4. Applicability of Stranded Cost Recovery and Net Cost Allocation NBCs
4.1. Forecasted Departing Load
Whether or not departing load should be forecasted and reflected in the IOUs' load forecasts is not an issue in this track of the proceeding. The structure of the load forecasts used in developing the LTPPs has already been addressed in Track 2, and any related issues have been reconciled in D.07-12-052.18 Now in Track 3, we are considering the implications of any forecasted departing loads, as determined in D.07-12-052, on the applicability of NBCs to certain customer groups. This has been raised as an issue in the context of both MDL and CGDL and recognized as an issue within the scope of this track of the proceeding.
The IOUs have taken the position that the Commission has already determined that departing load forecasts should not be a basis for releasing departing customers from having to pay the NBCs. In D.04-12-048, the Commission stated, "In general we agree that the utilities should be allowed to recover their net stranded costs from all customers, which may require the application of additional cost responsibility surcharges or other non-bypassable surcharges." (Finding of Fact 33.) Furthermore, in D.06-07-029, we stated, "It is reasonable, and consistent with law, for the Commission to adopt this limited and transitional cost allocation mechanism to support the development of new generation by having the costs and benefits shared by all customers." (Conclusion of Law 5.) We continue to support these general determinations.
As a part of determining the cost allocation, we need to examine and determine the fair share of certain customers, in particular MDL and CGDL, because of the implications of the LTPP load forecasts that anticipate departing load based on historical trends. This consideration of the fair share is necessary to ensure bundled customer indifference and the proper alignment of benefits and cost responsibility.19 Based on such examination, as discussed below, we have considered the extent to which MDL and CGDL customers will be subject to both the D.04-12-048 and D.06-07-029 NBCs.
In general, the issue revolves around the position of certain parties20 that the IOUs' load forecasts should reflect reasonable amounts of MDL and CGDL, the IOUs should not be procuring for that forecasted DL, there should therefore be no associated costs, and consequently the proposed new generation NBCs should not be imposed on those departing customers.
The principal objections to this proposal have been raised by the IOUs and TURN. PG&E argues this proposal should be rejected because:
1. the Commission has already declined to exclude departing load that is forecast;
2. for policy and planning reasons, forecasting is not an appropriate basis for exceptions;
3. the intervenors have not demonstrated that PG&E has forecast specific departing loads;
4. parties that are not willing to bear the burden of incorrect forecasts should not be excluded; and
5. allowing exceptions based on forecasts will lead to endless litigation and disputes.
SCE adds that departing load introduces additional uncertainty and error into the utility's load forecast and results in additional costs. To avoid unfairly shifting these risks and costs to remaining bundled service customers, according to SCE, all customers taking bundled service at the time resource commitments are made should be responsible for the above-market (stranded) costs of those resources, if any, either through paying the bundled service rate or a CRS designed to recover these costs. SDG&E makes a similar argument.
TURN argues that when the average cost of a utility's supply portfolio is higher than the current market price of power, any departing load - forecasted or not - will increase the average cost to the remaining bundled service customers, resulting in stranded costs.
As noted by the IOUs, the Commission has previously stated that the D.04-12-048 net stranded costs should be recovered from all customers, and the CAM was adopted in D.06-07-029 to support the development of new generation by having the costs and benefits shared by all customers. However, in considering the effects of forecasted departing load on the applicability of the NBCs, we must ensure the outcome of our determination is, to the extent possible, consistent with the preservation of bundled customer indifference and cost recovery from customers on whose behalf resources were procured. In that regard, we must determine the fair share of the departing load for the costs the IOU incurred on behalf of that load. In D.04-12-048, the Commission stated:
A major issue in this proceeding is the extent to which the utilities will be compensated for investments or purchases that they must make in order to meet their obligations to provide reliable service to their customers. The implementation of CCA, departing municipal load, and the potential for lifting, in some form or another, the current ban on allowing new DA all create a great degree of uncertainty as to the amount of load the existing utilities will be responsible for serving in the future. Given the potential for a significant portion of the utilities' load to take service from a different provider, the utilities are concerned that they could end up over-procuring resources and incurring the stranded costs associated with these resources.
