5. Revenue Allocation and Common Rate Design Issues

5.1 Objectives and Principles

The allocation of revenue requirement responsibility to the various customer classes and the establishment of specific rates and rate structures to collect the adopted revenue requirement are the primary objectives of this decision. We consider the extent to which the revenue allocation and rate design principles adopted in D.01-05-064 are appropriately applied in this decision. For example, in setting rates for residential customers, we look first to the five-tier structure adopted for Edison and PG&E, determine whether there are any grounds for adopting a different approach for SDG&E, and approve such approach in the absence of such grounds. Similarly, we consider whether class caps for agricultural customers such as those adopted in D.01-05-064 should be adopted for SDG&E's agricultural customers.

We note that to varying degrees, the parties have generally taken a similar approach in making their recommendations in this proceeding. In D.01-05-064 we "...called upon our institutional expertise and experience as well as our understanding of law and policy to make hard choices based on the law, California energy policy and the record before us." (D.01-05-064, p. 9.) Because we call upon that same expertise and experience here, it is reasonable to expect that we will reach similar conclusions on issues having the same or similar policy and factual contexts.

Our objectives in this decision include the following:

1. Establish rates that will ensure the DWR revenue requirement attributable to SDG&E's ratepayers is collected through SDG&E's retail rates, consistent with D.01-09-___.

2. Observe and give effect to the second sentence of § 332.1(f) by considering the extent to which rates for SDG&E's large customers should be adjusted based on comparisons of the comparable rates for large customers of Edison and PG&E.

3. Observe and promote principles of equity in the allocation of revenue responsibility and rate design. As we stated in D.01-05-064, "...equity transcends the application of simple mathematical formulas. We therefore evaluate rate design proposals considering customers' ability to pay and the hardship that rate increases impose on particularly vulnerable customers." (D.01-05-064, p. 14.)

4. Promote energy conservation in the establishment of rate structures. For example, by assigning more costs at peak usage hours, we may encourage consumption reductions during those hours. This in turn could mitigate the energy crisis by potentially reducing the frequency and duration of rolling blackouts that endanger the public health and safety. To the extent that wholesale market prices are at least somewhat responsive to reductions in demand, this could also mitigate high prices.

5. Observe and recognize legal requirements and practical constraints. For example, Water Code § 80110 prohibits rate increases for residential customers for usage up to 130% of baseline allowances in existence when ABX1 1 was enacted. Also, rate design proposals that might otherwise be meritorious should not be adopted if metering or billing systems cannot accommodate the proposals in relevant time frames. Similarly, in fairness to the parties, the constrained opportunity for full consideration of novel and complex proposals in this expedited proceeding weighs against approving such proposals at this time.

We have historically applied cost-based principles when assigning revenue responsibility to customer classes and designing individual rate structures, but we are severely limited in our ability to do so here. We have had insufficient opportunity to obtain and analyze evidence concerning the cost components underlying the rate increases that we adopt today. The development, analysis, and application of marginal cost data are typically time-consuming and highly contested undertakings in revenue allocation and rate design proceedings. The fast pace of this abbreviated proceeding suggests that we should view with great caution proposals for customer class allocations that rely on detailed cost analyses.

5.2 Assignment of Utility-Retained Generation
(URG)

SDG&E proposes that separate average rate increases be established for small and large customers based on an assumption that URG is assigned solely to small customers. With respect to the August 7, 2001 DWR revenue requirement update, SDG&E proposes that the average rate increase for small customers be set at 2.64 cents per kWh and that the average rate increase for large customers be set at 3.16 cents per kWh. In connection with this proposal, SDG&E also proposes that large customers be exempt from ongoing Competition Transition Cost (CTC) charges on the basis that CTC charges are associated with URG. If the CTC rate were not removed from the rate increases that result from implementing the DWR revenue requirement, the average large customer rate increase would be 3.82 cents per kWh.

Other parties oppose SDG&E's proposed treatment of URG, finding no practical or legal justification for it. Because SDG&E's proposal is complex and given the expedited nature of this phase of the proceeding, we are not able to fully consider the proposal at this time. Thus, we will not address this proposal in this phase of the proceeding and will continue with our usual ratemaking practice and allocate URG to all customer classes.

5.3 Revenue Allocation

SDG&E proposes to use an equal cents per kWh allocation of the DWR increase. Except for FEA, the parties that take a position on revenue allocation support SDG&E's approach. FEA on the other hand proposes to allocate revenue responsibility based on loss adjustments, assumptions regarding demand and energy splits in generation, a "market adjusted allocation ratio" method for incorporating peak and off-peak prices into the allocation of DWR energy, and a "top 100 hours" allocation of the demand-related portion of URG.

