Word Document

COM/JLN/ccv

      ALTERNATE DRAFT

    Item 1a

  
   

      9/7/2000

     

Decision ALTERNATE PROPOSED DECISION OF COMMISSIONER NEEPER

(Mailed 8/2/00)

BEFORE THE PUBLIC UTILITIES COMMISSION OF THE STATE OF CALIFORNIA

Application of SOUTHERN CALIFORNIA EDISON COMPANY (U 338-E) to: (1) Consolidate Authorized Rates And Revenue Requirements; (2) Verify Residual competition Transition Charge Revenues; (3) Review the Disposition of Balancing and Memorandum Accounts; (4) Review Generation Cost Jurisdictional Cost Allocation; (5) Review the Reasonableness of the Administration of the Low Emission Vehicle Program; (6) Review the Administration of Special Contracts; and (7) Present a Proposal for the Inclusion of Long Run Marginal Costs in the Power Exchange Energy Credit.

Application 99-08-022
(Filed August 9, 1999)

And Related Matters.

Application 99-08-023
Application 99-08-026

(Filed August 9, 1999)

(Appearances are listed in Appendix A.)

TABLE OF CONTENTS

Page

OPINION 2

Summary 2

I. Background and Procedural History 2

II. The PX Credit 5

III. RMR 44

IV. Other Contested Issues 51

V. Stipulations 53

VI. Uncontested Issues 54

VII. Motion of SDG&E to File Advice Letter 54

Findings of Fact 56

Conclusions of Law 71

ORDER 76

Appendix A - List of Appearances

Appendix B - Stipulation Among the Federal Executive Agencies, the Office of Ratepayer Advocates, The Utility Reform Network, and Southern California Edison Company Regarding Jurisdictional Allocation Issues in the 1999 Revenue Adjustment Proceeding (Application No. 99-09-022)

Appendix C - CPUC Recommendation Re RMR Allocation

OPINION

Summary

In this decision we adopt a Power Exchange (PX) credit adder of 0.034 cents per kilowatt-hour (¢/kWh) for all three utilities. We find that it is unreasonable to exempt wholesale customers from paying their fair share of Reliability Must-Run (RMR) costs and we put Southern California Edison Company (Edison) on notice that retail ratepayers will not bear the burden of 100% of the RMR costs in the future. We discuss the confusion regarding the definition and computation of long-run marginal costs (LRMC), and hold that the proper method for determining the PX credit is by use of long-run marginal costs, which was the method used by ORA and ARM.

I. Background and Procedural History

In our opinion on Cost Recovery Plans, Decision (D.)  96-12-077, we recognized the need to streamline utility cost recovery mechanisms to effectively implement a restructured electric utility industry in accordance with Assembly Bill No. 1890 (AB 1890). Accordingly, we created the Revenue Adjustment Procedure (RAP) to review, track, and compare each utility's authorized revenue requirements with actual recorded revenues and to approve any necessary adjustments or updates to authorized revenues. Such adjustments are associated with the performance-based ratemaking (PBR) mechanism and decisions addressing such issues as power purchase contracts, public purpose programs, nuclear decommissioning, and transition costs.

A number of issues were added to the 1999 RAP by D.99-06-058, the first RAP:

In addition, we designated the 1999 RAP as the proceeding to consider (1) requests for authorization to use recorded monthly jurisdictional allocation factors for assigning recorded system generation costs between retail and wholesale requirements customers; and (2) LRMC of energy procurement for inclusion in the PX credit provided to utility customers that elect direct access.

D.99-06-058 described the context of PX price issues as follows:


Under the restructured electricity market in California, customers may subscribe to "bundled service" from the utility or "direct access" service from a competitive energy provider. Customers who purchase bundled service from the utility pay a PX charge to cover the utility's power supply costs, while customers who elect direct access service receive a credit on their bills called the "PX credit" that offsets the energy costs included in the bundled rate. (Id. at 20.)

In the first RAP, the Office of Ratepayer Advocates (ORA) recommended that all on-going costs related to energy procurement from the PX by the utilities on behalf of bundled-service customers including costs of maintenance, refinement, and enhancement of utilities' systems should be recovered through the PX price, and that direct access customers should be given a credit for those costs. D.99-06-058 adopted refinements to the method of calculating the PX credit, but concluded that there was not an adequate record to adopt changes to its composition at that time, and ordered its composition to be reviewed in the current RAP.

Edison, Pacific Gas and Electric Company (PG&E), and San Diego Gas & Electric Company (SDG&E) filed their 1999 RAP applications in August 1999. The Alliance for Retail Marketers (ARM)1 and ORA filed protests. The applications were consolidated for hearing.

The Federal Executive Agencies (FEA), ARM, ORA, and The Utility Reform Network (TURN) submitted direct testimony; Aglet Consumer Alliance (Aglet), the California Large Energy Consumer's Association (CLECA), FEA, TURN, the Coalition of California Utility Employees (CCUE) and applicants submitted rebuttal testimony. Evidentiary hearings began on February 7, 2000 and concluded on February 22, 2000. The proceeding was submitted on April 17, 2000 with the filing of concurrent opening and concurrent reply briefs.2

At a prehearing conference, the presiding administrative law judge (ALJ) directed the utilities to address how they intend to proceed at FERC to recover a fair allocation of RMR costs from the utilities' wholesale transmission customers. The ALJ directed that "each utility shall state what percentage of its transmission load is represented by its wholesale transmission customers, what steps each utility intends to pursue at the FERC to recover a fair share of RMR costs from the utility's wholesale transmission customers, and how soon each utility expects to begin [recovering] a fair share of RMR costs from its wholesale transmission customers." We affirm those instructions.

In their testimony, ORA and TURN addressed the allocation of RMR costs among retail customers, an issue that had not been added to the proceeding. PG&E filed a motion to strike this testimony. The motion was granted. Similarly, Aglet served rebuttal testimony addressing the allocation of RMR costs between wholesale and retail customers during the entire transition period (retroactive to April, 1998). Edison filed a motion to strike Aglet's testimony. The motion was granted.

II. The PX Credit

A. Background

Competition for California retail electric service is frustrated and handicapped when that competition is based solely on the wholesale price of power. The UDCs' price for power, as represented by the PX Credit on direct access customer bills, is essentially equal to the wholesale price of power from the California Power Exchange. Therefore, ESPs are required to compete on a retail basis with a PX Credit which only represents wholesale costs. Meanwhile, the UDCs' retail costs of procuring and serving electricity to their bundled customers are currently excluded from the PX Credit. The Commission recognized this discrepancy in D.99-06-058 (the "1998 RAP Decision"), when it ordered that the long run marginal costs ("LRMC") associated with the utilities' procurement and associated retail marketing and customer service functions should be included in the calculation of the PX Credit.

We will direct the utilities to include the long run marginal costs of these functions in future calculations of the PX Credit, that is, in the utilities' 1999 RAP applications. Recognizing that long run marginal cost studies would be a difficult undertaking in the near term, we will require the utilities to use actual April 1998-April 1999 recorded costs or 1999 budgeted or forecasted costs as proxies for long run marginal costs. The actual recorded costs should include allocations of overheads. It is our intent to review these additional PX Credit items on an expedited basis in the 1999 RAP. 3

Inadequate or understated PX Credits harm both ESPs and direct access customers. Customers are denied the credits which would enable them to incur greater energy cost savings by electing to utilize an alternate, direct access supplier. This provides a disincentive to participate in direct access because of the widespread perception among customers that direct access is not worth the trouble because the savings are so small. ESPs are harmed because this disincentive to customer participation in direct access reduces the potential market which is available.

A key policy principle established by D.97-08-056 is that "costs associated with one function will not be allocated to other functions." That is, unbundling utility rates and services is one of the primary means by which efficient markets may develop for utility products and services. To the extent that prices reflect the costs of associated products and services, sellers will offer the most efficient quantity and variety of these products and services. Buyers will then be able to make purchasing decisions that best serve their interests. In pursuing a policy to promote more efficient generation markets, D.97-08-056 rejected proposals to allocate to monopoly functions any costs associated with services that are or will be subject to competition, specifically stating an intent to not permit allocations of generation cost to distribution customers.4 It found that allocation of generation costs to distribution customers would compromise market efficiency by producing artificially low utility generation rates (or utility profits which do not correspond to utility risk), and would provide competitive advantages that would stifle competition to the utilities.

D.97-08-056 rejected Southern California Edision's (SCE) argument that allocation of fixed administrative and general (A&G) costs to generation would improperly disallow appropriately incurred costs because of SCE's perception that it could not recover fixed costs in competitive generation markets, and rejected similar arguments by Pacific Gas and Electric Company (PG&E) and San Deigo Gas & Electric Company (SDG&E). The Commission stated its policy on allocation of these costs in D.97-08-056, as follows:

"Some utility costs do not vary over some period of time. They are incurred notwithstanding the utility's output. It does not necessarily follow, however, that distribution customers should assume liability for all such costs even if the utilities will continue to incur them. The utilities' argument that they will be unable to recover these costs in generation markets is not convincing. Their competitors also incur fixed costs. Arguably, competitors' fixed costs are higher per unit of output than the utilities' because many competitors will not realize the economies of scale or scope which the utilities enjoy. A utility's generation system, whether it is owned and operated by the utility or any other entity, will continue to incur fixed costs which must be allocated to generation. .... Consequently, allocating to distribution customers all fixed costs would create a competitive advantage to the utilities at the expense of captive customers, contrary to our stated objectives and the requirements of AB 1890."

D.97-08-056 stopped short of distinguishing between procurement (costs of providing power to end-use customers) and generation (including costs of generation plant that could be transferred to new owners, and costs of marketing a plant's output), because the extent of ultimate divestiture of utility-owned generation could not be predicted at that time. The three utilities' divestiture of very significant portions of the generation that they once owned means that D.97-08-056's principle of recovering generation-related costs through markets must now evolve to distinguish among the costs that D.97-08-056 allocated generally to "generation", between those that are associated with power production and those that should be allocated to procurement for end-use customers.

Indeed, D.99-06-058 in the first RAP confirmed that D.97-08-056 did not resolve the composition of the PX credit on a permanent basis. D.99-06-058 added to the Commission's previous statements that costs of services that are or will be subject to competition will not be assigned as costs of monopoly functions, stating at pp. 23-24: "[U]tility commodity prices must ultimately recognize those costs which the utilities must recover in the long run as any other provider. ... As part of that strategy utility pricing must eliminate any `competitive advantage created by an institutionalized removal of costs otherwise intrinsic to the provision of a service,' as DGS' comments describe the utilities' proposals."

The issue of the PX credit - both its constituents and the process of its calculation - was one of the most contentious issues in the 1998 RAP, with intervenors such as Enron and the Alliance for Retail Markets (ARM) advocating the inclusion of many more categories of costs in the credit. D.99-06-058 rejected most of these proposals, but it directed Edison, PG&E and SDG&E to prepare studies of certain costs for inclusion in the credit in the 1999 RAP case. Specifically, D.99-06-058 ordered the UDCs to study the Long-Run Marginal Costs (LRMC) of account managers, customer service representatives, self-provision of ancillary services and the financing costs of Power Exchange purchases.

B. Testimony of Parties

The UDCs prepared the required studies, but two intervenors, ARM and ORA, filed testimony contending that the UDC studies are inadequate and misleading. TURN also offered some brief testimony.

1. UDCs

The UDC approach may be stated as follows:

    · Marginal cost is defined as the change in cost associated with a small change in output (Helgens/PG&E, Tr. 1/12; Marcus/CCUE, Tr. 6/869).

    · A "small change" in output is defined generally as a one-unit change in output. For many firms, the nature of production is such that it is very possible that the change in costs associated with a one-unit increase could be the same as that associated with a one-unit decrease, although that general rule is not always true (Helgens/PG&E, Tr. 1/12, 13).