One solution to this problem, discussed above, is the adoption of load forecasts that seek to address, to the extent possible, the uncertainties over the future load that the utilities will be responsible for. Another solution is for the utilities to be entitled to recover any stranded costs occurring as a result of their efforts to meet their load obligations.
Given these two possible solutions to offset the effects of departing load, it is necessary to first determine what types of departing load are reflected or not reflected in the adopted LTPP load forecasts.
Issues related to load forecasts were litigated in Phase II, Track 2 of this proceeding and are addressed in detail in D.07-12-052. In determining the appropriate load forecasts, D.07-12-052 relied heavily on the CEC's Integrated Energy Policy Report (IEPR) process and states the following:
The last LTPP decision, D.04-12-048, directed the IOUs to prepare a Medium-Load Plan Scenario in future LTPPs using the CEC's IEPR base case load-forecast scenario or an Alternative Base Case load-forecast scenario, if the utility chose to file one. In R.04-04-003, the predecessor LTPP rulemaking that resulted in D.04-12-048, the assigned Commissioner issued a ruling on March 14, 2005 (hereinafter referred to as the "IEPR Ruling") directing all parties interested in the IOUs' load forecasts for 2006 to participate in the CEC's 2005 IEPR process since the Commission did not intend to re-examine specified issues resolved during the IEPR process. (D.07-12-052, p. 22, footnote omitted.)
We clarify in this decision, and will reiterate in the OIR for the next LTPP proceeding, that the IOUs are to use the CEC's forecast in their LTPPs. The CEC's IEPR process is the proper forum to litigate and contest issues related to each IOU's demand forecast. If an IOU believes that the CEC's forecast is too "conservative" or that the CEC should use different forecasting models, data or other inputs, that IOU must bring those issues up and have them resolved in the IEPR proceeding. (D.07-12-052, pp. 27-28.)
We find it prudent to review load forecast sensitivities, but for purposes of granting procurement authority, need determination should be based on the CEC's base forecast under baseline (1-in-2) temperature conditions pursuant to D.04-12-048. (D.07-12-052, p. 28.)
We concur with many of the concerns raised by the CEC and other parties. To address these concerns and conform to our own policy directives, we base the IOU need determination tables on the CEC's base case, 1 in 2 summer temperature demand forecast (the three need tables, PGE-1, SCE-1, and SDGE-1, all use the forecasts from CEC's 2007 IEPR issued on November 21, 2007). (D.07-12-052, p. 29.)
While we recognize that the 2007 IEPR forecast estimates were not vetted in this proceeding, many aspects of the IEPR forecasting process were. The IEPR process is a public one, involving many of the same participants that are parties to this proceeding, and the IEPR document is a public document. We find it prudent to update the forecast estimates used as inputs in this decision based on the most current public information available to us, particularly given the long time lag that has occurred since the LTPPs were developed. The California Energy Demand Forecast, 2008-2018, the underlying load forecast which the 2007 IEPR assumes, had not been officially adopted by the CEC, as of the mailing of this Proposed Decision. We note that the incorporation of the draft 2007 IEPR demand forecast into our overall needs analysis may give certain parties concern, however, we believe that the draft forecast provides a better `snapshot' of the current needs of the system. (D.07-12-052, footnote 38, pp. 29-30.)
As indicated above, the LTPP load forecasts adopted by D.07-12-052 were based on the November 2007 California Energy Demand 2008-2018 Staff Revised Forecast21 (2007 IEPR Demand Forecast) and are shown in Tables PGE-1, SCE-1, and SDGE-1 of that decision. Each table indicates the forecasts were based on the CEC's 2007 IEPR 1-in-2 peak demand. For PG&E and SCE the tables indicate that the service area calculation includes bundled and DA customers and excludes POUs. This is verified by examination of the 2007 IEPR Demand Forecast, on which the D.07-12-052 load forecasts are based. For example for the year 2007, D.07-12-052 represents the 1-in-2 Service Area Summer Temperature demand to be 19,845 MW for PG&E. This is the load used in D.07-12-052 to determine PG&E's Service Area Surplus (Deficit) for that year. That demand amount can be traced back to the 2007 IEPR Demand Forecast by adding PG&E's bundled load of 18,827 MW and PG&E direct access load of 1,017 MW, as shown on Form 1.5b.
