XVI. CRS Mitigation: Capping or Levelizing CRS

Various parties representing DA interests propose that the Commission consider the cumulative economic impact on DA customers of imposing DA CRS charges, and the potential risk of making DA uneconomic for its program participants. These parties propose that the DA CRS be capped at a prescribed amount to limit the adverse economic effects on DA customers that would otherwise result from the increases in electricity charges that would be required to fully fund DA CRS, including the Bond Charges. The shortfall representing the difference between DA CRS costs and the revenues provided by DA participants would be someone else's responsibility, at least for a while. This proposal is a classic example of a cost shift that is inconsistent with our previous adopted policy determinations mandating bundled customer indifference, and contravenes sound policy considerations articulated by the Legislature and Governor. Further, the imposition of a cap may result in an unlawful discrimination. As we discuss below, a subsidy of at least $1.5 billion dollars of DA customers by bundled customers has already occurred; adding to that subsidy is neither good policy nor lawful.

A. Policy and Legal Considerations

Recently, the Commission has expressed the view that the DA program has value for California, and that efforts should be undertaken to avoid making DA uneconomic for the customers who participate. While the Legislature suspended DA by enacting the provisions of Water Code 80110, it did not end DA nor did it repeal or modify the provisions of AB 1890 directing the

Commission to "... take actions as needed to facilitate direct transactions between electricity suppliers and end use customers."⁹⁷ The cost shift issue posed by the DA CRS cap proposal is therefore not about ending or preserving direct access as a matter of philosophy. Rather it is about the use of subsidies, like the kind of cap proposed by the parties, to prop up DA under the conditions imposed on California by the energy crisis. We reject the proposal for caps at this time.

The Governor and the Legislature have made it very clear that cost shifting and subsidies are not permissible devices for use in propping up DA. In September 2001, a few days after our suspension of DA in D. 01-09-060, the Governor stated in a veto message accompanying the return of proposed legislation that would have created a significant exception to the DA suspension mandated in AB 1X,⁹⁸ the Governor said:

"I am returning Assembly Bill 9XX without my signature.

This bill would authorize end-use customers to aggregate their electric loads as individual consumers with private aggregators or as members of their local community with community choice aggregators.

Last June, approximately two percent of the customer load in the territory served by the three investor-owned utilities (IOUs) was receiving power from direct access providers. The Public Utilities Commission (PUC) recently suspended direct access, but the percentage of load subject to direct access transactions grew to as much as 13 percent or more prior to the suspension. That growth creates a significant and unfair cost burden for those customers who continue to receive power from the IOUs and the Department of Water Resources.

This rapid growth in direct access necessitates more concise cost-containment provisions for the remaining IOU customers than those contained in this bill, and those provisions should apply to all direct access contracts.

Moreover, this bill does not clearly authorize fees to cover costs that may result when direct access customers return to service with an IOU, which would create new and unanticipated procurement obligations for the IOU. Those new procurement obligations could come about solely because the direct access provider no longer chooses to provide service to its customers because of rising electricity costs, and instead passes that burden on to the IOU and its customers.

Any efforts to allow direct access must be equitable for all stakeholders."

The Legislature has similarly expressed its intent that all customers pay their fair share of energy costs, regardless of the identity of their supplier. Recently, the Legislature enacted AB 117, which was signed into law on September 24, 2002. Stats. 2002, Chapter 838, effective January 1, 2003. AB 117 provides a limited exception to the suspension of DA mandated in AB 1X by permitting community aggregation programs. In enacting this limited exception, the Legislature expressed its intention that all DA customers pay their fair share of energy costs, without cost shifting.⁹⁹ Public Utilities Code section 366.2(d), as added by that statute provides:

(d) (1) It is the intent of the Legislature that each retail end-use customer that has purchased power from an electrical corporation on or after February 1, 2001, should bear a fair share of the Department of Water Resources' electricity purchase costs, as well as electricity purchase contract obligations incurred as of the effective date of the act adding this section, that are recoverable from electrical corporation customers in commission-approved rates. It is further the intent of the Legislature to prevent any shifting of recoverable costs between customers.