    · The term "long-run" is typically defined as a period of time where all factors of production such as labor, plant, equipment resources, and natural resources can be varied, assuming they are capable of being varied (Helgens/PG&E, Tr. 1/13, 14; Marcus/CCUE, Tr. 6/869, 870).

    · The "short-run" is defined typically as a period of time where at least one factor of production that is capable of being changed over the long term remains fixed (Helgens/PG&E Tr. 1/13, 14).

    · The concepts of long-run and short-run vary from firm to firm depending upon the nature and business of the firm, and the period of time necessary for the firm to have the flexibility to vary all factors of production that are capable of being varied (Helgens/PG&E Tr. 1/13, 14).

    · When one combines the concepts of "long-run" and "marginal costs," one is referring to the change in cost from a small reduction (or increase), perhaps a one-unit reduction (or increase), in output where all costs related to the factors of production that are capable of becoming variable over time are in fact variable (Helgens/PG&E, Tr. 1/13; Marcus/CCUE, Tr. 6/870).

    · The determination of LRMC doesn't occur in a vacuum, one must look at the firm and the industry to determine if there are any constraints (defined by such things as production technologies or legal and regulatory requirements) that will affect the ability of the firm to reach an optimum input mix by varying all factors of production (Helgens/PG&E, Tr. 1/14; Croyle/SDG&E, Tr. 3/385).

    · The UDC's default commodity procurement responsibility (which applies to all its distribution customers, direct access and non-direct access alike) is a type of constraint that affects the UDC's ability to vary all its procurement related costs in the long-run (Croyle/SDG&E, Tr. 3/385, 386).

2. ARM

As it did in the previous RAP proceeding, ARM contends that ESPs are placed at a disadvantage in the California market by having to compete against a UDC PX credit that does not include all the relevant costs. That is, ARM contends that some UDC costs associated with the (competitive) power market are recovered via the (monopoly) distribution rates. According to ARM witness Chris King, "the UDCs' price of power, presented as the PX credit on direct access customer bills, is equal to the wholesale price of power. All of the UDCs' retail costs of procuring and serving electricity to its bundled customers are excluded from the PX credit." (p. 3)

King depicts the Retail Energy functions as performed by a company existing within the UDC (the "utility-operated ESP"), and notes that such a company would have all the administrative overheads (personnel, billing, legal, DP) as any other business. Therefore, says ARM, a "structural separation" is the only sure way of removing from distribution rates all costs that relate to ESP activities being conducted by the UDCs. However, within the context of the RAP, ARM offers its own estimates of "Retail Energy" Long-Run Marginal Costs (LRMC) and urges the Commission to adopt these in lieu of the LRMCs proposed by the UDCs.

ARM witness Weisenmiller takes issue both with the UDCs' methods of calculating the PX credit, and the spectrum of costs included in the credit. In the 1998 RAP, ESP intervenors claimed that they could not obtain adequate data from the UDCs to construct an accurate alternative PX credit. In this year's RAP, Weisenmiller states that he does not have sufficient information to devise a comprehensive credit, but has developed a "partial" alternative credit incorporating some costs which the UDCs do not include in their credit calculations. Contending that the UDCs' studies are inadequate, ARM asks that the Commission direct the UDCs to undertake "comprehensive" studies that would identify all costs of power procurement.

Weisenmiller briefly analyses the procedures four other states (Connecticut, Georgia, New Jersey and Pennsylvania) have employed to calculate credits for customers who buy electricity from ESPs. He concludes that employing comparable procedures in California would allocate between $3.60 and $10.50 per MWh to the PX credit as a procurement component (pp. 13-18) [by comparison, SCE proposed a credit of $0.02/MWh] He also cites CPUC experience in the gas and telecom areas as support for policies that would reflect "retail costs" in prices. (pp. 8-12)

Weisenmiller says the UDCs' estimates of procurement costs are deficient both because the UDCs did not identify enough cost categories as being related to procurement, and because the UDCs claimed much of the relevant costs to be unavoidable and therefore did not incorporate them as part of marginal costs. He also faults all three UDC studies for not incorporating any shared or common costs, such as A&G.

Weisenmiller critiques each UDC's LRMC study (pp.21-33) and says all three share the fault of treating most procurement costs as unavoidable (non-marginal) due to the UDCs' default service obligations; Weisenmiller says this is erroneous in a LRMC study such as that mandated by the 1998 RAP; a LRMC study should treat all costs as variable.

From his analysis of the UDC filings and other data, Weisenmiller proposes procurement credit estimates of 59¢/MWh for PG&E, 67¢/MWh for Edison, and 65¢/MWh for SDG&E (as noted earlier, though, he says these estimates are not all-inclusive). Customer service and customer accounts expenses appear to be the largest single elements of his estimated credits, followed by costs of load forecasting, bidding and interfacing with the Power Exchange.

3. ORA

ORA witnesses James Price and George Cluff filed testimony on the PX credit issues. On the central issue of estimating power procurement expenses for the UDCs, ORA follows the same line of reasoning as ARM, stating that the UDCs' obligations as default providers should not preclude the full unbundling of procurement costs from distribution rates; like ARM, ORA would have two sets of rates: distribution rates for the costs of true monopoly services, and separate procurement rates to be paid only by bundled service customers.

ORA identified a list of sixteen categories of UDC costs it considers related to procurement activities rather than the "wires" function. The lists includes items such as customer accounts, load bidding to the PX, energy portfolio management, finance costs of PX purchases, and energy-related advertising (ORA, Ch. 6, p. 7). ORA's analysis arrived at estimated procurement rates of 34¢/MWh for Edison and 42¢/MWh for PG&E.

The following table summarizes the LRMC estimates of the parties:

 

Estimates of Procurement Costs (cents/MWh)

     

Company

  
   

PG&E

Edison

SDG&E

 

UDC

0

.02

3

 

ORA

48

40

15

 

ARM

59

67

65

[Comparisons between companies might not be appropriate, because the companies did not use the same calculation procedures.]

4. TURN

TURN's witness Florio touched very briefly on the matter of the PX credit. Florio characterized the UDCs' cost estimates as "absurdly low," and "an almost invisible amount the brokerage fees that this Commission has adopted for the gas utilities, which are analogous in concept to the credits at issue here, are in the range of 26 to 38 cents per MWh on an energy-equivalent basis." (p. 9)

5. UDC Rebuttal

The three UDCs filed extensive testimony in rebuttal to ORA and ARM, and another party, the Coalition of California Utility Employees ("CCUE") filed rebuttal as well.

PG&E witness Pars says ARM is erroneous in claiming that long run marginal cost is the difference between serving all customers and serving no customers. Pars reiterates that marginal cost is defined as the cost of producing one more unit of product, and asserts that most procurement costs do not vary even with substantial changes in output: "if PG&E's purchases from the PX decreased by 75 percent, PG&E would still have to use the same computer software, processes and procedures to perform forecasting and submit demand bids to the PX." (p. VFP-4) Pars says that ARM is, in effect, advocating a fully-allocated costing approach and calling it marginal cost (p. VFP-6). She adds that most of the added categories of costs proposed for inclusion in the PX credit by ARM are inappropriate because they do not vary with the quantity of power PG&E buys from the PX (p. VFP-7); also, Pars says that PG&E customer account managers do not perform procurement-related functions, and that only a tiny fraction of customer-service calls has to do with direct-access matters.

Pars says the ORA recommendations to create separate rates for the competitive procurement default service functions are too vague and are based on inappropriate cost data. (pp. VFP-11 - VFP-16).

In Edison's rebuttal to ARM and ORA, witness Jazieri asserts that the UDCs' default provider role necessarily makes a UDC very different from an ESP, which has no obligation to serve all customers; therefore, the UDC's costs will necessarily be higher than the ESP's (pp.9-10). And Edison disputes the comparability of the gas, telephone and out-of-state electric company margins cited by ARM (pp.11-13).

As did PG&E, Edison says that ARM has misstated the nature of marginal cost, and in fact is using a fully-allocated cost analysis rather than LRMC; Edison says that most of its procurement costs do not vary with the quantity of power it buys from the PX and, therefore, these costs should not be included in a marginal cost analysis.

Edison says ARM and ORA arrived at their proposed PX procurement credits by including non-marginal costs, as well as costs that do not relate to energy procurement at all, e.g. legal and regulatory expenses (pp. 25-35).

SDG&E's rebuttal sounds the same themes as PG&E's and Edison's. SDG&E witness Croyle says that ARM has not actually developed an "implementable" definition of LRMC and instead is using a fully-allocated cost analysis. Croyle also says ORA witness Price has made a incomplete analysis, arriving at a total procurement rate, but not taking the final step of allocating part of it to bundled customers. (p. 16). Therefore, Croyle says, Price's analysis so far as it goes, is not inconsistent with SDG&E's. However, Croyle says ORA witness Cluff has developed arbitrary cost allocation proposals and offered no supporting evidence.

Two other SDG&E witnesses offer more specific criticisms of ARM's cost study. SDG&E witness Garcia says ARM's cost study made a number of mistakes in its analysis of specific SDG&E costs (e.g. using the wrong variable for labor costs). According to Garcia, SDG&E is in general agreement with ORA's method of calculating the commodity procurement rate (p. 2), but differs on details related to 376 costs and certain anticipated expenses under a proposed procurement PBR. (pp. 2-3)

SDG&E witness Osborne examines ARM's analysis of customer service and information expenses, and asserts that ARM based its allocations of these costs upon assumptions rather than actual cost behavior; for example, ARM claimed that 23% of customer service calls were procurement related; Osborne says the correct proportion is less than 1%. Similarly, she says ARM exaggerated the proportion of account manager time spent on procurement-connected activities. Finally, Osborne says customer information expenses (e.g. the costs of processing billing data or changing account information) do not diminish when more customers choose direct access; therefore, these costs should not figure in a marginal cost analysis of procurement expenses.

The Coalition of California Utility Employees (CCUE) also submitted rebuttal testimony in response to ARM and ORA. CCUE's witness Marcus essentially repeats the UDCs' criticisms of ARM and ORA, but in a more concise presentation. Marcus does not do any numerical analysis of his own, but instead points out what he describes as basic flaws in ARM's and ORA's reasoning. Central to his testimony is the contention that ARM witness Weisenmiller's approach to LRMC in fact confuses LRMC with Long-Run Average Cost (LRAC) by incorporating all costs, whether or not they vary with the quantity of power purchased by a UDC. Consequently, such an approach will generate a unit cost greater than the true LRMC.

Marcus also finds fault with Weisenmiller's deriving the "procurement" share of customer service expenses by allocating these expenses on the basis of overall revenue. Marcus points out that this kind of allocation bears no relation to how customer service personnel time is spent. And Marcus claims the examples Weisenmiller cites from other states are not relevant because those most of those rates were not based on LRMCs.

Regarding ORA's testimony, Marcus has minimal criticism of Price, but observes that there is a huge difference between Price's estimates of average procurement costs and Cluff's proposed credits.

Having asserted that ARM's and ORA's proposed credits are far above LRMC, Marcus says that if such credits were adopted, UDCs' revenue losses from customers switching to direct access service would exceed any cost savings. Hence, he says, UDCs would be given the incentive to dissuade customers from taking direct access service, contrary to the Commission's goal of greater competition in the state's electric industry.

C. Discussion

6. The Economic Arguments Of The Utilities Are Flawed

In considering the utilities' assessment of their procurement costs, one fact stands out: all three utilities have claimed that the long-run marginal costs ("LRMCs") of procurement are zero. While the utilities may understandably want to understate their procurement costs in this proceeding, 5 the claim that LRMCs are zero is not credible. This result can be traced to several flawed economic arguments advanced by the utilities. First, they improperly try to present short-run methodologies as a long-run analysis. Second, they incorrectly define and overstate the importance of their default provider obligations. Finally, they have relied on a rigidly simplistic and impractical definition of "marginal cost."