In the IEPR process, IOUs are required to quantify and document their assumptions about migrating load. This information is needed to support compliance with AB 1723 (PRC 25302.5), which require all LSEs to provide the CEC with their "forecasted load that may be lost or added" by a POU or CCA or served by an ESP. The CEC is to perform an assessment of migrating load in each IOU service territory and submit the results to the CPUC.
That bundled load does not include POU load (and the associated MDL)22 is demonstrated on Form 1.4b in the 2007 IEPR Demand Forecast which shows bundled and direct access loads separately from POU loads. In general, forecasts of demand, including that for MDL, reflect historical consumption, economic and demographic projections, weather adjustments and specific inputs from LSEs.23
That CGDL is also reflected in the D.07-12-052 adopted load forecasts can be verified by examination of the 2007 IEPR Demand Forecast in the discussion related to Self Generation. It states, "As discussed in Chapter 1, the peak demand forecast is reduced by the projected effects of the SGIP, CSI and other similar programs. The effects of these programs are forecast based on recent trends in installations." (2007 IEPR Demand Forecast, p. 74.) 24 Historic and forecasted CGDL peak demand is shown as "Total Private Supply" on Form 1.4 of the 2007 IEPR Demand Forecast.
The above discussion of how departing load is reflected in the adopted load forecasts is also consistent with our understanding of how departing load is considered in the load forecasts prepared by the IOUs.25 In D.07-12-052, we stated:
Regarding parties' concerns over PG&E's assessment of departing load, we concur with PG&E's response that its analysis of system need is not impacted by possible future load shifting due to DA and CCA, and that future DG and MDL is captured by historical trends used to develop the forecast.26
Similar statements are made with regard to SCE and SDG&E departing loads.27
Based on the load forecasts adopted in D.07-12-052, it would be appropriate to employ both of the solutions expressed in D.04-12-048 related to IOU procurement for departing load. For IOU customers that are eligible to, and do, choose DA service from an ESP and for customers that decide to use a CCA, D.07-12-052 indicates that their loads are included in the adopted load forecasts on which the LTPPs are based. Therefore, the IOUs would be procuring resources on their behalf, and NBCs should be imposed on these customers when they cease taking procurement services from the IOUs, in order to maintain bundled customer indifference. Imposition of the NBCs is appropriate, because CCA, and to an extent customers currently eligible to return to DA, create uncertainty regarding what loads the IOUs will be required to serve. At this time, there is insufficient history of such transactions and limited knowledge of customers' intent to pursue such transactions in the future, for the IOUs to use in determining how much, or how long, power should be procured on such customers' behalf. Planning for these customers' needs and imposing the NBCs if and when these customers choose alternative procurement services is a reasonable way to address the problem.28
On the other hand, D.07-12-052 indicates that future CGDL and MDL are captured by historical trends used to develop the load forecasts. Therefore, the forecasted loads associated with MDL and CGDL customers are not included in the D.07-12-052 adopted load forecasts. This is consistent with the solution expressed in D.04-12-048 whereby the Commission would adopt load forecasts that seek to address, to the extent possible, the uncertainties over the future load that the utilities will be responsible for.
We note that the use of historic information and trends to reflect future departing load reduces some risk to the IOUs of possibly adopting overly optimistic estimates and tends to limit the dispute and litigation related to what the appropriate levels of departing load should be. For instance, PG&E states that MDL bypass is no longer expected to be materially different than recent trends captured in the historic data.29 While there may be differences between the amounts of departing load implicit in the load forecasts and the amounts recorded on a year-by-year basis, over time any such variations should level out and bundled customer indifference will be maintained. Also, as long as historic information and trends are the basis for reflecting the departing load in the load forecast, unexpected annual variations between actual and assumed departing loads will result in the assumed forecast departing load levels being adjusted up or down in the future based on the historic amounts, again resulting in bundled customer indifference being maintained over time.