(2) The Legislature finds and declares that this subdivision is

consistent with the requirements of Division 27 (commencing with

Section 80000) of the Water Code and Section 360.5, and is therefore

declaratory of existing law.

This language parallels the language of Ordering Paragraph 3 of D.02-03-055, which ordered that:

3. Direct access surcharges or exit fees shall be developed in A.00-11-038, et al. so that there is an equitable allocation of the DWR costs, so that direct access customers pay their fair share of DWR costs.

The Legislature's adoption of our ordering language in D.02-03-055 evidently reflects their expectation that we will make good on our commitment to develop a DA surcharge that will result in bundled customer indifference and will prevent cost-shifting. Imposing an arbitrary cap on the surcharge frustrates that expectation and breaks faith with the commitment in D. 02-03-055.

A cap cannot limit the ability of the utilities and DWR to recover their costs. If the DA customers do not pay, then the bundled customers will. The Commission has covenanted in the Rate Agreement that there will be no shortfalls in DWR's recovery of power and bond charges. The Commission cannot permit any shortfall, and if a shortfall should occur the Commission must take some action to avoid it including imposing all the costs on bundled customers. Otherwise it could be argued that the Commission has breached this covenant, which would have consequences for the bonds and could possibly trigger a claim of default or lawsuits against the Commission. An argument that the utilities - rather than ratepayers -- could cover the shortfall is contrary to AB 1X, which provides that the end use customers are responsible for payment of DWR charges, not the utilities.¹⁰⁰

Based on the evidence before us, and as more fully discussed below, the caps proposed by various parties will be result in massive cost-shifting and will violate the principle of bundled customer indifference. As TURN points out, a cap effectively requires the bundled customers to subsidize DA customers, contrary to the stated intent of the Legislature and the Governor and contrary to this Commission's commitment in D. 02-03-055.

Further, Public Utilities Code Section 453 constitutes a legal barrier to adopting the kinds of cap proposed by the various parties.¹⁰¹ This statutory provision prohibits granting unreasonable preferences to any customer or class of customer; it is another obstacle for the proponents of a cap. Pub. Util. Code, §453.

The purpose of the DA CRS is to assure that the DA customers as a class and as individual customers pay their fair share of the costs of the service provided to them by DWR and the utilities during the energy crisis. The Governor, in the veto message quoted above, noted the enormous cost shifts that had occurred as the result of rapid migration to DA service during the Summer of 2001. When the Commission voted to ratify that migration in D.02-03-055 it made a commitment to cost recovery from DA customers through a surcharge "in lieu of" a roll-back.

The effect of a cap is to create an exemption from payment of the full amount of those costs by DA customers for some period of time.¹⁰² As a result, bundled customers pay both their own share of costs and some portion of the DA customers' share. This is as true of the bundled customer businesses which are direct competitors of DA customer businesses as it is of the captive residential customers. This circumstance creates a "preference, disadvantage, prejudice...," U.S. Steel v. PUC, 29 C.3d 603, 611 (1981), which violates section 453, Andersen v. Pacific Bell, 277 Cal App. 3d 277, 285 (1988), unless a rational basis for the discrimination can be shown. U.S. Steel v. PUC, 29 C.3d 603, 610-14 (1981) and cases cited therein.

The only rationale offered for a cap is that in the absence of cap the DA program will fail because DA contracts will be rendered uneconomic. In other words, the cap proponents are explicitly contending for a preference in electricity charges in order to sustain an otherwise non-viable program of which they are the sole beneficiaries. The only stated rationale is therefore, discrimination and preference for their own sake. This violates the statute. As the California Supreme Court said in the U.S. Steel case:

...

The constitutional bedrock upon which all equal protection analysis rests is composed of the insistence upon a rational relationship between selected legislative ends and the means chosen to further or achieve them. This precept, and the reasons for its existence, have never found clearer expression than the words of Justice Robert Jackson, uttered 30 years ago. 'I regard it as a salutary doctrine,' Justice Jackson stated, 'that cities, states and the Federal Government must exercise their powers so as not to discriminate between their inhabitants except upon some reasonable differentiation fairly related to the object of regulation. This equality is not merely abstract justice. The framers of the Constitution knew, and we should not forget today, that there is no  [*612]  more effective practical guaranty against arbitrary and unreasonable government than to require that the principles of law which officials would impose upon a minority must be imposed generally. Conversely, nothing opens the door to arbitrary action so effectively as to allow those officials to pick and choose only a few to whom they will apply legislation and thus to escape the political retribution that might be visited upon them if larger numbers were affected. Courts can take no better measure to assure that laws will be  ]  just than to require that laws be equal in operation.'