7. The UDCs are Confusing the Definitions of "Short-Run" and "Long-Run" with Respect to Marginal Costs.

ARM and ORA have applied a definition of "long-run" consistent with economic theory and Commission precedent, namely that "long-run" implies that "all aspects of the economic equation can be changed, including fixed assets (plant), fixed obligations under contracts, and all variable inputs." 6 The UDCs have espoused the same definition, but their analysis has consistently betrayed it by applying what in effect are "short-run" arguments.

All three utilities claim that their procurement costs are fixed, that they in fact do not vary significantly given any change in the level of demand. For example, asked if there would be no significant reduction in procurement costs if SDG&E's "procurement requirement declined by 75 percent," SDG&E witness Croyle replied, "That's generally going to be true, yes." 7 PG&E, in its Opening Brief, maintains that its "surveys and interviews produced evidence that PG&E's procurement-related costs don't change as a result of a small change in either 1) the volume of power PG&E purchases from the PX or 2) the number of customers relying on PG&E to meet their electric procurement needs." 8 SCE witness Vail stated:

Q Mr. Vail, suppose SCE lost one-third of its bundled service customers to ESPs tomorrow, would its cost of procuring energy be affected?

A No.

Q Why not?

A The process of procuring energy involves estimating and bidding the load of the aggregate of our customers. 9

But in this case the phrasing of the first question, posed by SCE's own counsel, betrays a fundamental misconception held by all three utilities. The question is not whether the "cost of procuring energy" would be affected "tomorrow" given a change in demand, but whether in the long-run-given the opportunity to optimize business operations and vary all inputs to production - costs would be different.

In its Opening Brief, SCE cites the testimony of CUE witness Marcus, who presented what SCE claims is "an excellent analogy which explains why the marginal cost of energy procurement is zero." 10 The analogy involves lighting for a classroom with 20 students. Witness Marcus noted that if two students moved away and the class retained only 18 students, "[the] lighting bill wouldn't change. I'd still turn on the lights in the room." 11 In this case, he argued, the marginal cost of lighting is zero. Unfortunately, this is a short-run analysis. In the long-run, one must assume that "all aspects of the economic equation can be changed." In this analogy, this means that as the class size changes, we must consider how costs would change given every possible reconfiguration or re-sizing of the room and its equipment to optimize lighting for the remaining students. Lighting costs would only remain fixed in the short-run.

This type of confusion about short- versus long-run costs discredits much of the three utilities' analysis. It casts doubt, for example, on the credibility of the surveys used to interview employees about potential changes in procurement costs. The greater concern about the surveys is whether they were at all useful for estimating long-run marginal costs.

PG&E's surveys, for example, asked employees how "labor hours" and "capital or expense related" costs might change given 10, 50, and 100 percent reductions in the number of bundled customers. 12 But they do not explicitly ask how costs might change given total flexibility to vary these inputs, a possibility which, as in the classroom lighting analogy, might not be immediately obvious. There is no way to tell whether respondents thought in terms of short-run scenarios, or managed to entertain a hypothetical long-run where all inputs could be varied and optimized. As Edison and SDG&E were not forthcoming with their survey instruments in this proceeding, it is entirely possible they suffer from the same deficiency.

Indeed, the notion that in the long-run all costs are fixed, and that the LRMC of procurement is equal to zero is not credible. While some procurement activities may by their nature seem independent of total load or customers, it is unreasonable to expect that savings are not achievable across all procurement functions, including for example, the activities of customer account managers and customer service representatives.

Contrary to SCE's claims in its Opening Brief, 13 ARM and ORA presented evidence in this proceeding strongly suggesting that LRMCs cannot be equal to zero. The procurement budgets of SCE and PG&E are between five and ten times higher than SDG&E's, even though SCE and PG&E both have about four times the customer base of SDG&E. 14 If LRMCs were truly equal to zero, then these procurement budgets should be roughly equivalent. Such large discrepancies are unlikely to be the result of accounting differences. Either there are serious problems with the data submitted by the utilities in this proceeding, or one has to conclude that increases and decreases in the number of customers cause procurement costs to go up and down over the long-run.

In their direct testimony, the utilities advocated a PX credit based on the short-run marginal cost of procurement, 15 despite clear direction from the Commission to use LRMCs. In claiming that substantial portions-indeed, all-of their procurement costs are fixed, the utilities are still clinging to a short-run conception of marginal costs. We must reject LRMC estimates based on what are in fact short-run analyses.

8. The UDCs Misconstrue their Default Provider Obligation and Incorrectly Assume it Causes Procurement Costs to be Fixed in the Long-Run.

The most problemmatic theme of the utilities is that they cannot exit the procurement market because of their default provider obligation, and therefore must incur procurement costs even if 100 percent of their customers choose direct access. 16 In other words, their default provider status effectively causes certain procurement costs to be fixed and unavoidable even in the long-run.

This reasoning is fallacious in several ways. First, as explained above, when looking at long-run marginal costs, one must assume that "all aspects of the economic equation can be changed," including among other things, "fixed obligations under contracts." 17 There is no dictum in either economic theory or Commission precedent stating that certain obligations cause costs to be fixed in the long-run. From a long-run perspective, the default provider obligation is no different from "obligations under contracts."

More specifically, it is not correct to assume that the utilities' default provider status will always be required, and thus that certain procurement costs are in the long-run "unavoidable." In fact, the Commission has already announced that it intends to reconsider the "role of the utility" in the distributed generation proceeding (Rulemaking 98-12-015). Further, the Commission noted in the 1998 RAP proceeding that "[f]uture rate design should recognize changes which might occur with regard to the utilities' obligation to serve." 18 While the default provider role of the utilities may or may not change as a result of future proceedings or Legislation, we do not necessarily view the default provider obligation as fixed and immutable. When conducting a long-run analysis of marginal costs, the utilities are incorrect to present certain costs as unavoidable because of this obligation.

Moreover, the default provider obligation has played no role in determining LRMCs for the utilities' distribution services. In fact, SCE Witness Glenn Vail stated that in calculating the marginal costs associated with transmission and distribution, far from considering its default service provider role, SCE will "hypothesize the competitive market, let us say, T&D, and determine what these components would cost in terms of marginal cost pricing." 19 Just as SCE "hypothesizes" a competitive market when it calculates the marginal costs associated with its T&D system, one needs to "hypothesize" a competitive market, and assume free entry and exit, when calculating the marginal costs associated with retail procurement.

Some parties have argued that regardless of how one treats fixed costs in the long-run, the default provider obligation nevertheless imposes a cost that should be paid by all customers. While we agree in concept, the question whether this cost is at all significant. In this proceeding the utilities have continually mischaracterized or misconstrued the nature of default service. SCE, for example, claims that because of the default provider obligation, utilities "must be ready to provide basic commodity service to all customers at all times." 20 This is simply not true. It would imply that the utilities must always incur the costs of serving all customers regardless of how many had chosen direct access-even if 100 percent chose direct access. As noted by the Department of General Services (DGS), "reason itself indicates that the UDCs' obligation merely requires a reasonable capability to service such customers that may seek bundled service." 21

Default provider status "acts like an `insurance policy'" for ratepayers, as utilities claim. 22 But just as insurance companies do not have to retain funds sufficient to compensate losses incurred by all of their customers at once, the utilities do not have to maintain the capacity to provide commodity service to "all customers at all times." SDG&E witness Croyle acknowledged as much when he stated:

What we're saying is that how we handle that default provider obligation does not necessarily require us to retain 100 percent of the resources that -- as if we were providing the service to all customers. 23

The question is how large the default provider obligation really is. As the ALJ observed, the cost of the obligation appears to be "minimal."24

9. The UDCs are Adhering to an Overly Simplistic and Impractical Definition of Marginal Cost.

Both PG&E and SCE concede that certain of their departments would accrue savings if they exited the procurement business altogether. 25 From a long-run perspective, exiting the procurement business is a very real prospect, utility arguments about their default provider obligations notwithstanding. The challenge in this proceeding is how to account for these kinds of savings that do not occur strictly "on the margin," but are nevertheless the result of avoiding incremental, non-fixed procurement costs.

The answer is to employ a method of calculating marginal costs endorsed by SDG&E and whose validity has been implicitly or explicitly acknowledged by every party in this proceeding. That method is to measure the increment in costs between serving no customers and serving all customers. Some parties have characterized this method as producing an "average" instead of a "marginal" cost, but it is appropriate for estimating marginal costs when inputs are "bulky" (i.e., when a single increment serves many customers) and their cost insensitive to changes in demand over large increments. "Bulky" inputs are not the same as "fixed" inputs from a long-run perspective.

Both PG&E and Edison have presented arguments during this proceeding that acknowledge the legitimacy of this "large increment" approach. 26 Furthermore, witnesses for SDG&E have explicitly stated their endorsement of this method. SDG&E witness Croyle stated the following:

Q Mr. Croyle, which of these [LRMC calculation methods] are you, in fact, recommending?

A We're recommending the -- the -- the large increment, the long-run marginal cost of serving all customers. 27

SDG&E witness Garcia reiterated this recommendation:

Q Again, is San Diego actually recommending to the Commission the LRMC of serving one more customer or the LRMC of serving all customers?

A Serving all customers. 28

Furthermore, the Commission has adopted the increment between serving no customers and serving all customers as a method for calculating LRMC in other proceedings. In developing costing principles for open access in telecommunications, the Commission decided that "[t]he increment being studied shall be the entire quantity of the service provided, not some small increase in demand." 29 In considering marginal costs for transmission and distribution, which also have "bulky" production inputs, the Commission has consistently found them to be positive and significant. In fact, Edison found in the Post-Transition Rate Design proceeding that its T&D marginal costs exceeded its revenue requirements. 30 Just as the utilities have developed non-zero transmission and distribution marginal costs, it is incumbent upon the Commission to adopt methods that result in non-zero LRMCs for retail procurement.

Both Edison and PG&E argue in their Opening Briefs that ARM and ORA have made proposals that harm bundled customers. Their proposals to have the Commission adopt a PX credit of zero, however-or of some negligible amount "if the Commission believes that additional PX credits are appropriate" 31-will instead harm direct access customers. Indeed, the purpose of this proceeding is to rectify the harm that is being caused to direct access customers as a result of the subsidies already provided by these customers to the utilities' bundled customers. At bottom, the utilities' arguments are anti-competitive because they do not allow bundled customers to obtain accurate credits that might convince them to switch their service to utility competitors.

10. The Commission Should Reject the Approach of the Utilities

The utilities, while proposing slightly different conceptual approaches to approximating the procurement aspect of the PX credit, each fail to provide the Commission with a viable approach that meets our objectives. The utilities' methods are not sufficiently similar to long-run marginal cost analysis. The utilities overstate the importance of and the implications of the default provider role. The utilities ultimately propose methodologies which will result in PX credits for procurement activities that are too low, and will harm the competitive marketplace.

PG&E's argument that any PX procurement credit greater than zero subsidizes ESPs ignores the Commission's findings in the 1998 RAP Decision and D.00-02-046. PG&E disregards the fact that the "existing PX credit" for procurement-related activities is zero and that adopting a PX credit of zero is inconsistent with the 1998 RAP decision and D.00-02-046. In the 1998 RAP decision, the Commission directed the "utilities to include the long-run marginal costs of these functions in future calculations of the PX credit, that is, in the utilities' 1999 RAP applications." (p. 23). Had it been our intent to adopt a PX procurement credit of zero, we would have said so in the 1998 RAP decision, saving the utilities and intervening parties the time and expense of the participating in present proceeding. PG&E's argument that its costs do not vary because of its default service role overlooks the fact that long-run marginal costs typically do not take into account the default service role. The real costs of default service are likely to be minimal.

For SCE and SDG&E, we also find that the methodologies made by these utilities do not meet our definition of long-run marginal cost. We also note that a number of categories of costs are missing or are well below reasonable levels. SDG&E's conceptual approach is the best of the three utilities. However, SDG&E did not follow through from its conceptual approach to its specific recommendaions. In reality, SDG&E's approach in theory is more closely effectuated by the ORA and ARM methodologies.