Forecasting the effects of CGDL and MDL has been done in the past30 and, as discussed previously, is done as part of the CEC's IEPR process which, at least for the foreseeable future, will be the basis for Commission's LTPP load forecasts. We reiterate the guidance provided in D.07-12-052, that the CEC's IEPR process is the proper forum to litigate and contest issues related to each IOU's demand forecast, including any concerns related to the accuracy of the predicted MDL and CGDL in the CEC forecast.
What we must consider now is (1) what it means for this departing load to be reflected in the load forecast, and (2) given that meaning, whether these departing load customers should be fully responsible, partially responsible, or not responsible at all, for the new generation NBCs established by D.04-12-048 and D.06-07-029.31 This is integral to our determination of the departing load's fair share.
Exclusion of MDL and CGDL from the load forecast can only logically be interpreted to mean that the LTPP, which uses that load forecast to determine resource needs in the forecast year, does not include any resources to serve that departing load in that forecast year and beyond. Accordingly, in such circumstances, it would be reasonable to determine that the fair share of departing load for paying the new generation NBCs would be zero. Stranded costs are avoided for these forecasted departures via the combination of (1) the layering of generation procurement by the IOUs (both in terms of procurement of longer term, shorter term and "spot" market resources and in terms of the sequenced procurement of resources which in turn results in resources regularly dropping out of the portfolio as contracts expire) and (2) forecast increases in load from new and existing customers.
As discussed below, in applying this departing load fair share concept, we have considered new generation resources that become operational (1) during the year that these customers depart and beyond and (2) before these customers depart.
For those new generation resources that become operational during the year that MDL and CGDL customers depart and beyond, those departing load customers should not be responsible for any new generation NBCs. That is because when the commitments for those resources are made the load forecasts on which procurement needs are based do not include loads related to MDL and CGDL. Such departing loads have been forecasted and are not included in the load forecasts used in determining the need for those resources. Those resources are therefore not procured on behalf of these departing load customers for any time period and their fair share of the costs should be zero.
We must also consider cost responsibility related to those new generation resources that become operational and begin to provide energy prior to the date that these customers depart the IOUs' systems. For transferred MDL and CGDL customers, they would have taken bundled service from the utility, for some period of time, prior to the year in which they depart. For the time that they are bundled service customers, they would pay for any operative new generation resources as part of their bundled service rates. However when they depart, their cost responsibility for such resources should end. That is because, at the time the resource commitments are made, (1) the LTPP load forecasts exclude forecasted amounts of MDL and CGDL; (2) these customers will eventually become the departing customers for which those amounts of MDL and CGDL are forecasted; and (3) therefore, in effect, these customers' loads are only reflected in the LTPP load forecast for the years in which they are bundled service customers. Therefore, (1) the IOUs' procurement needs related to these customers are only identified and planned for in the years in which they are bundled service customers; (2) the IOUs' procurement commitments are made on behalf of these customers only for the time that they are on bundled service; and (3) these customers' fair share of the costs related to these resources should be zero after they depart.
Consistent with our overall guiding principles for resolving NBC implementation issues, these departing customers should not pay any NBC related to new generation resources that were not procured on their behalf, as these customers' fair share would be zero. We will not impose either the D.04-12-048 or D.06-07-029 NBCs on MDL and CGDL, since these classes of departing load are reflected in the load forecasts on which the LTPPs are based. Also, since there are no resources or associated costs in the forecast year related to the load departing in that year, there is no cost shifting to bundled customers when these departing customers leave.