For Edison and PG&E, we adopted an equal cents per kWh revenue allocation approach to revenue allocation in D.01-05-064. We have considered whether this approach should be carried forward in this proceeding, and conclude that it should be. FEA's methodology is complex, has not been adequately examined in this expedited proceeding, and requires acceptance of assumptions about underlying generation costs that we are not prepared to accept on the basis of this record. Our reasons for deferring action on a cost-based approach to revenue allocation include the following:

FEA uses market index prices for South Path 15 (Southern California) for the year 2000 to establish an on-peak/off-peak price ratio. Notwithstanding the confidence that FEA witness Brubaker has in contending that 2000 is a reasonable proxy for 2001 for the May - December period, we find insufficient basis for assuming that the defective and seemingly random market prices seen in the past year will be repeated in the future. Nor can we find that the South Path 15 price indices reflect actual costs underlying the DWR revenue requirement at issue in this decision. We concur with SDG&E that unpredictability of the current market with its flaws and defects is reason to simplify cost allocation methods.

FEA's use of "6x16" and "other" classifications to represent on-peak and off-peak blocks of energy, respectively is inappropriate since it covers the on-peak and semi-peak periods for commercial and industrial TOU rate schedules. Essentially, this requires an implicit assumption that there is no price differential between the on-peak and semi-peak prices. Also, the residential on-peak period is only six hours, and for agriculture TOU, the on-peak period is eleven hours. The assumption of a sixteen-hour on-peak period for commercial, industrial, residential TOU, or agricultural customers is clearly inappropriate.

FEA's calculations are based on proxy data and assumptions because the actual costs underlying the DWR-determined revenue requirement are not available.

To develop demand and energy-related cost components, FEA made certain assumptions with respect to the demand-related costs associated with the SONGS costs. However, under the "ICIP" ratemaking mechanism, SONGS-related costs are treated as 100% variable, or energy-related. FEA's assumption regarding a demand component for SONGS costs is not consistent with current ratemaking.

In applying the "top 100 hours" method, FEA applied the Schedule A allocation factor to Schedules A-TOU and AD rather than the AL-TOU allocation factor. FEA has also used allocation factors that are not consistent with the "top 100 hours" allocation factors currently used in developing current rates.

For the reasons detailed above, we reject FEA's allocation methodology and we adopt SDG&E's proposal for an equal cents per kWh allocation of revenue responsibility, (before allocations of revenue shortfalls are taken into account). While we generally believe that it is appropriate to allocate revenue responsibility among customer classes on the basis of cost causation principles, FEA's proposals rely on too many unproven assumptions to justify application of those principles here.

5.4 Limiters, Rate Caps, and other Exclusions

ORA proposes that we adopt "bill limiters" or "average rate limiters" as a safeguard against the possibility of individual customers within a class or category facing extremely high bill increases. According to ORA, average rate limiters would act as a general safety net to protect against such increases. ORA's proposal is to limit each customer's rate to a multiple of the average rate for each rate schedule: 300% for industrial rates and 250% for agricultural rates. SDG&E opposes average rate limiters as unjustified cross-subsidies that are inefficient and inconsistent with the need to send strong conservation signals to customers. Rather than adopt this broad safety-net approach to deal with unforeseen problems, which could unnecessarily blunt conservation pricing signals, we prefer to address any such problems that might arise on a case-by-case basis. We recently reconsidered and rejected the use of average rate limiters in D.01-06-040.

CFBF requests that we cap rate agricultural rates increases at 15% for TOU rates and 5% for non-TOU rates as proposed by Governor Gray Davis. (State of California: Meeting the Energy Challenge, Governor Gray Davis, April 5, 2001.) Alternatively, CFBF proposes capping agricultural rate increases at the 20% TOU/15% non-TOU rate levels adopted in D.01-05-064.

SDG&E faults the CFBF proposal as being an unwarranted subsidy, but the company fails to articulate sufficient reason why we should adopt a different approach for agricultural rate subsidies in the SDG&E service territory than we approved for agricultural customers in the Edison and PG&E service territories. We believe that with the language in § 332.1(f) with respect to inter-utility rate comparability, the legislature encourages the commission to consider adopting the same approach in the absence of evidence to the contrary. Also, any claim that an agricultural rate cap is inappropriate simply because it shields customers from market prices ignores the fact that all SDG&E customer rates now are capped at 6.5 cents, and the fact that CARE-eligible customers and residential customers using less than 130% of baseline are exempt from rate surcharges, further masking the full market price. We have also recognized that exposing customers to full market prices could have a "significant negative impact on business and the California economy." (D.01-05-064, p. 27.) The agricultural customer class makes up less than 1% of SDG&E sales and revenues, and the impact of implementing a cap on agricultural rates will be minimal. We therefore will approve caps on increases for agricultural rates of 20% for TOU rates and 15% for non-TOU rates. While we approve agricultural rate caps as a matter of policy, we note that with a system average increase of 1.46 cents per kWh, the allocation to the agricultural rate schedules does not trigger the caps.

Water Code § 80110 provides that residential customer usage up to 130% of baseline quantities is exempt from increases in electricity charges. We give effect to this requirement in structuring a tiered residential rate structure. SDG&E supports the exemption of CARE-eligible customers as well as the exemption of medical baseline customers. No party contests these residential customer exemptions, which we hereby adopt.

5.5 Allocation of Revenue Shortfalls

The caps and exemptions adopted in the previous section create revenue shortfalls that must be assigned to other customers (or consumption levels) in order to meet our objective of setting rates sufficient to collect the DWR revenue requirement.