29 C.3d 603 at 611, emphasis in original.

As the quantitative analysis in Subsection C below demonstrates, the magnitude of the discrimination is such that bundled customers will be suffering a permanent shift of DA customer cost burden. The stated objective for establishing a DA CRS in the first place will be defeated.

B. Parties' Arguments For And Against A Cap

The complexity of the cap determination on a record that is at best incomplete for this purpose is a further obstacle to approval of a cap on the DA CRS at this time. Before a DA CRS cap can be adopted, we must first address (1) what level of cap should be set, (2) under what conditions should the level of the cap be reevaluated, (3) what rate components does it cover, and (4) in what order are costs collected? Questions also arise concerning how the deferred collections in excess of the cap should be financed, and by whom. What interest rate should be applied to the deferred charges, and how can the responsibility for funding the interest be assigned to preserve bundled ratepayer indifference?

D.02-07-032 authorized SCE to establish a "Historical Procurement Charge" (HPC) in the matter of A.98-07-003. SCE was thereby authorized to apply the HPC to DA customers by reducing the DA customers' generation credit by 2.7 ¢/kWh until the effective date of a Commission decision implementing a DA CRS in the instant rulemaking (R.02-01-011). This reduction in the DA surcharge credit was intended to generate $391 million in revenues, thereby providing for equivalent contributions between bundled and DA customers for the recovery of SCE's past procurement cost undercollections.

In D.02-07-032, we noted the likelihood that DA customers would be subject to DA CRS in this proceeding, bond charges in A.00-11-038 et al., "tail" CTC associated with Public Utilities Code Section 367, in addition to the HPC. We observed that the "pancaking" of surcharges in different proceedings may lead to DA contracts becoming uneconomic. We noted that there was a risk of DA contracts becoming uneconomic, and stated in D.02-07-032 that "there should be a cap on the total surcharge levels imposed on DA customers (including the impact of any changes to PX credits)." But the Decision did not set a specific overall cap, and deferred the particular issue to other proceedings for further consideration. (D.02-07-032, pp. 23-25.) A recitation of the arguments and evidence in this proceeding suggests how complex determining the cap would be, and how adopting the kind of cap proposed by the parties might result in discrimination and unfair preference, especially with the lack of adequate record.

CLECA and CMTA argue that a cap should be imposed on the maximum annual CRS that would be billed to DA customers. They claim that the combined effect of SCE's HPC, a charge to recover the DWR historical costs, a charge to recover the DWR Indifference Costs, and a charge to recover the above-market URG costs could make DA uneconomic.¹⁰³ Both parties argue that this is inconsistent with the direction of the Commission.¹⁰⁴ CLECA proposes caps of 2.0 ¢/kWh for PG&E and 2.25 ¢/kWh for Edison and 2.75 ¢/kWh for SDG&E. Because of SDG&E's relatively higher costs, CLECA recommends a 20-year recovery period rather than a 15-year period. It was on the basis of the figures on Table 2 of CLECA's exhibit that Dr. Barkovich concluded that its proposed caps would accommodate full recovery of the HPC, the Bond Charge and the DWR charges, with the significant caveat that recovery would be "over time." The changes CLECA anticipates in these figures does not alter its conclusion, but its numbers only represent approximations. CLECA believes the Commission should utilize the actual figures in the ongoing DWR revenue requirement

proceeding to develop utility-specific DWR exit fees for 2003, and combine them with the approved Bond Charges and the HPC, if one is applicable, under an overall cap. CMTA proposes a uniform cap of 2.0 ¢/kWh be adopted, along with balancing accounts to reconcile exit fee revenues and allocated costs.