In supporting the IOUs, TURN argues that the question of whether or not some amount of future departing load may have been reflected in a prior forecast sponsored by a utility or adopted by this Commission or the CEC should be irrelevant to the applicability of the NBCs at issue in this proceeding. It is TURN's position that when the average cost of a utility's supply portfolio is higher than the current market price of power, any departing load - forecasted or not - will increase the average cost to the remaining bundled service customers, resulting in stranded costs. We do not agree with TURN's conclusion that therefore under all such circumstances departing load customers should be assessed an NBC.
The more important consideration is the appropriate measure of ratepayer indifference. All other things being equal, exclusion of forecasted departing load from the LTPP load forecasts and exclusion of MDL (with the exception of large municipalizations) and CGDL customers from cost responsibility for new generation resources after the customers depart leaves existing bundled customers with the same cost responsibility as was anticipated when the LTPP load forecasts were made. That is simply because the forecasted departing customers were not anticipated to be served after they depart because their loads are excluded from the forecasts on which the procurement decisions are based. The fact that the forecasted departing customers actually depart does not affect the costs to the bundled customers when compared to costs associated with the assumptions in the sales forecasts and procurement plans associated with the new generation resources. In that regard, bundled customers are appropriately indifferent to the departure of the forecasted departing customers.
To summarize, as opposed to DA and CCA, MDL and CGDL do not create a large degree of uncertainty regarding what loads the IOUs will be required to serve. The adopted load forecasts directly address the effect of MDL and CGDL, and the consequent LTPPs are not developed to serve those departing loads. These forecasts justify our determination that the fair share of these departing load customers will be zero. Accordingly, imposition of the D.04-12-048 and D.06-07-029 NBCs is not necessary for MDL or CGDL customers. However, with a large municipalization, we take a different approach as discussed in Section 4.1.4 below.
4.1.3. TURN's Recommendation for a Binding Notice of Intent Process
TURN argues that a binding notice of intent (BNI) process provides a much more robust way of dealing with the uncertainty regarding future departing loads than endless debating over who included or should have included which potential departing loads in a past forecast. TURN notes Commission has adopted this approach for CCA load.
In general, we agree with TURN's position that if a potential departing customer is not willing to commit to a firm departure date via a BNI, then that customer should remain liable for the potential stranded costs associated with any commitments the utility enters into prior to the date of the actual departure. However, that customer should only be responsible for commitments that were made on its behalf. This principle is embodied in the determination of the fair share. In the case of CCA, the IOU's are procuring and making procurement commitments on behalf of potential CCA customers until the specific dates indicated by the BNIs. That is because loads associated with these customers are included in the IOUs' load forecasts on which their procurement decisions are based. That is not the case for MDL and CGDL customers with respect to the new generation NBCs. As indicated by D.07-12-052, the IOUs exclude these departing loads in their forecasts. As stated previously, for this reason, the IOUs are essentially not procuring on behalf of MDL and CGDL customers in the year they depart and beyond. Accordingly, it is reasonable to determine that the fair share for the new generation costs would be zero. Although the BNI process may be a viable approach for determining when IOU procurement on behalf of certain customers ends, it is not relevant in addressing the NBC applicability issue of whether these customers should be assessed any NBC at all under a fair share analysis.
As discussed above, our analysis of the fair share cost responsibility for MDL is based in large part on our determination that such load is reasonably reflected in the historical trends used in developing the adopted LTPP load forecasts. However, at some point the historical trends of MDL may no longer reasonably represent the amounts of MDL that will occur. This point would be reached if there is a "large municipalization" in the forecast year. While there is no precise measure of what constitutes a "large municipalization," in the context of this decision, we are defining "large municipalization" as any portion of an IOU's service territory that has been taken control of or annexed by a POU where the amount of load departing the IOUs' service territories due to the municipalization is of such a large magnitude that it cannot reasonably be assumed to have been reflected as part of the historical MDL trends used in developing the adopted LTPP load forecasts. SCE states that its long-range retail load forecasts use historical data starting from 1991 and that all sizeable annexations occurred prior to 1991.32 PG&E indicates that it would likely remove any large municipalizations from the historical data but adds that it is difficult to quantify what "large" would be.33 SDG&E indicates that it has no existing or planned municipalization at this time.34
Therefore, if a large municipalization occurs in a particular year, the associated departing load would logically have been part of that year's LTPP load forecast, and the IOUs would have been making new generation resource commitments on behalf of those departing customers up until the time they depart or provide appropriate notice of departure. Unlike that for MDL and CGDL, which are reflected in the LTPP load forecasts, it cannot be argued that large municipalization load, which is not forecasted for LTPP purposes, should be excluded from new generation cost responsibility. That is because when the commitments for resources that were procured prior to these customers' departure are made, there is no forecasted information that would indicate that customers would be departing due to large municipalizations at any time over the lives of the resources. Procurement would have been planned accordingly. Therefore, under the principle of allocating fair share, large municipalization departing customers should be fully responsible for the new generation NBCs.