ORA and ARM propose methods that, while perhaps imperfect, adhere more closely to previous Commission Decisions and our specific directions in D.99-06-058. With our rejection of the recommendations of the utilities, we are left with the recommendations of ORA and ARM. These parties do not agree on all figures, but both analyze the utilities' costs in similar ways and based on similar premises. We accept the underlying premises of both ORA and ARM that all costs are variable in the long-run, that the default provider role of the utility does not preclude costs from being considered variable, and that procurement costs should be borne by the bundled customers who are being served by the utilities and not the direct access customers who are not. In the next section, we discuss the analytical approaches to determining the specific credit levels for each utility. In that section we will focus only on the ORA and ARM methodologies, having now rejected the approaches of the utilities.

D. Analytical Approaches

ARM's Analytical Approach

Because of the difficulty in unraveling the specific elements of Retail Energy cost, ARM believes that the utilities must functionally separate their Retail Energy function from their distribution function, so that these functions deal with each other on an arms length basis. The utilities should then be required to file separate, unbundled, bottoms-up distribution and retail energy tariffs that will establish how their costs will be recovered for both these functions in the future. This should be done immediately for SDG&E because it has completed its rate freeze period and should be done by both PG&E and SCE in time for the beginning of their respective post transition periods. In the meantime, ARM recommends the Commission adopt its LRMC Retail Energy cost estimates.

ARM believes the key is to develop a retail price for a "Utility-Operated Energy Service Provider" (UOESP). Basically, the Commission should be seeking a retail price that allows the customer to compare fairly an ESP's price for electricity with the UDC's price (PX Charge/PX Credit). In this way, the customer has full, transparent information and can intelligently shop the market. Embedded in each of the utilities is a UOESP which serves approximately 98% of the customers in their service areas. For customers to make a fair comparison between ESP and UDC electricity prices, they need to see the costs of each utility's ESP as if it were a stand alone ESP business, operating with its own profit and loss statement.

The UOESP has the exact same cost areas that any other ESP would have. These costs include wholesale procurement, Customer Service Functions, various corporate support functions, and various general and administrative costs. For example, it must have human resources professionals to develop salary and vacation policies, recruit staff, and administer benefit programs. It must have payroll staff and systems for paying its staff and accounting staff and systems for paying its bills and keeping its books. It must have information technology staff to devise computer systems, develop communication systems, run UOESP internal and external web sites and ensure Y2K compliance for itself and its vendors. The UOESP must have educational materials for customers and public relations staff to notify interested parties about its activities or address public issues that arise. It must have a staff of attorneys and regulatory specialists to file advice letters, engage in cases and respond to complaints regarding bundled rates. It needs executive staff to determine company strategy, manage overall finance and operations, and provide customer service. It must hire consultants to assist in areas like corporate strategy and operations, risk management, and market analysis. The UOESP must have cars, computers, telecommunications systems, telephones, pagers, personal communication devices, and furnished offices. These are only examples and do not represent a comprehensive assessment of all UOESP costs. None of these UOESP costs are currently included in the PX Credit.

Default service is what a customer gets if it does not make a choice to purchase power from an alternate supplier. The number of bundled service customers a UDC has to serve is naturally affected by these individual customer decisions. As the default supplier, a UDC gets a steady stream of new customers every month, as people and businesses move into the service territory for the first time. Unless these customers are direct access savvy, they will automatically "default" to bundled UDC service. In addition, customers who have already chosen direct access, but move to a new location within the UDC service territory are likely to "default" back to bundled UDC service for a period of time before they can return to direct access. In addition, the UDC welcomes back individual customers who have chosen a direct access provider but who have subsequently changed their mind and wish to return to the utility. Finally, UDCs also must serve direct access customers that an ESP has decided no longer to serve.

In considering the credit for all utilities, ARM also identified additional costs that should be included in the UDCs' PX credits, including 1) legal and regulatory costs; 2) block forward market financing costs; and 3) advertising costs, among numerous others. Testimony filed in this proceeding has clearly demonstrated that there are many costs, above and beyond those acknowledged by the UDCs, that should be considered for calculation of the PX Credit. 32 ARM's recommendations are based on the analysis of what it sees as the UOESP's costs. ARM makes the assumption that the utility need not provide service to all customers under current rules, and that default service does not necessarily mean that the utility must stand ready to serve all 100% of customers at any point in time.

ORA's Analytical Approach

ORA suggested that the Commission establish "a standard definition of pertinent functions," after which the utilities must then "record their costs in a way that is directly translatable to the Commission's standard definition of procurement functions, so that audits can be conducted to verify that all pertinent costs have been recorded in this transparent manner." 33 In an attempt to facilitate a sufficiently detailed identification of the tasks that may now comprise commodity procurement, ORA analyzed the cost categories discussed by the utilities, found a suitably-detailed listing of functions in SDG&E's testimony in this proceeding, and supplemented SDG&E's list using testimony presented by parties in past proceedings. ORA identified the following items as procurement-related functions rather than related to wires company functions or market facilitation: 34 (ORA, Ex. 49, p. 6-7.)

a. Load bidding to PX, including daily, hourly, and near real-time (e.g., 10 minute) load forecasts to define power needs

b. Energy portfolio management of energy, ancillary services, and real-time markets, including block-forward, day-ahead, day-of, and 10 minute markets, including self-provision of ancillary services

c. Use of load curtailment, including or spinning reserve equivalent, to lower costs

d. Managing existing purchase contracts

e. Purchase and management of Firm Transmission Rights (FTRs) contracts

f. Duties of Customer Account Managers and Customer Service Representatives

g. Financing costs for purchasing power from PX

h. Processing settlement statements

i. Processing payments to PX

j. Filing settlement disputes

k. Meter reporting for load

l. Calculating schedule PX charges

m. Interfacing with PX and ISO

n. Programming for commodity service

o. Capital lease system

p. Energy-related advertising

We agree that this list of functions is appropriate for development of SDG&E's procurement adder to the PX credit. Although the listing above was useful for analysis of SDG&E's costs, PG&E and SCE apparently do not record or forecast costs in the listed categories. However, SCE provided an alternative listing of functions that allows its procurement costs to be analyzed:

ORA accepts this list of functions as being appropriate for SCE, and we find it useful as well. While SCE's definition of these activities is useful, it has the result of a comparable analysis for SCE and SDG&E not being possible due to the differences in the identification of pertinent costs. Nevertheless, the functions for each are comparable enough to allow consideration of credits based on their analysis. ORA was unable to separate PG&E's procurement costs from a multitude of other costs. Given this lack of usable data for PG&E, ORA recommends a proxy for PG&E's procurement rate. Because PG&E and SCE operate in the same statewide market and are of similar size, ORA recommends that SCE's total procurement-related revenue should be divided by PG&E's annual sales, to produce a recommended total procurement rate for PG&E. If PG&E-specific data is to be used instead of the proxy recommended by ORA, ORA supports ARM's analysis because it reflects the extensive record considered in PG&E's recent General Rate Case.

We will not accept ORA's recommendation for PG&E. We strongly prefer to adopt credits that reflect the specific costs of the individual companies; ORA was unable to perform such an analysis. However, ARM did just that. Below, we will consider ARM's analysis of PG&E's costs in detail.

The difficulties encountered by parties other than the utilities in determining their recommended rate adjustments and fully supporting their showings affect the breakdown of the overall costs that are examined in ARM's and ORA's testimony into marginal costs versus other cost components, and do not warrant rejection of recommendations other than the utilities' own proposals. However, the imperfections of the data require us to scrutinize the ORA and ARM proposals carefully in order to determine the credit levels.

This proceeding cannot be considered to have said the last word on the establishment of a rate component for commodity procurement, since issues under consideration in other proceedings will not have been decided in time for inclusion in the rates being set in this proceeding. First, SCE's request to present an alternative basis for the procurement rate35 may warrant consideration, since long-run incremental cost-based ratemaking may be consistent with the outcome of future proceedings. It would be useful; to promote consistent analyses by all utilities, to specify the cost categories to be used, such as the categories that SCE presented in its data response to ORA. In addition, proposals that are being considered in the Revenue Cycle Services/ Direct Access Service Fee (RCS/DASF) proceeding may establish charges that apply to customers who are served on bundled utility service. If such charges were established, the total revenue responsibility for bundled service would need to consider the utility's costs of enrolling and serving customers on a bundled-service status. The RCS/DASF proceeding may also affect the allocation of shared and common costs, which may also be involved in parties' proposals in this proceeding. Assuming that the commodity procurement rate that results from this proceeding is not altered again in the RCS/DASF proceeding, the Commission should allow parties to the next RAP proceeding (or its successor) to propose adjustments to the commodity procurement rate, to reflect the outcome of the RCS/DASF proceeding.

Allowing these potential issues to be considered in future proceedings will be necessary to ensure reflection in rates of the costs of requiring the utility, in its role as a provider of competitive services, to interact with other parts of the utility through same mechanisms as other ESPs. One aspect of requiring comparable treatment between ESPs and the utility's provision of services that are subject to competition can be adopted in this proceeding: as the joint parties have recommended in the proposed SDG&E commodity-procurement settlement in the PTR proceeding, the utilities should be required to publish their hourly forecasts of delivery service area load, which are used to produce distribution loss factors. These forecasts may have a variety of uses within distribution operations and producing them is appropriately treated as a "wires company" function, but allowing different accessibility to different providers of commodity service (i.e., the utilities' scheduling staff vs. other ESPs) would not promote a competitive market. Posting these forecasts simultaneously with the posting of distribution loss factors can provide additional information to assist market participants in the market, such as insight as to whether the distribution company expects unusual load conditions. There should be no expectation that, by providing this forecast of delivery service area load, the utilities have made representations regarding the accuracy of these forecasts or should be held liable in connection with any entity's use of or reliance on these forecasts.

ORA proposes a methodology called the Electric Service Provider Customer Credit (ESPCC) model. ORA's methodology for the utilities costs of being market facilitators for direct access assumes that direct access customers are the primary beneficiaries of the direct access program. Accordingly, these costs are allocated to classes based on the probability that customers in a class will chose direct access service. These cost further benefit the utility in giving them a role in the direct access market that helps them market their image, and as such have a strategic marketing component.

The key features of ORA's ESPCC methodology include the following:

ORA requests that the CPUC adopt the ESPCC Methodology and Initial Values as an appropriate next step in the evolution of a competitive California Electric Industry. ORA observes that the ESPCC is an approximation to the long-run incremental cost associated with the IOU's strategic marketing activities for all IOU services. In turn, the long-run incremental costs for these activities can be readily converted into per customer credit for those IOU customers selecting the Direct Access Option. Until the CPUC adopts a long-run incremental cost methodology for the IOUs, the ESPCC can serve as a transition methodology for this key embryonic stage of evolution for California's Electric Industry.

E. Adopted Analytical Methodology

We have already rejected the utilities' arguments that the long-run marginal costs of procurement are zero or near zero. The utilities in reality conducted short-run marginal cost analyses and generally did not consider how its factors of production would change in the long-run if it served fewer customers.

Both ORA and ARM presented much more complete and appropriate studies. ORA's list of cost categories presents a good basis for analysis of procurement costs. The utilities criticize ORA and ARM for certain imperfections in their methodologies. However, due to the lack of responsiveness by or information from utilities, ORA's overall methodology provided conservative figures by leaving out certain costs that likely belong within the PX credit. While ARM's studies provide a good basis for a determination of the credit and could be adopted, we believe ORA's detailed and comprehensive studies provide a slightly better basis for establishing the credit. To the extent that ORA or ARM are incorrect and have overstated costs in any way, accepting ORA's lower figures protects both ratepayers and utilities from error.

ORA needed to recommend a proxy for PG&E's procurement rate. Because PG&E and SCE operate in the same statewide market and are of similar size, ORA recommended that SCE's total procurement-related revenue should be divided by PG&E's annual sales, to produce a recommended total procurement rate for PG&E. If PG&E-specific data is to be used instead of the proxy recommended by ORA, ORA supports ARM's analysis because it reflects the extensive record considered in PG&E's recent General Rate Case.