CMTA proposes that the Commission sequence the recovery of the various categories of costs under the cap with the HPC procurement costs receiving the highest priority, followed by uneconomic DWR and URG costs. Total charges would remain at the capped level until direct access customers had fulfilled their HPC obligation and were current on their contribution to uneconomic DWR and URG costs. CMTA argues that its recommendation in this regard is consistent with the Commission's recent decision concerning SCE's HPC.¹⁰⁵

SCE believes that adopting a cap is appropriate, and consistent with the Commission's intention to maintain DA as a viable customer option. SCE believes, however, that a 2.0 ¢/kWh cap is too low, and that the cap should initially be set at a level to at least allow the recovery of SCE's HPC and the Bond Charge. SCE believes that setting the cap at 3.0 ¢/kWh will allow recovery of both of these items, with the condition that the first part of the revenues go to the Bond Charge (and to DWR) and the rest of the charges go to recovery of SCE's PROACT. Recovery of the PROACT will help SCE regain its credit worthy standing which was a top priority of the Settlement. Once the PROACT is recovered, SCE can reduce its charges to reflect the underlying cost of service, benefiting all customers. Setting the cap at 3.0 ¢/kWh will also accelerate the recovery of PROACT and allow the DWR above-market costs to be recovered sooner, which will benefit bundled service customers.

But this avoids the issue of ongoing DWR costs and utility costs for which DA customers are responsible. The DA surcharge cap proposed for adoption in this proceeding would cover the surcharges considered in this proceeding; the ongoing CTC, the DWR Bond Charge, and the DWR power. When the Commission addresses PG&E's Historic Undercollection Charge (HUC), we must also consider how the DA surcharge cap could relate to those charges.

SCE argues that it should not be required to finance any deferred collections of DWR revenue requirement attributable to DA customers in excess of a cap. Because the amounts collected for DWR power are the property of DWR, and not the IOUs, SCE argues that DWR should be the entity financing these undercollections. DWR disagrees, however, arguing that DWR has no ability to issue additional bonds or to borrow additional monies to carry shortfalls in DA CRS obligations. DWR proposes that it be paid first from any funds collected under a cap, with IOUs bearing the risk for covering their remaining costs through any remaining funds.

PG&E believes that a cap of 4 ¢/kWh would be reasonable, based on the comparative level of bundled rates that would be the alternative for DA customers. PG&E proposes that the Ongoing CTC be deemed to be recovered first, then the DWR Bond Charges, leaving any shortfall attributable to the DWR. PG&E also proposes that the cap be differentiated by voltage level for Rate Schedule E-20, consistent with underlying rates themselves, to reflect the differing line losses at different voltage levels.

Edison and PG&E are attempting to avoid the costs associated with fronting the money needed to cover the DA customers' costs. PG&E and Edison argue that they should not front and should not carry those costs. SCE argues that i it has no financial capability.

If a DA surcharge cap limits the revenues recovered from DA customers, then the DA shortfall must be collected from bundled customers. In the latter event, however, bundled customers would pay more than was allocated to them under the indifference calculation for that year.¹⁰⁶

PG&E proposes that DWR issue bonds to finance that shortfall. It is within DWR's authorized purpose for issuing bonds. Further, the $11.9 billion total bond issuance contemplated by DWR,¹⁰⁷ which does not take the effects of a cap into account, is well below the statutory limit of $13.4 billion set on DWR's total bond issuance.¹⁰⁸ This approach would require the active participation of DWR in developing the bond issuance to finance the cap. PG&E notes that DWR understands the concept, and did not raise immediate objections.¹⁰⁹ With DWR funding the shortfall, PG&E asserts that customers would then be able to take advantage of the interest rate at which DWR can issue bonds. According to PG&E, under this approach, bundled customers provide the same amount each year as they would to DWR if there were no cap. DA customers pay less in the early years, and more in the later years as they bear responsibility for the bonds issued to finance the effects of the DA surcharge cap.

