Discussion

Although we think that both Edison and ORA have cast each other's positions in the most extreme light, we must agree with ORA that Edison has failed to meet the burden of proof established in Resolution E-3544 for recovering TRRRMA costs in distribution rates. In essence, Edison is making the same argument for setting distribution rates "residually" that it made in 1997, an argument we rejected in D.97-08-056. (See 74 CPUC2d at 19.) We continue to think that before allowing recovery of the A&G and G&I costs booked in TRRRMA, Edison should be required to offer some proof that these costs are reasonable and distribution-related.

Even though Edison disclaims such an intention, it also seems clear that in this application, Edison is asking as a practical matter to be relieved of the results of its agreement to use the multi-factor allocation methodology. Apart from the fact that we accepted this methodology in D.97-08-056, we are unwilling to undercut it by granting the relief Edison has requested because the multi-factor methodology is the product of a quid pro quo between Edison and ORA.⁷ As Edison's counsel explained at the PHC, ORA had criticized Edison's use of the labor ratio allocation methodology in its first nongeneration PBR application, A.93-12-029. In response to this criticism, the multi-factor approach was jointly devised by Edison and ORA in workshops under the auspices of the Ratesetting Working Group. (PHC Tr., pp. 9-10; Edison Brief, pp. 5-6.)

While we know the nature of the new allocation methodology that resulted from these workshops, we do not know what other consideration Edison may have received in exchange for its agreement to use the multi-factor allocation approach in its filings. This adds to our reluctance to grant relief that would disturb the bargain apparently reached between Edison and ORA.

In addition to our reluctance to disturb this bargain, we are troubled by the fact that Edison has not attempted to offer any proof that the TRRRMA costs it is seeking to recover are either reasonable or distribution-related, as required by Resolution E-3544. Instead of offering such proof, Edison has merely (1) repackaged the residual approach, and (2) insisted that representations made by our staff during the transmission rate proceeding caused FERC to reject the multi-factor methodology, because FERC believed the TRRRMA costs would be recovered here at the Commission.

Neither of these arguments adequately addresses the clear requirements of Resolution E-3544 and D.97-08-056, from which the resolution was derived. If Edison believed that it could not meet these requirements, then - as ORA pointed out on page 16 of its September 7, 2001 testimony - Edison should have filed a petition for modification of D.97-08-056. However, no such petition has been filed.

Edison's continued reliance on the residual approach rejected in D.97-08-056 is apparent from the following passage in Edison's reply to ORA's protest:

"In [D.97-08-056] the Commission found reasonable and adopted a nongeneration revenue requirement, based on 1995 GRC authorized A&G and G&I plant costs, a portion of which is now recorded in the TRRRMA. FERC adopted a transmission revenue requirement that did not include the TRRRMA costs based solely on the use of a different allocation methodology. Logically, (1) the TRRRMA costs have been determined by the Commission to be reasonable nongeneration costs, (2) the FERC found them not to be transmission related and neither the CPUC [n]or FERC disallowed the costs from recovery,[⁸] therefore, by definition, (3) they are distribution-related costs." (Reply to Protest, p. 4.)

There are several flaws with the conclusion in this syllogism. First, although it is true that D.97-08-056 concluded it was reasonable to treat the A&G and G&I costs in question as nongeneration costs, the decision emphasized that it was adopting these costs as an interim measure, and that the Commission expected further proof would be offered in the event FERC refused to include some of the claimed transmission costs in transmission rates:

"Just as we have declined to reduce the distribution revenue requirements in this proceeding to account for costs associated with activities the utilities may no longer conduct,[⁹] we decline to increase the distribution revenue requirements to account for [adverse] FERC decisions. In each instance, the utilities will have an opportunity to make their case with regard to specific revenue requirements changes in their PBR proceedings, or, for PG&E, general rate case. In the interim, we will adopt the revenue requirement for distribution that each utility proposes here with the adjustments we make in subsequent sections, consistent with law and policy. To the extent necessary, we will revisit these revenue requirements at a later date . . ." (74 CPUC2d at 19.)