We do not wish to use SCE data to calculate PG&E's PX credit. We have an extensive record from ARM analyzing PG&E's costs, and we shall use that record.

PG&E disagrees with ARM's analysis in a fundamental way. However, we have rejected PG&E overall analysis. PG&E also provides comments on ARM's study in three specific cost categories.

PG&E argues that the costs not currently recovered in rates should not be part of the PX credit. PG&E cites three specific costs under this category. They are gas and electric supply expense36, working cashing, and incremental restructuring implementation costs appropriately collected per P.V. Code Section 376 ("376 costs"). PG&E believes ARM has improperly included these costs in the PX credit.

Our goal for the PX credit was clearly articulated in the 1998 RAP decision (D.99-06-058):.

"If we are to promote competition in generation markets, utility commodity prices must ultimately recognize those costs which the utilities must recover in the long run as any other provider. Our long term strategy is to create an industry structure in which utilities are one of many competitors. As part of that strategy utility pricing must eliminate any "competitive advantage by an institutionalized removal of costs otherwise intrinsic to the provision of a service," as DGS' comments describe the utilities proposals."(D.99-06-058, pp23-24)

To implement this strategy, we ordered the utilities to file LRMC studies.

"PG&E, SDG&E, and Edison shall include in their respective 1999 revenue allocation proceeding (RAP) a PX credit that reflects the long run marginal costs of customer account managers, customer service representatives, self-provision of ancillary services and financing costs for purchasing power from the PX. The PX calculation should also include an estimate of other expected long run marginal costs as set forth herein." (Id. p49, OP4, emphasis added)

Therefore, the relevant criterion is all the expected long run marginal costs, not whether a specific cost is in rates. We agree with ORA and Aglet, whether a particular cost is currently in rates is generally not pertinent. The issue is whether the cost is incurred in the procurement process. As Aglet pointed out in its opening brief, future test year ratemaking as it is currently practiced does not require a utility to incur specific costs and does not authorize a utility to achieve specific revenues to recover those costs. Between test years, it is inevitable that the utility will incur some costs that were not anticipated and not incur some costs that were expected. Furthermore, it is possible for an utility to recover costs without them being in rates. For example, Externally Managed Electric Restructuring cost is recovered by PG&E through monthly debit entries to its Transition Revenue Account (TRA) in accordance with the settlement adopted in D.99-05-031.37

Costs associated with gas and electric supply function are relevant to commodity procurement. D.00-02-046 adopted a total revenue requirement of $10.012 million for both expense and capital associated with this category. This amount was not included in distribution rates because these functions are related to generation, not distribution. The decision further concluded that" [a]t least in the significant part they [functions under gas and electric supply category] are the same functions provided by scheduling coordinators and energy service providers on behalf of direct access customers."38 The GRC decision also stated that at that time, it would not determine the appropriate mechanism for recovery but would" entertain proposals by PG&E for recovery of these expenses from customers on whose behalf they are performed."39 It is appropriate to include electric supply costs in the PX credit because they are directly applicable to the procurement function. However, it is not clear that gas supply costs should be included in the electric credit.

Similar to its argument regarding gas and electric supply costs, PG&E contends incremental 376 costs should not be part of the PX credit because these costs are not in existing rates. PG&E withdrew these costs from its 1999 GRC request but subsequently filed an Electric Restructuring Costs Account (ERCA) application to recover them. PG&E also has not demonstrated where ARM has included these costs in ARM's estimates. It is possible that incremental 376 costs are improperly included in ARM's estimate.

PG&E notes that its 1999 GRC decision adopted a negative revenue requirement for working cash, implying this cost is not included in rates because it is negative. Working cash is composed of two major components: operational working cash requirements less amounts not supplied by investors and funds needed to pay expenses in advance of collecting revenues (the lag component). The net effect of these two components as adopted in D.00-02-046 is a negative revenue requirement. The lag component of the adopted working cash revenue requirement, which is what is reflected in ARM's PX credit calculation, is positive. ARM's estimate of working cash due to lag time is actually less than what was adopted in the GRC. Therefore it is appropriate to include working cash in the credit.

There being questions based on PG&E's responses to modify ARM's analysis, we will not adopt ARM's figures for the procurement adder to the PX credit. Instead, we will continue our conservative approach and adopt ORA's credit recommendation as a proxy for an adjustment to ARM's recommendation.

In summary, we believe the appropriate credit is defined by ORA's figures for each utility.

However, we will not adopt these credits. Instead, we adopt a single credit for all three utilities. The advantage of a single credit approach is that ESPs - which likely do not themselves have different procurement costs in different utility territories -- will face similar circumstances when soliciting customers, and not have to design different approaches for each company. Also, there will not be a movement of ESPs to or away from customers in one utility or another.

The credit level will be at the average of the three credit levels discussed above, specifically 34 cents per MWH, or 0.034 cents per kWh.

III. RMR

RMR generation is generation the Independent System Operator (ISO) determines is required to maintain a reliable transmission system, including generation to meet reliability criteria, load demand in constrained areas, and voltage and security support needs. Before the Commission initiated restructuring of the California electric industry with D.95-12-063, energy users paid the costs of those same transmission support functions. The vertically integrated utilities used generation resources as a substitute for certain transmission facilities because generation and transmission could be planned and operated on a coordinated basis. The costs of such uneconomic dispatch of generating units for transmission reliability purposes were reflected in increases in the costs of energy.

After restructuring, with generation participating in markets rather than being subjected to cost-based regulation, it was necessary to mitigate the market power generating units might otherwise exert when needed for reliability purposes. The ISO designated RMR units and executed FERC-approved RMR contracts with generators to provide the reliability services that are now necessary due to the restructuring of the electric service industry in California.

Under FERC-jurisdictional tariffs, the ISO bills the utility for the costs of RMR units in the utility's service area. The ISO's tariff requires the utility to pay RMR costs invoiced to it by the ISO -- "Each Responsible Utility shall pay the amount due under each Responsible invoice by the due date specified in the Responsible Utility invoice...." (ISO tariff § 5.2.7.)

The utility's recovery of the RMR invoices it receives are governed by the FERC-jurisdictional Transmission Owner (TO) Tariff which specifies that the RMR costs charged to the utility by the ISO are to be recovered from "End Users": "Must-run contract costs payable by a utility that is a Participating TO pursuant to Section 5.2.7 of the ISO Tariff shall be recovered from End-Users located in the Service Area of that utility." (TO tariff § 15.) End-User is synonymous with retail customer -- "A purchaser of electric power who purchases such power to satisfy a Load directly connected to the ISO Controlled Grid or to a Distribution System and who does not resell the power." (ISO Tariff Master Definitions Supplement.)

Under the TO tariff, the particular retail end-use customers from whom Edison is authorized to collect the RMR costs, are those located in its service area:


Service Area: An area in which, as of December 20, 1995, an Investor-Owned Utility (IOU) or a Local Publicly Owned Electric Utility was obligated to provide electric service to End-Use Customers. (Exh. 70.)

As of December 20, 1995, Edison provided electric service to retail end-use customers in its service territory, but it did not provide electric service to the retail customers inside areas served by municipal utilities. However, Edison provided transmission service to the municipal utilities.

Edison claims it is authorized by this Commission to fully recover from Edison's own retail customers the RMR costs billed by the ISO to participating transmission owners in two decisions - D.97-12-109 and D.98-04-019:


"The Commission should grant the petition to modify D.97-08-056 filed by Edison with regard to must-run costs to the extent it would account for the costs in the TRA for purposes of calculating "headroom."40

Edison currently allocates 100% of RMR costs to retail customers, even though both retail and wholesale customers benefit from system wide voltage support and the mitigation of thermal loading of transmission facilities resulting from RMR generating unit reliability. ORA recommends that Edison promptly file at FERC proposals to recover RMR costs from its wholesale customers and in the event it fails to file, ORA recommends that the Commission impute an allocation of RMR costs to wholesale customers by limiting the level of RMR costs recorded in the TRA.41

Prior to D.98-04-019, the Participating TOs (PTOs) had filed at FERC TO tariffs which included a section relating to the "Recovery of Must-Run Contract Cost." TO tariff § 15 states:


"Must run contract costs payable by a utility that is a participating TO pursuant to Section 5.2.7 of the ISO Tariff shall be recovered from End Users located in the Service Area of that utility. Such utility shall file with the Commission and/or the appropriate Local Regulatory Authorit(ies) a mechanism for such cost recovery."

ORA argues that the plain text of TO tariff § 15 states that RMR costs will be recovered from end-users, but most importantly, it does not state that they will be exclusively or as Edison asserts "strictly" recovered from retail transmission customers. The TO tariff further establishes the concurrent jurisdiction of FERC and/or the CPUC to determine the mechanism for RMR attributable cost recovery. Given what the TO tariff does not say, the express language of the TO tariff and the PTO's voluntary election to seek recovery of RMR costs at the CPUC, ORA contends that it is baseless for Edison to claim that the CPUC does not have regulatory jurisdiction over the TO tariff when in fact the TO tariff enables the CPUC to establish a mechanism for RMR attributable cost recovery if the PTOs so avail themselves of CPUC jurisdiction.

ORA believes Pub. Util. Code § 451 is applicable to this issue. Section 451 states that "[e]very unjust or unreasonable charge demanded or received for such product or commodity or service is unlawful." ORA claims the exemption of wholesale transmission customers from paying a fair share of RMR costs results in retail transmission customers footing the entire RMR bill. Because retail customers are subsidizing wholesale customers, the present RMR cost recovery mechanism is unjust and unreasonable. In addition, Pub. Util. Code § 453 states that "[n]o public utility shall establish or maintain any unreasonable difference as to rates, charges, service, facilities, or in any other respect ... between classes of service." ORA asserts that by not paying anything toward RMR costs, Edison's wholesale customers receive unduly discriminatory and preferential rate treatment contrary to the proscriptions of § 453. Similarly, §§ 205 and 206 of the Federal Power Act, 16 U.S.C. §§ 824(d) and 824(e), prohibit unjust and unreasonable rates and unduly discriminatory or preferential rates

ORA recommends that the Commission require Edison to immediately file proposals at FERC to recover a fair allocation of RMR costs to its wholesale transmission customers, to apply to the remainder of the utility's rate freeze period.42 If Edison does not make such a filing, ORA recommends that the Commission should impute an allocation that represents a proxy for what a FERC-adopted RMR cost allocation may achieve. This imputed allocation would be used to limit the amount of RMR costs debited to the TRA for ultimate recovery from retail customers. Under ORA's recommendation the imputed allocation would limit Edison's recovery of future RMR costs to 87% of the total paid to the ISO and be applicable for the remainder of Edison's rate freeze period.

Edison claims that ORA's recommendations were previously raised by ORA and rejected by the Commission in D.97-12-109, in A.96-12-019. Edison reminds us that in A.96-12-019, ORA argued before this Commission that,


"The beneficiaries of must-run generation in Edison's transmission service area may include wholesale customers as well as retail customers, but recovery of the ISO's must-run billings through retail rates would place the entire burden of these costs on retail customers. The Commission should not accept this result without further analysis."

Edison responded that,


"However, the FERC Transmission Owners (TO) Tariff states that these costs '. . . shall be recovered from end-users located in the Service Area of that utility . . . .' (Revised Pro Forma Transmission Owners Tariff Section 15, August 15, 1997.) Wholesale customers are not end-users and will therefore not pay these costs."

However, when we issued D.97-12-109 granting permission to recover RMR costs from retail customers, we did not explicitly address this issue. Therefore, there is no basis for citing this discussion as a precedent.