DWR has no authority under AB 1X and SB 31X to finance recovery of DWR-related costs for DA customers. Thus, we believe this approach suggested by PG&E is not viable.¹¹⁰

PG&E states that under the other approach, bundled customers would provide more to DWR in the early years, relative to the uncapped calculation, and less in later years. An "interest rate" would have to be established, to determine how much additional cost responsibility DA customers would have to bear in the future to "pay back" bundled customers for the extra amount they bore in the early years.

SDG&E favors levelization of annual fixed charges as a preferred approach to mitigating DA CRS, particularly given the relatively higher DWR costs experienced within its service territory. Levelization defers the impact of high-cost contract obligations in the early years to later years. SDG&E is also amenable to an overall cap on DA CRS in conjunction with levelization of the DWR component. SDG&E believes that a 2.7 ¢/kWh cap, encompassing the individual components of the DA CRS, DWR Bond Charge, HPC Charge, and ongoing tail-CTC, would more than cover its costs if its positions were adopted, as set forth below:

DWR Ongoing 1.26 cents

DWR Bonds 0.51

HPC 0.00

CTC 0.70

2.47 cents

However, based upon updated DWR revenue requirements, SDG&E believes the Commission may well adopt a DWR Bond Charge higher than that proposed by SDG&E, pursuant to the terms of the SDG&E-DWR Servicing Agreement and/or Rate Agreement. To the extent that this occurs, and results in the aggregate sum of the components exceeding the 2.7 ¢/kWh cap, such a cap would result in an under-recovery of one or more SDG&E components under the cap.¹¹¹

SDG&E states that an under-recovery would result from the fact that, once adopted, the DWR Bond Charge becomes a non-bypassable charge that must be recovered pursuant to the SDG&E-DWR Servicing Agreement. In much the same fashion, the ongoing tail-CTC is also a non-bypassable charge that must be recovered pursuant to Pub. Util. Code sections 367, 369 and 370. For PG&E and SCE, an HPC charge is expected to remain fixed for a period of one or more years. Consequently, the only remaining element to be under-recovered is the DA CRS.

To the extent that a DA CRS revenue recovery shortfall is caused by the cap, SDG&E believes the shortfall should then be recovered from that IOU's bundled customers and tracked for that IOU. At such time that adequate headroom exists under the cap, DA customers should reimburse bundled customers for that shortfall with interest calculated at the 90-day commercial paper rate. This headroom would develop over time as a result of the completion of the collection of the HPC charge, and possible changes in the level of the DWR Bond Charge and ongoing tail-CTC.

The suggestions that would require either the utilities or DWR to finance the recovery of DWR-related costs for DA customers are not viable options. TURN and ORA raise the further concern as to how capping the DA CRS could adversely affect bundled ratepayers who could potentially be burdened with shouldering the financing costs of excessive deferrals of DA cost responsibility as well as fronting payment of ongoing DWR costs. In effect, their argument is that creating a preference for DA customers comes at a price for the non-participant customers, without any offsetting system benefits. There is no rational basis for capping the DA CRS at this time.

TURN and ORA argue that the Commission must address the risk a cap places upon bundled customers. Financing any revenue undercollection produced by a cap must come from somewhere. (See PG&E cross-examination in McDonald/DWR, Tr. 1, pp. 15-120.) Bundled customer will pay the financing costs by default if another group or entity can not. (RT. 3, pp. 299-302, Marcus/TURN.) The financing will occur at the short-term balancing account rate, which TURN has calculated to be about 7%. (Ex. 18) Depending on the initial level of the cap and the resulting shortfall in revenues, this could result in a significant electricity charge increase for remaining bundled service customers given the magnitude of the DA customer load.

C. Magnitude of DA CRS Cost Responsibility and of Undercollections Caused by a Cap

It would be unfortunate if the Commission were to approve the use of a cap on the DA CRS, even on an interim basis, without understanding the potential impact of that cap on bundled customers. Yet, as the ALJ indirectly acknowledges in the Proposed Decision, the record in this proceeding is insufficient to make that determination.¹¹² In the face of proposals to adopt a cap nonetheless, we must look at the scant information available to determine if there is reason to be concerned. What we can do is consider the nature of the already-exiting undercollection and, using the Navigant model that is part of the record in this proceeding and URG rates adopted in D.02-06-016, develop a crude understanding of the possible ratepayer exposure in the form of DA CSR undercollections.