Proposals to allocate the shortfalls from the CARE-eligible customer and medical baseline customer exemptions to all other customers (but not to residential consumption below 130% of baseline) on an equal cents per kWh basis are uncontested. These proposals are generally consistent with our allocation of shortfalls in D.01-06-064, and we extend their use to SDG&E.¹³ Similarly, while SDG&E opposes caps on agricultural rate increases, it does not oppose an equal cents per kWh allocation of any revenue shortfall to all other customers, other than to CARE-eligible and medical baseline customers, and other than residential consumption within 130% of baseline. We therefore intend that agricultural shortfalls be allocated on this basis. As noted above, the adopted agricultural caps are not invoked with the system average increase implemented in this decision.

SDG&E and FEA propose to allocate the shortfall from the 130 % of baseline exemption within the residential class, but not to CARE-eligible customers and medical baseline customers. Aglet and ORA propose that this shortfall be allocated to all customers, other than CARE-eligible and medical baseline customers, as was done in D.01-05-064.

We first note that parties in this proceeding are in agreement that the 1/3 residential; 1/3 commercial; 1/3 industrial allocation method adopted in Decision 01-05-064 does not work for SDG&E. SDG&E has a combined commercial/industrial classification and a much different mix of customer classes (e.g., far fewer commercial/industrial customers as compared to residential) than either PG&E or Edison.

In D.01-05-064 we allocated the shortfall from the 130% of baseline exemption to all other non-exempt consumption, including consumption by commercial and industrial customers. We did so out of concern that allocating the entire revenue requirement shortfall within the residential class would create rate spikes that are too severe. (D.01-05-064, p. 26.) Also, we noted the importance of establishing rates that send appropriate conservation signals to all customer classes, and noted that shifting costs to non-exempt residential consumption would undermine that objective because such consumption represents only 11% of total consumption for Edison and PG&E combined.¹⁴ (Id., p. 22.)

We find that similar concerns are applicable here. Allocating the 130% of baseline shortfall only within the residential class would require far greater increases for non-exempt residential consumption. For example, we have calculated that for the fifth residential tier that we establish today, the current rate would have to be increased by more than 50%. We do not believe that increases of that magnitude should be necessary to promote conservation, even for heavy residential users. At the same time, our preferred approach does not create undue consequences for the other customer classes that will assume a share of the shortfall, as shown in the following table.

Comparison of Average Rate Increase Percent by Customer Class

(Does Not Reflect Subsidies for Medical Baseline Customers)

Customer Class	130% of baseline shortfall
	Allocated to residential only	Allocated to all customers
Residential *	23.63%	12.49%
Small commercial	14.96%	17.94%
Medium/large commercial	14.55%	18.41%
Street lighting	11.50%	14.55%
Large commercial/industrial	15.04%	19.03%
Agricultural	11.42%	14.45%

* Does not include customers with usage up to 130% of baseline quantities

Allocating the 130% of baseline shortfall to all other non-exempt consumption, on an equal cents per kWh basis, produces an equitable overall allocation and appropriate conservation pricing signals. We therefore adopt this approach.

5.6 Tiered Rate Design - Non-Residential

The concept of applying tiered rate design to non-residential rates was raised in the Edison/PG&E rate stabilization proceeding (A.00-11-038, et al.) An ACR issued on March 29, 2001 in this proceeding set a prehearing conference in this case to be heard on a common record with the prehearing conference on rate design in the rate stabilization docket. After that prehearing conference, it was determined that rate design issues for SDG&E should be heard on a separate schedule. However, the April 30 ACR in this proceeding noted that pursuant to § 332.1(f), consideration of tiered rate design for large customers is within the scope of this proceeding.

On this issue, the statements from the public participation hearings, the evidence in this case, and our decision on rate design for Edison and PG&E converge. Simply put, there is no support for this approach, and the record shows that adopting it could be inequitable. In particular, the evidence shows that the level of usage by non-residential customers does not reflect or demonstrate a customer's efficient use of electricity. We conclude that there is no basis for adopting tiered rate design structures for non-residential rates at this time and on this evidentiary record.

5.7 Non-Residential TOU Rate Design

In D.01-05-064, we adopted an approach for designing non-residential TOU rates which spreads the rate increase over all hours, with a slight differential increase on summer on-peak usage (D.01-05-064, Section VI.C). In designing these rates we set the summer on-peak energy rate approximately 5 cents/kWh higher than the average rate increase for that schedule, and allocated the remaining increase to the semi- and off-peak periods (see for example Appendix C of D.01-05-064 for Edison's TOU-8 rates).

We apply a consistent approach here. However, in this case if the summer on-peak rate were set at 5 cents/kWh above the average increase for the schedule, a negligible increase would result for the semi- and off-peak periods. This would not provide the appropriate incentive to conserve during these time periods. Therefore, to provide the appropriate incentive to conserve during all hours, we have designed non-residential TOU rates by setting the summer on-peak rate 2 cents/kWh higher than the average increase for the schedule.