proceeding, at the utility's risk, to allow for responsive testimony from the other parties." (D.04-03-039, mimeo. at 84-85; footnote omitted.)

⁴³ In its August 13, 2007 opening comments, GSWC points out that, contrary to the assertion in the PD and alternate PD, a statement that 31% of the salary for the new Call Center Support Analyst position was being charged to ASUS did appear in the company's direct testimony.  The statement was reflected in Ms. Darney-Lane's comment that 31% of the new position's salary was being "directly charged to non-regulated cost centers."  (GSWC Opening Comments, p. 11; Ex. 5; Darney-Lane, p. 5.)

While we acknowledge that this reference to charging a portion of the new analyst's salary to ASUS did appear in GSWC's direct testimony, it should be noted that the statement was easy to miss. Moreover, the statement concerning the 31% appeared in an introductory discussion by Ms. Darney-Lane of the general methodology she had used to forecast labor expense for the company's Region II and general office operations.  No mention of the 31% was made in the actual discussion of the Call Center Support Analyst position, which was very brief.  (See, Ex. 5, Darney-Lane, p. 23.)  Of note: the data requests attached to GSWC's August 20, 2007 reply comments are associated with GSWC's rebuttal testimony.

⁴⁴ In D.04-07-022, in commenting upon Edison's failure to include certain capital additions to a Customer Service Business Unit project in its direct testimony, we said:

The Commission has held that it is not permissible for utilities to hold back on the presentation of salient information until the submission of rebuttal testimony. We would be well within our rights and responsibilities if we were to disallow these capital additions on procedural grounds as advocated by ORA. (Mimeo. at 157.)

However, the capital additions were not disallowed in D.04-07-022 because the Commission expressly found that Edison's failure to include them in its direct showing was inadvertent, and not part of a litigation strategy. (Id. at 156-57, 335.)

See also, Pacific Gas and Electric Company, D.87-03-034, 24 CPUC2d 45, 1987 Cal. PUC LEXIS 544 (denying PG&E's petition for modification where the utility alleged that accelerated tax depreciation it had previously taken made it impossible to recover authorized plant investment, where the evidence PG&E had originally offered in the proceeding did not suggest there was a significant difference between the plant's tax basis and its book basis, even though such evidence was apparently available.)

⁴⁵ In the same ruling, the FERC ALJ noted that trying to deal with the problem of sandbagging by permitting the aggrieved party to submit surrebuttal testimony is often not a good solution:

Allowing such evidence into the record on rebuttal invariably gives rise to the need for the undertaking of additional discovery and for the filing of surrebuttal evidence by the prejudiced party, and thus, at times, interminably delaying the conclusion of the hearings. To paraphrase . . . another proceeding, if the material was not included in the initial submission due to lack of diligence, or was knowingly withheld, then merely ordering further discovery and surrebuttal would be tantamount to rewarding the slothful and recalcitrant litigant and unduly burdening the more assiduous participants." (Id.)

Notwithstanding this general caution, the FERC ALJ did allow some of the intervenors to submit surrebuttal testimony on specific, narrowly-focused issues where Edison had a plausible argument that it could not have included the disputed material in its direct testimony. (Id. at pp. 65,067-65,068.)