As discussed above in Section VIII, the costs to be recovered by the DA CRS include several elements - the DWR Bond charge, ongoing DWR power cost obligations and the "un-economic" portion of utility retained generation (URG). We have discussed the applicability of these elements of the DA CRS to differently situated DA customers. The URG element derived from AB 1890, is applicable to all DA customers, regardless of vintage (when the DA arrangement took effect.) The Bond Charge element is applicable to all DA customers except those who were continuously on DA both before and after January 17, 2001 (who never took electric energy service from DWR.) The DWR Power Charge is also applicable to all DA customers except those continuously on DA service.

As we noted above at pages 61-62, the URG and DWR Power Charge elements for 2001 and 2002 for which DA customers are responsible have already been billed to bundled customers and paid by them. They are now essentially a "matter of history." That is, bundled customers have paid amounts over and above their own costs representing costs caused by DA customers but not paid by those customers. The magnitude of those amounts is illustrated as follows for the respective utilities:¹¹³

PG&E SCE SDG&E TOTAL

2001- DWR Power 36,996 30,960 10,403 78,359

2001- URG 28,486 46,846 4,907 80,239

2001- Total 65,482 77,806 15,310 158,598

2002- DWR Power 410,711 374,570 120,395 905,676

2002- URG 114,762 177,209 21,201 313,172

2002- Total 525,473 551,779 141,596 1,218,848

2001-02 Total 590,955 629,585 156,906 1,377,446

The total amount of DA customer costs paid by bundled customers through the end of 2002 is approximately $1.4 billion dollars. If a carrying cost of the utilities' authorized rate of return on invested capital is applied, the total subsidy is nearly $1.5 billion.

A DA CRS that takes effect on January 1, 2003 must cover at least these costs already advanced so that bundled customers may receive the bill credit that we describe (at pages 61-62) as necessary to maintain bundled customer indifference with respect to those costs that they have already advanced for the DA customers. The advance occurred directly as a result of our order in D.02-03-055, which permitted customers who switched to DA during the period between July 1 and September 20, 2001 to avoid paying their share of costs through bundled rates, but with the requirement that they would pay their fair share through a DA CRS.

Given the estimates of DA load calculated by Navigant for each utility, this results in DA CRS surcharges for 2003, representing only repayment of 2001-02 advances by bundled customers, as follows:¹¹⁴

PG&E -- 5.60 ¢/kWh (on DA load of 10,545 GWh)

SCE -- 7.99 ¢/kWh (on DA load of 7,878 GWh)

SDG&E -- 8.51 ¢/kWh (on DA load of 1,844 GWh)

However, a DA CRS that does not cover current costs for 2003 and beyond - including the bond charge that begins effective January 1, 2003 - will result in a new advance by bundled customers, who will be paying the utility and DWR charges for DA customers in their current bills. For 2003, the Navigant model, Scenario 8, develops the following levels of cost responsibility:

PG&E SCE SDG&E TOTAL

Bond Charge 52,945 39,555 9,259 101,759

Power Charge 413,975 450,454 117,731 982,162

URG 109,613 167,593 20,009 297,215

Total 576,533 657,602 147,088 1,381,223

The Navigant-developed estimates of 2003 DA load yield the following levels for DA CRS required to recover 2003 costs during 2003:

PG&E -- 5.47 ¢/kWh (on DA load of 10,545 GWh)

SCE -- 8.35 ¢/kWh (on DA load of 7,878 GWh)

SDG&E -- 7.98 ¢/kWh (on DA load of 1,844 GWh)

For DA customers to be current on their charges in 2003, DA CRS would have to be set at levels approximating the sum of 2001-2002 catch up and 2003 current:

PG&E -- 11.07 ¢/kWh (on DA load of 10,545 GWh)

SCE -- 16.34 ¢/kWh (on DA load of 7,878 GWh)

SDG&E -- 16.49 ¢/kWh (on DA load of 1,844 GWh)

For SCE, an additional 1 cent must be added pursuant to D.02-07-032 to recover the Historical Procurement Charge (HPC) amount of $391 million over time. (D.02-07-032, Findings of Fact #12)

Any cap on the DA CRS results in an ongoing cost shift (discrimination) against bundled customers and a preference for DA customers so long as the DA customers' cost responsibility exceeds the revenue generated by the DA CRS as capped. The DWR Power Charge element of the revenue requirement that must be met by DWR's customers (and by DA customers) is projected to decline as the DWR contracts expire or are renegotiated. At some point, a levelized DA CRS will begin to generate positive revenues as compared with costs so that the bundled customer subsidies are repaid.