Imposition of new generation NBCs on customers departing due to large municipalizations shall be accomplished through a separate application filed by the affected IOU. Customers' NBC cost responsibility shall be determined through a fair share analysis based on the record of that proceeding. The IOU has the burden to show the departures are within the definition of a large municipalization, especially as it relates to how the large municipalization is or is not reflected in the adopted LTPP load forecasts.
For purposes of determining when the IOUs should stop procuring new generation resources for these departing customers, a BNI process similar to that established for CCAs is reasonable.35 As determined in D.04-12-048 and D.05-12-041, customers choosing CCA service will be responsible for new generation NBCs associated with the resources procured prior to departure if no BNI is submitted. If a BNI is submitted, the customer will pay only the NBC associated with new generation resources procured prior to the date the BNI is submitted to the IOU. In the event that the CCA cannot meet the BNI date, the CCA will be liable for any net incremental procurement costs incurred by the utility. The IOUs should make this process available for large municipalizations.
If the large municipalization entity does not wish to provide a BNI, the actual departure date is a reasonable date to determine large municipalization NBC cost responsibility, similar to that for CCAs. While we prefer this BNI process, we will not strictly impose its conditions, recognizing that there may be a reason, possibly having to do with the timing of the processes in finalizing a large municipalization, for the entity not to choose either the use of the BNI or the actual departure date, but to recommend some alternative date instead. However, we will impose the burden on the large municipalization entity to demonstrate the reasonableness of using its proposed date as opposed to a BNI date or the actual departure date for determining when the IOU should no longer procure new generation resources for the departing customers.36 If that burden is not met, the actual departure date will be used for that purpose.
4.1.5. New Western Area Power Administration (WAPA) Departing Load and Split Wheeling Departing Load
PG&E requests that it be made clear that the new generation NBCs also apply to new WAPA departing load37 and split wheeling departing load,38 consistent with D.03-09-052 and D.06-05-018.39 In those decisions, the Commission determined that:
A CRS shall be imposed on split wheeling preference power customers to the extent they received a portion of their power through PG&E bundled service to the extent such power exceeds the customer's CRD in the manner contemplated under the existing provisions of Contract 2948A. (D.03-09-052, Ordering Paragraph 4)
The following new Ordering Paragraph (OP) 8 is added (reordering current OPs 8 and 9): "PG&E is directed to promptly file an advice letter with the appropriate amendments to its tariff to bill and collect CRS and other applicable nonbypassable charges from preference power customers consisting of `Additional Customer Load' relating to the specific list of delivery points listed in Appendix C of the WDT Agreement, that have taken bundled service from PG&E on or after February 1, 2001, and subsequently reduced or terminated such service to take electric service from WAPA or another similarly situated entity." (D.06-05-018, Ordering Paragraph 5)
In its testimony, PG&E included such loads as being subject to the D.04-12-048 NBC.40 No party provided responsive testimony or other evidence that shows such loads should be excluded from that charge.
Regarding the D.06-07-029 NBC, the Commission stated:
In summary, Section 380 allows an IOU to recover the costs it incurs to sustain "system reliability and local area reliability" from all customers "on whose behalf the costs are incurred." We construe benefiting customers as defined in Section IV.B.1 as those customers on whose behalf the costs are incurred. (D.06-07-029, p. 41.)