The second flaw in Edison's syllogism is the proposition that it is reasonable to treat all costs eligible for inclusion in TRRRMA as distribution-related.¹⁰ The Commission clearly rejected a variation of this argument in D.97-08-056, noting that "one of our criteria for determining the reasonableness of a proposal is whether it allocates the costs of a given function to that function's revenue requirement." (Id.) The Commission took this position largely because of concerns that since distribution is a monopoly function, the utilities would be tempted during the unbundling process to allocate excessive costs to that function:

"In pursuing a policy to promote more efficient generation markets, we reject proposals to allocate to monopoly functions any costs associated with services that are or will be subject to competition. Specifically, we will not permit allocations of generation cost to distribution customers. To do so would compromise market efficiency by producing artificially low utility generation rates . . . and provide competitive advantages, which would stifle competition to the utilities. Moreover, any allocation to monopoly customers of costs associated with competitive products would be unfair to monopoly customers because they would, in effect, be required to subsidize shareholder profits." (Id. at 15.)

In addition to disagreeing with the proposition that all costs booked in TRRRMA should automatically be considered distribution costs, we disagree with Edison's assertion that the Commission staff effectively promised FERC that any costs excluded from transmission rates due to FERC's use of the labor ratio allocation methodology would be recovered by Edison in the distribution rates subject to our jurisdiction. Although Edison doubtless wishes that our staff had made such promises, it is clear from an examination of the November 30, 1999 reply comments in FERC Docket No. ER97-2355-000 that no such representation was made. After describing the circumstances leading to the authorization of TRRRMA, staff's comments concluded that "if Edison is able to subsequently demonstrate that these costs are reasonable, distribution-related costs (as opposed to generation-related costs), Edison can recover these costs in distribution rates." (Emphasis added.) This statement stops well short of an unqualified promise.

We also think that changes in the California electric industry brought about by restructuring -- changes that were only beginning to take place when D.97-08-056 was decided -- make it reasonable to continue to hold Edison to the burden of proof for TRRRMA cost recovery set forth in Resolution E-3544. D.97-08-056 itself contemplated that the changes caused by restructuring would make it appropriate to reexamine utility revenue requirements:

"We are aware that the utilities' activities will change in the next few years. For example, the ISO will take on dispatch and management of electric loads. The utilities may eliminate or redefine some of their customer relations and generation activities. Even if we do not create new forums to consider these potential cost reductions, we recognize that these types of changes in activities will affect utility revenue requirements in the near future. We find nothing in AB 1890 to restrict this Commission's authority to adjust revenue requirements as long as the changes are otherwise consistent with the statute's provisions." (74 CPUC2d at 15.)

In fact, in both Edison's transmission proceeding at FERC and in PBR filings at this Commission, some reexamination of Edison's nongeneration revenue requirement has taken place. A noteworthy example of this occurred in the FERC ALJ's Initial Decision in ER97-2355-000. In that case, our staff had challenged Edison's decision to compute A&G expenses for the period beginning January 1, 1998 (the so-called "Phase II" period) by escalating 1995 data. The FERC ALJ agreed that because of restructuring developments, Edison should be required to recompute its A&G expenses based on actual 1997 data:

"As CPUC asserts, the fundamental purpose of this proceeding is to determine just and reasonable rates, particularly, in this proceeding where SCE has drastically restructured and downsized its previous utility operations and has turned over its transmission facilities to ISO control. Clearly the 1997 recorded A&G amounts, with the adjustment for divested generating plants, will more likely yield just and reasonable rates than SCE's poorly founded projections. Accordingly, SCE's projected 1998 A&G expenses are found to be unjust and unreasonable. SCE will, therefore, be directed to file a compliance filing substituting its 1997 actual data for A&G expenses for Period II, with the appropriate adjustment for divested generating plants." (86 FERC at pp. 65,176-77.)

The annual data that Edison has reported on its FERC Form 1, as well as Commission resolutions concerning the awards Edison has requested under its PBR mechanism, also suggest that since electric restructuring went into effect, Edison has experienced considerable volatility with respect to its distribution expenses. For example, Table A-1 to Resolution E-3772 (which deals with Edison's request for a PBR award for calendar year 2000) shows Edison's PBR results-of-operation for 1997 through 2000. For the operation and maintenance expense data for distribution shown in Table A-1 (which are set forth below), the figure for 2000 is 62% higher than the one for 1997, and although the amounts increase with the passage of time, they do not reflect a simple trend from year-to-year:

1997 $172,299,000

1998 $277,127,000

1999 $243,964,000

2000 $278,065,000¹¹

Although the data is not directly comparable, a similar (although smaller) variability can be seen in the operation and maintenance expense data for distribution that Edison reported to FERC on its Form 1 for the years 1996 to 2000, data of which we take official notice:

1996	$ 161,688,00012
1997	$ 177,924,000
1998	$ 203,754,000
1999	$ 179,611,000
2000	$ 201,689,000

In view of the significant effect that electric restructuring has apparently had on Edison's distribution expenses, and the variability these expenses have shown since restructuring took effect, it is not unreasonable to require that before Edison is allowed to recover the TRRRMA balances in rates, it should offer some proof that the costs making up these balances are both reasonable and distribution-related. ¹³ However, Edison has elected not to offer any such proof, relying instead upon the syllogism quoted above.

We recognize that it will probably not be easy for Edison to offer the proof required by today's decision. As Edison's counsel stated at the PHC, "I don't have a witness to put on the stand who can point to a particular dollar in this $24 million a year and say, `[t]hat's definitely a distribution dollar.'" (PHC Tr., p. 9.)¹⁴ We also recognize that the task of offering proof has been made more difficult by the apparent gaps in Edison's accounting data.¹⁵ However, the burden of proof that Resolution E-3544 places upon Edison is clear, and it was obviously derived from the discussion in D.97-08-056. Under these circumstances, Edison cannot claim that the standard of proof it must meet in order to recover TRRRMA balances is, in ORA's words, "an unpleasant surprise." (ORA Testimony, p. 16.)

We close by noting that although Edison appears disinclined to offer this Commission the proof that Resolution E-3544 requires, Edison is not without some recourse. As noted above, the utility has filed at FERC a Conditional Request for Rehearing of Opinion 445, which sets forth both legal and policy arguments why FERC should reconsider its decision to affirm the ALJ's rejection of the multi-factor allocation methodology. At the PHC, Edison's counsel clearly stated that if this application were to be denied, Edison would be returning to FERC to pursue this alternative remedy. (PHC Tr., pp. 7-8.)

"This proceeding is not the appropriate forum for reaching the potentially complex issues relating to changes in revenue requirements. In D.96-10-074, we ordered the utilities to file revenue requirements `based on our last authorization and separate this total between transmission and distribution' . . . By this, we stated our intent to consider existing revenue requirements in this proceeding. We have accordingly emphasized allocations of existing costs to utility functions in this proceeding[,] rather than seeking to accomplish the more ambitious task of reviewing revenue requirements." (74 CPUC2d at 15; emphasis in original.)

"SCE is not requesting that rates be set residually by treating as `distribution-related, all costs that FERC declined to include in transmission rates.' SCE did not record in TRRRMA all costs that FERC declined to include in transmission rates. As opposed to the approximately $24 million recorded annually in TRRRMA, SCE's application did not also request recovery of an additional $14 million removed from SCE's FERC rate request . . . . . . If SCE had done so, the TRRRMA revenue requirement request would have been approximately $38 million annually, not the $24 million that SCE is actually requesting. . . By not including the $14 million per year in its TRRRMA costs, SCE is requesting recovery of only distribution costs that have been found reasonable by this Commission and FERC." (Edison Comments, pp. 5-6; emphasis in original, footnotes omitted.)

"[W]e were using a methodology here [i.e., the multi-factor methodology] that had several steps to it . . . some parts of it you're looking directly at certain costs and you're assigning them . . . "But for the most part, there comes a point where you're using a methodology that doesn't enable you to look at a particular dollar and put somebody on the stand and say, `Yep, that was in my business unit and I spent that dollar, and next year I'll need it and I'll spend it again next year.' "So we are at this position where we don't have a witness to take the stand to talk specifically to those costs." (PHC Tr., p. 14.)

"SCE has clearly not shown that its method provides a more accurate result. As demonstrated by Staff and CPUC, SCE's detailed analysis of costs is lacking. The records produced lacked function codes (information on the activity performed) or location codes (geographic information). While SCE made certain judgments in formulating these assignments, there is no evidence of how these assignments were made or why certain assignments were made. Direct assignments of A&G expenses were made without including any information as to the reasons for such assignments . . . Only two percent of the A&G expenses were allocated to the direct pool . . . indicating that there is insufficient data to make direct assignments." 86 FERC at p. 65,145 (citations omitted).