Edison argues that both FERC and the federal courts have recognized that rate filings at FERC are the utility's to make, and the state regulatory authority cannot legally order the utility to make a particular rate filing at FERC. (Western Mass. Electric Co., 23 FERC ¶61,025 (1983), affirmed Mass. Dept. of Pub. Util. V. U.S., 72 F.2d 886 (1st Cir. 1984).) In addition, Edison contends that because FERC has set the wholesale rate that the ISO charged to Edison, and because FERC has assigned 100% of such costs to Edison's retail customers, this Commission cannot legally conclude that the FERC-jurisdictional rate is unreasonable nor can it change the FERC-authorized allocation. Further, Edison maintains that the "imputed" allocation to wholesale customers and accompanying limit on costs recovered from retail customers proposed by ORA would result in a disallowance of FERC-jurisdictional costs. Thus, Edison asserts that a Commission decision implementing ORA's recommendation would violate the filed-rate doctrine established by the U.S. Supreme Court. (See Nantahala Power & Light Co. v. Thornburg, 476 U.S. 953, 966 (1986).)

However, this Commission does have jurisdiction over the costs Edison can recover from retail ratepayers. We are not convinced that these issues have been fully explained or litigated at FERC. FERC does set the RMR rates which are then charged to Edison, but FERC did not decide the RMR rates Edison could charge to others. In fact, there are no FERC-filed rates for which Edison can invoke the filed rate doctrine. Edison recovers its RMR costs (i.e., the charges assessed to Edison by FERC-imposed rates) by filing for recovery at this Commission. Therefore, we have jurisdiction over the costs Edison can reasonably collect from retail ratepayers. We can impute the amount of revenue that Edison could seek to recover from its wholesale customers. (Rochester Gas & Electric Corp. v. Public Service Commission of State of New York, 754 F.2d 99 (1985).)

After a careful review of the facts presented in this proceeding, we are of the opinion that to exempt wholesale customers from paying their fair share of RMR costs is to give them a free ride to avoid paying for benefits received. The ISO has determined that certain generating units require RMR designation in order to ensure the reliability of the transmission system in a utilities service territory. Yet, while RMR units benefit the entire transmission system (given the interconnected nature of the transmission system all transactions become mutually interdependent) the utilities charge retail customers 100% of those costs. In simple terms, wholesale customers get the entire RMR service for free, even though in the case of Edison, about 13% of Edison's total transmission service revenue is recovered from wholesale transmission services. This allocation is unjust and unreasonable. Edison has not justified, nor can it justify, this allocation. The amounts in question are significant. Edison's RMR charges in 1998 exceeded $115 million and in 1999 exceeded $142 million.

Because we did not address this issue in previous decisions, we will not retroactively allocate a portion of RMR costs to wholesale customers, nor disallow this recovery through the TRA. However, we put Edison on notice that it will no longer be able t prospectively recover 100% of its RMR costs in the TRA. In its next RAP application, we direct Edison to discuss the steps it has taken at FERC to address this situation. In light of this warning, we will seriously consider a disallowance of RMR costs attributed to wholesale customers in the next RAP.

IV. Other Contested Issues

A. The Distribution Energy Charge

Consistent with Commission policy, Edison adjusts distribution rates by increasing each rate component (customer, demand, and energy charges) by the PBR Update Rule, unless that increase would result in a particular charge exceeding the rate level in effect on June 10, 1996, in violation of the rate freeze. In the event a particular charge would exceed its June 10, 1996 level, Edison calculates the dollar amount of the differential and converts that to a per-kilowatthour adder to its distribution energy charge. FEA objects to this approach and contends that such a differential should be recovered through increases to all of the remaining rate components, as it is done by PG&E.

Edison's methodology has already been adopted by the Commission, even though it differs from PG&E's methodology, and has been in use in designing Edison's rates during the transition period. (D.97-08-056, p. 46.) There is no reason for us to change this methodology at this time. Accordingly, we reject FEA's recommendation and adopt Edison's methodology.

B. LEV Programs

Pursuant to D.99-06-058 in the 1998 RAP, Edison reported on the implementation of its LEV programs and associated costs during the record period. This report demonstrates that Edison's record period costs and activities are reasonable and within the guidelines of the Commission's LEV decision, D.95-11-035.

FEA recommends that Edison be ordered to include in its next RAP filings a comprehensive analysis of the LEV programs from the inception to the present. FEA contends that the current filings do not provide a sufficient basis to allow an informed judgment of the overall effectiveness of the programs. We believe FEA's contentions are misplaced. There is no need for such additional data to be included in the next RAP. An annual RAP addresses Edison's program activities and associated costs within the record period of that particular proceeding. The RAP is not the proceeding to address the overall effectiveness of the programs or the continuation of the LEV programs beyond the current authorized period, December 21, 1995 through December 31, 2001. In D.95-11-035, we specifically set forth the information that should be provided annually and biannually. We also specified that those reports should be submitted to the Commission Advisory and Compliance Division (now the Energy Division).

Since the issuance of D.95-11-035, Edison has provided a detailed report of its program activities in every annual LEV report. In addition, Edison has provided a detailed report which also includes program expenditures in every Energy Cost Adjustment Clause (ECAC) proceeding and RAP for the record period covered by the particular proceeding. In the aggregate, these overlapping reports provide a complete and very detailed picture of the LEV program activities. We have reviewed this data and have found it reasonable. ORA was able to review Edison's LEV program for the current record period and concluded that Edison's implementation of its LEV program and associated expenditures are within the guidelines set forth in D.95-11-035 and that the recorded costs are reasonable.

The sum of information from these sequential reports is more than sufficient to allow an informed judgment of the overall effectiveness of the LEV programs. Edison should not be burdened with any additional analysis or compilation of the data.

V. Stipulations

A. Jurisdictional Allocation Stipulation

FEA, ORA, TURN, and Edison resolved their differences regarding the jurisdictional allocation factor issues in this proceeding. The agreement is set forth in the Jurisdictional Allocation Stipulation (Appendix B). The parties state that this stipulation represents a reasonable compromise of the parties' positions. It also promotes an efficient and optimal use of the parties' and the Commission's resources, and fairly reflects Commission decisions which govern the issues that are being considered in this proceeding.

The Stipulation provides that:

The stipulation recognizes: (1) the generation-related nature of the costs; (2) the diminishing amount of wholesale service Edison provides since its last GRC due to the restructuring of California's electric industry and associated impact on the jurisdictional-based allocation of its costs; (3) the need for consistency with the Commission's previously adopted methodology for similar costs reviewed in Edison's ECAC proceedings; and (4) the need for consistency with the treatment of generation costs and market revenues recorded in the TCBA. The methodology is also consistent with the decision in the 1998 RAP, which adopted a 100% retail allocation factor for the PBR exclusions, nuclear decommissioning, and public purpose programs revenue requirements. Accordingly, the Jurisdictional Allocation Stipulation will be adopted.

B. RMR Cost Allocation Stipulation

PG&E, ORA, and Aglet have compromised their differences regarding the allocation of RMR costs between retail and wholesale customers. This stipulation is Appendix C. PG&E has agreed to file with FERC, on or before April 28, 2000, a mechanism to recover RMR costs that includes a fair allocation of such costs to wholesale customers and to request that the filing be effective within 61 days of the filing date. In turn, ORA and Aglet agree that PG&E's commitment to make a filing at FERC resolves their concerns expressed in this proceeding with regard to allocating RMR costs between wholesale and retail customers, and that the stipulation supersedes the recommendations contained in ORA's testimony. The stipulation is reasonable and will be approved.

VI. Uncontested Issues

Issues which are uncontested by the parties will not be discussed but are adopted in the Findings of Fact. We have reviewed these uncontested issues and are satisfied that the utilities' proposals are appropriate and reasonable.

VII. Motion of SDG&E to File Advice Letter

In this RAP, SDG&E has calculated the LRMC of providing commodity procurement service to be .003¢/kWh. SDG&E asserts that because it had already ended its rate freeze it is necessary to split the .003¢/kWh between a PX credit of .001¢/kWh (to benefit direct access customers only) and a PX charge of .002¢/kWh (to be charged to bundled commodity service customers only). The result is that direct access customers will benefit by the .003¢/kWh differential compared to the amount paid by bundled customers.

Because SDG&E is proposing that a PX charge of .002 ¢/kWh be included in the existing electric energy change on its customer bills, it contends that a slight and minor rate increase could result. Normally, a formal application to increase rates is required by Commission rules. However, the Commission's GO 96-A, in Section VI, allows a utility to avoid a formal application to increase rates where the rate increase is minor in nature. Specifically, the rules states in pertinent part: "In cases where the proposed increases are minor in nature, the Commission may accept a showing in the advice letter provided justification is fully set forth therein, without the necessity of a formal application." Therefore, SDG&E moves the Commission, should it agree with SDG&E's proposal for a PX charge, to allow SDG&E to implement the PX charge by filing an advice letter pursuant to GO 96-A.

TURN objects to granting SDG&E's motion on the ground that it seeks relief beyond the scope of this RAP.

We will deny the motion. GO 96-A provides for review by our Energy Division. As we understand SDG&E's motion, if we were to grant it, the Energy Division review would be omitted. Our order in this proceeding authorizes an advice letter filing to implement the 0.034 ¢/kWh credit. Should SDG&E require further relief by way of a minor increase in rates it may file under GO 96-A for review by the Energy Division.

Findings of Fact

A. Power Exchange Credit Issues

1. D.99-06-058 states that, under the restructured electricity market in California, customers may subscribe to "bundled service" from the utility or "direct access" service from a competitive energy provider. Customers who purchase bundled service from the utility pay a PX charge to cover the utility's power supply costs, while customers who elect direct access service receive a credit on their bills called the "PX credit" that offsets the energy costs included in the bundled rate.

2. D.99-06-058 adopted refinements to the method of calculating the PX credit, and concluded that there was not an adequate record to adopt changes to its composition at that time, and ordered its composition to be reviewed in the current RAP.

3. D.96-10-074 and D.97-08-056 separated each utility's last authorized rate base, revenue requirement, and rate design into generation, transmission, distribution, and public purpose rate components.

4. D.97-08-056 recognized that the purpose of unbundling is to promote the development of competitive markets for generation, and established critical policy principles that apply to rate unbundling.

5. D.97-08-056 found that unbundling promotes competition by providing customers with options for individual services and by sending customers price signals, which would permit them to make reasoned choices about their competitive options. It also found that the purpose of promoting competition, where it may be viable, is to assure the best use of the economy's resources, to assure that customers pay the lowest price for services, and to expand the array of services available to customers.

6. A key policy principle established by D.97-08-056 is that "costs associated with one function will not be allocated to other functions."

7. 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.

8. D.97-08-056 rejected SCE's argument that allocation of fixed A&G costs to generation would improperly disallow appropriately incurred costs because of SCE's perception that it could not recover fixed costs in competitive generation markets, and rejected similar arguments by PG&E and SDG&E.

9. D.99-06-058 states utility commodity prices must ultimately recognize those costs, which the utilities must recover in the long run as any other provider.

10. To charge direct access customers for procurement costs would be to charge them for costs of services that they have chosen to forgo, by selecting Energy Service Providers other than the utility.

11. Charging customers only for services that they choose to use (e.g., by remaining on bundled service) is consistent with the Commission's view of long run marginal cost-based pricing.

12. The functions performed by the utility in the process of procuring power for bundled-service customers need to be clearly defined in order to establish an appropriate procurement rate.

13. California's utilities are no longer the sole source of power production.

14. The matching of supply and the utilities' forecasted load is performed by the PX.

15. The utility's involvement in power procurement for bundled-service customers now generally consists of functions including preparing and submitting load forecasts to the PX, submitting meter data for settlements to be computed by the ISO and PX, processing settlement statements, and computing retail charges.

16. ORA identified the following items as procurement-related functions rather than related to wires company functions or market facilitation:

17. SCE provided an alternative listing of functions that allows its procurement costs to be analyzed:

18. None of the utilities attempted comprehensive cost separation studies, but only examined the categories of costs raised by the intervenors in the last RAP proceeding.

19. None of the utilities provided PX credit methodologies based on the Commission's long-run marginal cost approach.

20. Only SCE used recorded or budgeted data, as directed by D.99-06-058, as a proxy for estimating LRMCs.

21. SDG&E's costs of PX monitoring and analysis are twice as big as PG&E's (even though SDG&E is much smaller in terms of sales and customers).