Application of an arbitrary 2.7 cent cap on the DA CRS to estimates of costs and loads developed by the Navigant model, as proposed by the PD and the AD of Commissioner Peevey, establishes that even after expiration of all DWR contracts and retirement of all energy bonds in 2022-twenty years from now -- bundled customers of each of the three utilities will not have been rendered indifferent, in the sense that as a class they will not have been repaid for the DA customers' costs they paid for in their bundled rates. Many of the bundled customer businesses who compete with the DA customers may be out of business. The inter-temporal, inter-class inequities are never rectified, in the face of a policy commitment to do so. This is a classic example of undue discrimination.

A similar analysis demonstrates that for the 4 cent cap proposed by President Lynch, the bundled customers of PG&E will be repaid after 2011, but the customers of Edison and SDG&E may not be repaid until after 2022.

D. Issues For Further Consideration In Establishing A DA CRS Cap

The DA cost responsibility and DA CRS levels that must be established for 2003 are crushing, the direct result of our decision to permit expanded DA to persist and our delay in putting a DA CRS in place; DA customers escaped cost responsibility for far too long during 2001 and 2002. They are paying the piper now, or rather repaying the bundled customers who suffered high rates and the further indignity of subsidizing the DA customers' costs. We should explore ways to mitigate the impacts of repaying the bundled customers, while not continuing to foster the illusion that DA is viable if it is in fact not viable without massive subsidies.

Because of the lack of an adequate record and for the reasons stated above, we have rejected a cap at this time. However, there are several issues that might usefully be explored for use in a possible cap determination in the future, when bundled customer indifference is not an issue. This may be when the actual DA CRS has been developed and we have empirical evidence about its affect on DA customers. One consideration in setting a cap is to limit the charges imposed on DA to avoid making DA uneconomic. Yet, the evidence presented on this issue was limited to subjective judgment and anecdotal accounts of discussions with industry representatives. Based on this little evidence, which we believe to be inadequate, we find little basis to quantify the relationship

between the level of a cap and the number of DA contracts that may become uneconomic. In the absence of good empirical evidence concerning the economic sensitivity of DA to various levels of caps, we must weigh the potential impacts of adopting a cap at either the high end or low end of parties' recommendations. Not only do we consider the adverse impacts of imposing a cap that is either too high or to low, we also consider whether effects will be experienced now or in the future. Another consideration is who will pay the interest charges to finance the excess portion of the DA CRS above the cap. We conclude that in order to preserve bundled ratepayer indifference, the interest charges required to finance the cap must be borne by DA customers. If bundled customers were required to fund interest charges to finance DA customers' cap, they would no longer be indifferent since those interest charges would increase total bundled customers' costs. Therefore any cap that is imposed must include within it any interest charges required to finance the excess above the cap.

The timing is also a relevant consideration in setting a cap. The potential risk to bundled customers of setting a low cap is in the potential for large undercollections to build up to a point where bundled customers would be forced to absorb at least some of the debt because DA customers would be financially unable to pay it. This risk grows as a function of time. Thus, bundled customers' exposure to this risk is felt less initially and more over time as any potential undercollection builds up. The timing affects just the reverse in the case of DA customers. The potential risk to the DA program in setting a high cap is felt more at the front end when DA CRS is initially established. If DA contracts become economically non-viable in fact, the risk is that those DA customers will exit the DA program. Because the level of the DA CRS is projected to be lower in the latter years of the DWR contracts, there might be more flexibility to develop a cap in the future as compared with today when costs are comparatively high and the risks of cost shifting are great. That circumstance may change. Once the actual level of the DA charge is established and we have concrete empirical evidence of its impact on DA customers, the issue of a cap may become more amenable to resolution.