22. If this inconsistency between the showings of the three utilities were to continue in future reviews of their procurement costs, the Commission would not be able to completely resolve the separation of procurement costs from distribution costs.

23. ORA used information from the utilities' data responses to identify their annual costs of procurement activities.

24. Procurement activities total $2.105 million for SDG&E.

25. SCE's year 2000 budget for procurement activities is $27.2 million.

26. ORA's recommended PX costs for SCE and SDG&E are conservative estimates for the procurement rate to be adopted in this proceeding.

27. In order to convert the utility's costs of commodity procurement to a rate, the total costs allocated to bundled-service and to maintaining default-service capabilities would then be divided by the sales volume applicable to each of these services.

28. These procurement rate components would then be adjusted for common cost overheads, which maintain consistency with D.97-08-056.

29. Dividing SCE's total procurement-related cost of $27.2 million by SCE's annual sales of 79,470 GWh, produce a total procurement rate of 0.034 cents per kWh

30. Dividing SDG&E's total annual procurement cost of $2.105 million by SDG&E's annual sales, result in a total procurement rate component of 0.012 cents per kWh.

31. Budget data for PG&E proved to be unusable.

32. ORA used SCE's total procurement-related revenue of $33.4 million as a proxy for PG&E. ORA divided that by PG&E's annual sales of 78,631 GWh, to produce a recommended total procurement rate of 0.042 cents per kWh.

33. ARM proposed procurements adders of 0.065 cents per kWh for SDG&E, 0.067 per kWh for Edison, and 0.059 cents/kWh for PG&E based on its studies.

34. If all procurement costs are assigned to serving bundled service customers, the total procurement costs should be divided by bundled service sales. The resulting procurement rates are 0.048 cents per kWh for PG&E, 0.040 cents per kWh for SCE, and 0.015 cents per kWh for SDG&E, which are the procurement cost adders to the PX credit that should be adopted in this proceeding, under the ORA method.

35. The utilities as a provider of competitive services should be required to interact with other parts of the utility through the same mechanisms as other ESPs, including comparable access to information, thus ensuring a "level playing field" and providing a realistic test of whether substantial costs are required to serve bundled-service customers.

36. The Commission's policy determinations in other proceedings support only allocating procurement costs to the PX rate paid by bundled service customers, and not to rates that direct access customers would not be charged, since they have chosen to forgo the utilities' procurement services by selecting other ESPs.

37. D.99-06-058 (the 1998 RAP Decision) required that the LRMC of the utilities' energy procurement services be calculated so that it could be added to the PX credit.

38. LRMC is the change in cost associated with a small change in output, over a long enough time that all factors of production that are capable of varying can be changed.

B. RMR Issues

39. In D.97-12-109 and D.98-04-019 the utilities were authorized to record RMR payments made to the ISO in the TRA, to the extent that those payments are recovered from the revenues collected by each utility during the transition period.

40. D.97-12-109 and D.98-04-019 both contemplated that the RMR recovery mechanism established therein would end at the end of the transition period.

41. In A.96-12-019 the Commission did not address the propriety of retail customers paying 100% of RMR payments made by Edison to the ISO.

42. No party disputed that the RMR cost entries and related refund entries in the TRA by the utilities were inaccurate or reflected any amounts other than what was paid by the utilities or received in refunds by the utilities.

43. The rates, terms, and conditions of the ISO tariff and the Transmission Owner TO tariff are under FERC's jurisdiction.

44. The rates, terms, and conditions of the contracts under which various generators provide RMR service to the ISO are under FERC's jurisdiction.

45. Under the ISO tariff, the ISO invoices the utilities for the costs of RMR units located in their service area.

46. Under the TO tariff, the utilities can file to recover RMR charges from the retail end-use customers located in their service area, pursuant to either a FERC mechanism or a CPUC mechanism.

47. Edison has never filed at FECR for a mechanism to recover RMR costs from any of its customers, therefore, Edison does not have filed-rates for these costs.

48. As part of their regular operations, PG&E and Edison deliver electricity over their own transmission lines to wholesale customers. During that process wholesale customers benefit from the transmission system reliability added by RMR units, in the same way that retail customers benefit.

49. Wholesale customers benefit from RMR units whose costs are billed by the ISO to PG&E and Edison, but pay nothing toward those costs.

50. To exempt wholesale customers from paying their fair share of RMR costs is to give them a free ride to avoid paying for benefits received.

51. The allocation authorized by the FERC-approved tariff is unjust and unreasonable.

C. Other Edison Contested Issues

52. Edison's methodology of adjusting the Distribution Energy Charge so that an increase in the other distribution rate components does not violate the rate freeze is reasonable and consistent with Commission policy.

53. The administration and cost information Edison submits annually in the RAP regarding its LEV programs is reasonable and consistent with D.95-11-035.

54. Edison appropriately extended its Special Contracts with Mobil Oil Company and Dow Chemical Company in accordance with Pub. Util. Code § 372, and there is no reason to require Edison to submit additional information supporting these contract extensions in its next RAP.

D. Stipulated Matters

55. The "Stipulation Among the Federal Executive Agencies, the Office of Ratepayer Advocates, The Utility Reform Network, and Southern California Edison Company Regarding Jurisdictional Allocation Issues in the 1999 Revenue Adjustment Proceeding (Application No. 99-08-022)" is a reasonable compromise of the parties' positions, an efficient and optimal use of the parties' and the Commission's resources, and consistent with Commission decisions, and should be adopted.

56. The stipulation among PG&E, ORA, and Aglet regarding RMR costs is a reasonable compromise of the parties' positions, an efficient and optimal use of the parties' and the Commission's resources, and consistent with Commission decisions, and should be adopted.

E. Uncontested Edison Issues

57. Edison should be permitted to update its forecast revenue requirements provided in Table II-1 of its 1999 RAP Report to reflect the October 31, 1999 recorded balances in all memorandum and balancing accounts and should include the impact of all Commission decisions issued through the effective date of a decision in this proceeding.

58. Edison's entries into its TRA should be based on the PBR Distribution Exclusions revenue requirement, Transmission revenue requirement, Nuclear Decommissioning revenue requirement, and Public Purpose Programs revenue requirements, adopted as of the date of this decision.

59. Edison's PBR Distribution Exclusions revenue requirement is comprised of the following amounts which should be included in its 2000 Distribution revenue requirement:

a) A Reduced Capital Recovery Amount and Incremental Return authorized revenue requirement of ($57.098) million.

b) A portion of the Streamlining Residual Account (SRA) balance associated with the Non-Utility Affiliate Credits in the amount of ($22.639) million.

c) The Hazardous Waste Balancing Account balance of $16.522 million.

d) The Demand-Side Management (DSM) Incentives authorized revenue requirement adopted as of the date of this decision.

e) A portion of the SRA balance associated with the DSM Incentives in the amount of $1.781 million.

f) The portion of the PBR Distribution Rate Performance Memorandum Account associated with Edison's PBR net revenue sharing for 1997, including interest through April 6, 2000, authorized pursuant to Resolution E-3656. The amount to be returned to ratepayers through the distribution rate component will be included in Edison's compliance advice letter to be submitted on or before May 6, 2000.

g) The balance of the Affiliate Transfer Fee Memorandum Account in the amount of ($0.703) million, pursuant to D.97-12-088. Edison received notification on March 16, 2000 that Advice Letter 1289-E, which establishes the ATF Memorandum Account, was approved.

60. Edison's Transmission revenue requirement is comprised of the Base Transmission revenue requirement adopted as of the date of this decision and the Transmission Revenue Balancing Account Adjustment amount of ($32.494) million, and is appropriate.

61. Edison's Nuclear Decommissioning revenue requirement of $44.097 million is comprised of the following amounts and is appropriate:

a) The Nuclear Decommissioning Trust Fund revenue requirement of $25.0 million.

b) The San Onofre Unit No. 1 Shutdown Operation & Maintenance currently authorized amount of $11.522 million.

c) The Department of Energy (DOE) Decontamination & Decommissioning (D&D) Fee in the amount of $4.611 million.

d) A portion of the SRA balance associated with the DOE D&D Fees in the amount of $0.464 million.

e) The Spent Nuclear Fuel Storage (SNFS) Fee in the amount of $3.057 million.

f) A portion of the SRA balance associated with the SNFS Fees in the amount of ($0.557 million).

62. Edison's Public Purpose Programs revenue requirement of $191.925 million is comprised of the following amounts and is appropriate:

a) DSM, Research Development and Demonstration (RD&D), and Renewable amounts of $90.0 million, $28.5 million, and $49.5 million, respectively, as mandated by AB 1890.

g) The Electric Vehicle Memorandum Account balance in the amount of $0.758 million.

h) A portion of the SRA balance associated with Intervenor Compensation payments in the amount of $0.837 million.

i) Franchise fees associated with the above listed Public Purpose Programs in the amount of $2.153 million.

63. The 1999 sales forecast proposed by Edison should be used to update the nongeneration Equal Percent of Marginal Cost (EPMC) factors utilized in allocating the PBR Exclusions and to convert those allocated revenues to a cents-per-kWh rate.

64. In the event that Edison's cost of capital Trigger Mechanism results in a cost of capital change, the resulting change in the Distribution revenue requirement calculated on a 1996 basis should be allocated to each customer class by their respective 1996 nongeneration EPMC percentages.

65. Edison's CARE surcharge amount of $0.00079 per kWh should be included in the Public Purpose Program charge.

66. Edison's proposed 2000 retail sales forecast of 79,470 GWh should be used to calculate the PBR Exclusions, Nuclear Decommissioning, and Public Purpose Programs rate levels.

67. Edison's Optional Pricing Adjustment Clause Balancing Account balance should be transferred to its TRA once the Commission reviews the 1998 Flexible Pricing Options Annual Report and determines that the shareholder contributions have been correctly calculated.

68. Edison should eliminate the following accounts as of the effective date of this decision:

69. Edison's ISO/PX Implementation Delay Memorandum Account should be eliminated upon authorization of Edison's proposed disposition of any remaining balance pursuant to a Commission decision in Edison's 1999 Annual Transition Cost Proceeding, A.99-09-013.

70. Edison should eliminate the following balancing and memorandum accounts upon authorization of Edison's proposed disposition of any remaining balances pursuant to a Commission decision in Edison's Direct Access Service Fee application, A.99-06-040:

71. Edison should eliminate the following balancing and memorandum accounts upon authorization of Edison's proposed disposition of any remaining balances in this proceeding.

h) Electric Magnetic Field Balancing and Memorandum Account.

i) Jurisdictional Allocation Memorandum Account.

j) Women, Minorities & Disabled Veterans Memorandum Account.

72. Edison should be authorized to modify the Rate Group Tracking Memorandum Account to include the Trust Transfer Amounts and an imputed 10 percent rate reduction revenue amounts in the Rate Group CTC Revenue Memorandum sub-account each month.

73. Edison should be authorized to retain all of its existing balancing and memorandum accounts not addressed in these findings of fact.

74. Edison should be authorized to transfer the $1.069 million generation-related balance in the Catastrophic Event Memorandum Account to its TRA on the effective date of the decision in this proceeding.

75. On the effective date of this decision, Edison should be authorized to transfer residual balances recorded in the following balancing and memorandum accounts to its TRA, and adjust the appropriate revenue requirement in the operation of its TRA to ensure that the residual CTC revenue is determined correctly without having to adjust rate levels:

k) CARE Adjustment Account.

l) EMF Balancing and Memorandum Account.

m) Catastrophic Event Memorandum Account.

n) RD&D Balancing Account (1995 GRC Unspent Balance portion only).

o) Women, Minorities & Disabled Veterans Memorandum Account.

76. Edison's Administration of its LEV Program and associated costs is reasonable for the May 1, 1998 through April 30, 1999 Record Period.

77. As of the effective date of D.99-09-070, adopting Edison's Gross Revenue Sharing Mechanism, all Other Operating Revenue (OOR) generated from Edison's LEV Program activities from September 16, 1999 forward will be subject to treatment under the adopted mechanism.

78. For OOR generated from Edison's LEV Program activities prior to September 16, 1999, Edison should credit back the OOR amounts to Edison's Electric Vehicle Adjustment Clause Balancing Account.

79. Edison's administration of its Self-Generation Deferral Rate Contracts during the Record Period is reasonable.

F. PG&E Findings

80. PG&E's administration of special electric contracts for the record period ending December 31, 1998, was reasonable.

81. PG&E's total costs recorded in the EVBA do not exceed the allocated budget under D.95-11-035.

82. PG&E files annual reports with the Commission providing detailed information on its LEV program, including accomplishments, projects, and expenditures.

83. PG&E's 1998 costs for its LEV program are reasonable.

84. It is not necessary to perform a review of PG&E's LEV programs from inception to the present.

85. PG&E proposed retention of six remaining IRMA subaccounts in its RAP application because the Commission had not yet specifically authorized PG&E to record unanticipated restructuring implementation costs PG&E incurred in 1999 in the Electric Restructuring Costs Account (ERCA).

86. In Resolution E-3648, the Commission authorized PG&E to record these unanticipated restructuring costs in the ERCA.

87. PG&E should address the six subaccounts of the IRMA in their next RAP application.

88. PG&E allocates performance-based ratemaking exclusion items such as the EVBA, the HSM, and SRA using the non-generation EPMC methodology.

89. PG&E's proposal to amortize the balances in the EVBA, the HSM and the SRA, by establishing rate components for these items on an equal ¢/kWh basis complies with Commission ratemaking requirements and is uncontested.

90. For the record period June 1998 through June 1999, PG&E correctly transferred all residual CTC revenue from the TRA to the TCBA.

91. PG&E's incorporation of real-time post settlement adjustments and block forward market costs into the PX credit calculation as required by Resolution E-3618 is reasonable.

92. PG&E's special electric contracts and entries to the EVBA are reasonable.

93. PG&E's proposals with regard to elimination of memorandum and balancing accounts are reasonable.

94. PG&E's entries in the TRA for the June 1998 through June 1999 periods are reasonable.

95. The consolidated and unbundled revenue requirements adopted by the Commission in other proceedings for entry into the TRA are reasonable.

96. PG&E's revenue allocation and rate design proposals are reasonable.

97. PG&E's request to consolidate the revenue requirements authorized in pending proceedings impacting test year 2000, including the Annual Earnings Assessment Proceeding (A.99-05-007), the Cost of Capital proceeding (A.99-11-003), the § 368(e) proceeding (A.99-03-039), and the Catastrophic Event Memorandum Account proceeding (A.99-01-011) is reasonable.

98. PG&E's request to update the illustrative 2000 revenue requirements presented in this proceeding to include the balancing and memorandum accounts' latest recorded balances for recovery in the TRA is reasonable.

Conclusions of Law

1. All on-going costs related to energy procurement, including maintenance, refinement, and enhancement of utilities' systems, from the PX by the utilities on behalf of bundled-service customers (including Virtual Direct Access) should be recovered through the PX price, and direct access customers should be given a credit for these costs.

2. The purpose of unbundling is to promote the development of competitive markets for generation, and established critical policy principles that apply to rate unbundling.

3. Unbundling promotes competition by providing customers with options for individual services and by sending customers price signals, which would permit them to make reasoned choices about their competitive options. The purpose of promoting competition, where it may be viable, is to assure the best use of the economy's resources, to assure that customers pay the lowest price for services, and to expand the array of services available to customers.

4. Utility pricing must eliminate any competitive advantage created by an institutionalized removal of costs otherwise intrinsic to the provision of a service.

5. The appropriate allocation of procurement costs is to bundled service customers. To charge direct access customers for procurement costs would be to charge them for costs of services that they have chosen to forgo, by selecting Energy Service Providers other than the utility.

6. The following items are appropriate procurement-related functions rather than related to wires company functions or market facilitation:

7. SCE's alternative listing of functions appropriately allows its procurement costs to be analyzed:

8. PG&E and SDG&E failed to use recorded or budgeted data, as directed by D.99-06-058, as a proxy for estimating.

9. ORA's recommended PX costs for SCE and SDG&E are reasonable and conservative estimates for the procurement rate to be adopted in this proceeding.

10. Budget data for PG&E submitted in this proceeding should not be used to calculate the PX credit.

11. ARM's recommended PX costs for PG&E should be adjusted to take into account improper inclusions. ORA's recommended PX adder for PG&E is a reasonable proxy for an adjusted ARM figure.

12. There should be a single procurement factor added to the PX credit adopted in this proceeding for all utilities, which should be the average of the individual procurement adders for the three utilities.

13. This proceeding cannot be considered to have said the last word on the establishment of a rate component for commodity procurement, since issues under consideration in other proceedings will not have been decided in time for inclusion in the rates being set in this proceeding.

14. The Commission should allow parties to the next RAP proceeding to propose adjustments to the commodity procurement rate, to reflect the outcome of the RCS/DASF proceeding.

15. The utilities as a provider of competitive services should be required to interact with other parts of the utility through the same mechanisms as other ESPs, including comparable access to information, thus ensuring a "level playing field" and providing a realistic test of whether substantial costs are required to serve bundled-service customers.

16. The Commission's policy determinations in other proceedings support only allocating procurement costs to the PX rate paid by bundled service customers, and not to rates that direct access customers would not be charged, since they have chosen to forgo the utilities' procurement services by selecting other ESPs.

17. The difficult process of determining these amounts should be simplified in future proceedings by establishing a standard definition of pertinent functions, which the utilities should then be expected to use when recording their procurement-related costs in order to enable future audits to verify that all pertinent costs have been recorded in a transparent manner.

18. The procurement cost adders to the PX credit that should be adopted in this proceeding are 0.048 cents per kWh for PG&E, 0.040 cents per kWh for SCE, and 0.015 cents per kWh for SDG&E.

19. The single procurement cost adder to the PX credit that should be adopted and implemented for PG&E, SCE and SDG&E is 0.034 cents per kwh.

20. This Commission cannot legally order Edison to make a Federal Power Act Section 205 filing at FERC under Mass. Dept. of Pub. Util. v. U.S., 729 F.2d 886 (1st Cir. 1984). However, this Commission has jurisdiction to decide how much of Edison's RMR costs Edison may recover from its distribution customers.

21. The filed rate doctrine does not apply in this case because Edison elected to file for a mechanism to recover if its RMR costs at this Commission rather than at FERC.

22. Aglet's recommendation that the Commission allocate a percentage of total RMR costs incurred by Edison since April 1998 to wholesale customers and thereby disallow a portion of the RMR costs already paid to the ISO is denied.

23. Edison is put on notice that it will not be able to prospectively recover 100% of its RMR costs in its TRA.

24. SDG&E's request to segment the PX credit between a credit and a charge is denied.

25. SDG&E's request to increase rates is denied.

26. The stipulations set forth in Appendices B and C are adopted.

27. The uncontested issues described in the Findings of Fact are reasonable and are adopted.

28. The PX credit issue is severed from the RAP.

29. The utility distribution companies shall file their next PX credit adjustment proceeding September 2003.

ORDER

IT IS ORDERED that:

1. The Power Exchange credit adder to be credited to the electricity bill of each direct access customer is 0.034 cents per kilowatt-hour. This adder shall be credited in addition to the credit that offsets the wholesale procurement of energy for bundled customers. This credit is applicable to customers of Southern California Edison Company (Edison), Pacific Gas and Electric Company, and San Diego Gas & Electric Company (the utility distribution companies).

2. Within 15 days after the effective date of this order the utility distribution companies shall file tariffs implementing Ordering Paragraph 1.

3. In Edison's next RAP application, Edison shall delineate the efforts it has undertaken at the Federal Energy Regulatory Commission to recover a fair share of Reliability Must-Run Costs from its wholesale customers..

4. The utility distribution companies shall file their next Revenue Adjustment Proceeding (RAP) on September 1, 2000.

5. The PX credit issue is severed from the RAP.

6. The utility distribution companies shall file their next PX credit application in September 2003.

7. Application (A.) 99-08-022, A.99-08-023, and A.99-08-026 are closed.

This order is effective today.

Dated , at San Francisco, California.

1 ARM is an alliance of energy service providers who actively participate in the California retail electric market, including PG&E Energy Services, NewEnergy, Inc., Enron Corp., Utility.com, GreenMountain.com Company, and Shell Energy Services. ARM members sell directly to residential, commercial, and industrial end-use customers. 2 Briefs were filed by those parties submitting testimony, and the California Department of General Services (General Services), the California Farm Bureau Federation (Farm Bureau), and the Center for Energy Efficiency and Renewable Technologies (CFEE). 3 D. 99-06-058 [mimeo] at p. 24. 4 D.97-08-056 stated specifically, at p. 8: "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." 5 See, for example, the testimony of CUE witness David Marcus, Transcript at p. 882, lines 4-7. 6 ARM Opening Brief at p. 21; D.96-04-050, [mimeo], at p. 25. 7 Transcript at pp. 418-419. 8 PG&E Opening Brief at p. 12. 9 Transcript at p. 179, lines 5-12. 10 SCE Opening Brief, p. 15. 11 Transcript at p. 889, lines 13-14. 12 Exhibit 37, RBW-12, PG&E Response to ARM Data Request 1(8). 13 SCE Opening Brief, p. 14. 14 ARM Opening Brief, p. 27. 15 PG&E, for example, "recommends that the Commission rely on a short-run marginal cost methodology for including any costs in the PX credits...." PG&E Direct Testimony at p. 6-5. 16 For example, SCE Opening Brief at p. 19. 17 D.96-04-050, [mimeo], at p. 25. 18 Id. at p. 26. 19 Transcript at p. 205, lines 8-10. 20 SCE Opening Brief, p. 21. 21 DGS Opening Brief at p. 17. 22 PG&E Opening Brief, p. 13. 23 Transcript at p. 472, lines 27-28 and p. 473, lines 1-2. 24 Transcript at p. 1112, line 12. 25 PG&E Opening Brief, p. 12; SCE Opening Brief, p. 12. 26 ARM Opening Brief, pp. 23-24. 27 Transcript at p. 415, lines 14-27. 28 Transcript at p. 529, lines 15-18. 29 D.95-12-016, Appendix C, p. 2. 30 Transcript at p. 204, lines 25-26. 31 PG&E Direct Testimony, p. 6-13. 32 See testimony of Dr. James Price, Transcript at pp. 1085-1086, and Michael Florio, Transcript at pp. 757-759. 33 ORA Opening Brief at p. 11. 34 Analysis since ORA's data requests were issued has determined that management of existing purchase contracts (item D) is eligible for recovery as a Competition Transition Cost (CTC), thus ORA has not included this cost in its recommended commodity procurement rate. 35 Exhibit SCE-1, p. 154. 36 Even though the category is called gas and electric supply, it is in fact only for electric supply. 37 D.99-05-031, Attachment 1, page2. 38 D.00-02-046, p.110 39 Id., p.111 40 D.97-12-109, mimeo. p. 11, Conclusion of Law 3. See also D.98-04-019: "[PG&E], [Edison], and [SDG&E] are authorized to recover must-run payments made to the [ISO] and authorized by the [FERC] to the extent that these payments are recovered from the revenues collected by each utility during the transition period and as described herein." (D.98-04-019, mimeo. p. 5, Ordering Para. 1.) 41 PG&E currently allocates 100% of RMR costs to its retail customers. But it has acceded to ORA's recommendation and has filed at FERC to allocate a portion of those costs to wholesale customers. SDG&E has already made such a filing. 42 ORA originally made this recommendation for both Edison and PG&E, but because PG&E has made the request filing at FERC, ORA's recommendation is now limited to Edison. (See Appendix C.)

Previous PageTop Of PageGo To First PageNext Page