Further, parties failed to present any convincing evidence that this preliminary assessment is at an appropriate level. Parties proposing caps as high as 4 ¢/kWh did not provide convincing evidence that a cap at this level could resolve the policy dilemma of avoiding subsidies while avoiding making DA uneconomic. Although certain comparisons were made with bundled rates to argue that a 4 ¢/kWh cap would still be less than bundled rates, and that an increase of that magnitude would be less than the increase that bundled service customers sustained in July 2001, we cannot find such a comparison to constitute convincing proof that DA contracts could survive such an increase.

The other reason cited for the 4 ¢/kWh cap is to avoid the build up of excessively high DA undercollections that could become the burden of bundled customers. While we acknowledge the validity of concerns regarding the potential risk of bundled customers becoming burdened with excessively large undercollections, we view this risk as a potential problem that could grow over time.

On the other hand, a 2 ¢/kWh cap, as proposed by CLECA and CMTA, is clearly too low to cover the requisite components of CRS without triggering unduly large deferred balances. The cap must be high enough to recover the Bond Charge, the Power Charge and SCE's HPC. SCE's ability to regain creditworthy status, and resume procuring electricity to fulfill its net short, is directly linked to its ability to recover the PROACT balance. Therefore, it is important that the HPC is recovered from all DA customers in a timely manner. Pursuant to D.02-07-032, the Commission has adopted a 1 ¢/kWh HPC for SCE as part of the amount to be collected under the cap after a decision in this proceeding is issued.

To the extent that funds provided by DA customers under the 2.7 ¢/kWh are not sufficient to cover both the bond charge and to pay for DA customers' share of the 2003 DWR power charge, any shortfall would have to be remitted to DWR from bundled customers' funds. To the extent that any bundled customers' funds are used to remit any portion of the DA share of 2003 DWR power costs, an interest charge would have to be assessed on DA customers to reimburse bundled customers for the use of their money. The interest charges due to bundled customers for the advance of such funds would be deducted from the gross proceeds from the DA CRS paid under the 2.7 ¢/kWh cap, and credited against the bundled customers pay to DWR. To the extent that after payment of the DWR-related obligations, there were insufficient funds remaining to pay the utilities for above-market URG-related costs, the utilities would have to arrange financing for that amount. It would be an unprecedented and untenable imposition on the credit of the utilities to require them to finance a rate reduction for DA customers.

An initial cap set at the level of 2.7 ¢/kWh might represent an appropriately cautious starting point for a cap, particularly at the very beginning of instituting these charges. It would not impose any abrupt change from the level the Commission has previously referenced as possibly being a reasonable cap value. A cap at this level would promote a bridge on continuity with the preliminary assessment on this issue that the Commission made in D.02-07-032. Once the actual DA CRS is established and we have empirical evidence, as distinguished from hypothetical scenarios, about the impact on DA customers and their ability and willingness to sustain DA relationships, we can revisit the issue and determine whether a cap at the 2.7 to 3.0 ¢/kWh level affects the balance between preserving the viability of the DA program and avoiding subsidies and preferences. We reserve the option to develop a cap prospectively if we determine that such a cap will protect bundled ratepayers against the risk of excessive undercollections imposed by any cap level.

Consideration should be given to alternatives such as having DA customers provide some form of security or collateral to support the repayment of debt generated by the caps. The goal of such collateralized security will be to provide protection against bundled ratepayers bearing potential risk for nonpayment by DA customers, and to attract sources of financing for the debt under favorable arrangements.

As another measure to protect bundled ratepayers, we shall require that any DA customer that returns to bundled service must still pay off their share of the unrecovered charges resulting from the cap. We direct the ALJ to issue a procedural ruling on outstanding issues relating to the cap.

453.(a) No public utility shall, as to rates, charges, service, facilities, or in any other respect, make or grant any preference or advantage to any corporation or person or subject any corporation or person to any prejudice or disadvantage. ... (c) No public utility shall establish or maintain any unreasonable difference as to rates, charges, service , facilities, or in any other respect, as between localities or as between classes of service....