No party provided testimony or other evidence that would indicate that PG&E would not incur the D.06-07-029 NBC related costs on behalf of new WAPA departing load and split wheeling departing load customers while they are customers of PG&E. Therefore, we conclude that the new generation NBCs should apply to new WAPA departing load and split wheeling departing load consistent with our rationale for assigning generation related cost responsibility in D. 03-09-052 and D.06-05-018.
4.2. AReM's Request for Confirmation Regarding Customers Currently Eligible to Return to DA
AReM asks the Commission to confirm that bundled service customers who are eligible to return to DA should also be exempted from the NBC associated with D.04-12-048. PG&E, SCE, SDG&E, and TURN oppose AReM's request.
In D.04-12-048, the Commission authorized the IOUs to recover the stranded costs of new utility procurement resulting from departing load from "all customers, including departing [load/customers]."41 AReM argues that, when read in context, this wording specifically excludes customers that are currently eligible for direct access. That is, "departing load" and "departing customers," as used in D.04-12-048, do not include customers that are currently on direct access or customers that are currently on bundled service but are eligible for direct access.
The opposition's principal response to AReM's assertion is that the Commission in D.04-12-048 determined that the IOUs should be allowed to recover stranded costs from all bundled customers, including departing load customers. There are no stated exceptions.
AReM supports its conclusion by citing D.03-12-059 and D.04-06-011 where the Commission stated various customers that are currently ineligible for direct access should be obligated to pay for stranded costs for 10 years.
In reply, PG&E states that in D.04-12-048, the Commission referenced these two decisions to support its decision to limit NBCs to 10 years. It did not cite these decisions as a basis for excluding DA eligible customers. PG&E adds that notably, just before the language quoting these two decisions, the Commission states that stranded costs should be recovered from all customers, which would include DA eligible customers.
AReM also references D.05-09-022 which addressed various petitions for modification of D.04-12-048, including the petitions filed by AReM and ESPs in which it was argued that the Commission does not have the authority to impose NBCs on direct access customers for purposes of allowing the IOUs to recover stranded costs associated with new procurement commitments. The Commission held, "[W]e may lawfully hold future direct access customers responsible for the recovery of new generation costs."42 AReM emphasizes the word "future." AReM argues that the Commission made no reference to customers that are currently eligible for direct access, indicating that would have been a glaring omission if the Commission had actually intended for the stranded cost NBCs authorized in D.04-12-048 to apply to such customers.
In reply, TURN states that today's bundled service customers who happen to be eligible for DA and subsequently depart to take service from an AReM member are precisely "future" direct access customers as specified in D.05-09-022. TURN also states that the mere fact that the Commission did not single out "currently bundled customers who are eligible to return to direct access" from other types of departing load does not prove AReM's point. If anything it proves the opposite - that all types of departing load are subject to the NBC.
AReM also states that arguments for imposing the charge on DA eligible customers ignore the existence of the elaborate rules developed by the Commission to govern the movement of DA-eligible customers to and from direct access so as to prevent gaming and costs being shifted to bundled customers. Under those rules, if a customer that is on direct access wants to return to bundled service, it must provide the utility with six months advance notice and will only become eligible to receive bundled service from the utility at the same rate as other customers at the end of the notice period. In addition, the customer is required to remain on bundled service for a minimum of three years, and if the customer wants to go back to direct access after the end of its minimum three-year commitment period, it must provide the utility with six months advance notice.
AReM argues that the Commission left open the possibility that it would later extend the minimum commitment period beyond three years if there was evidence that a longer period was "necessary to avoid stranding long-term portfolio supply obligations undertaken to serve DA customers returning to bundled status...."43 According to AReM, the Commission has not seen a need to do so, because the Commission's rules to prevent cost shifting by DA customers also ensure that DA customers impose no costs on bundled customers. AReM states that any costs incurred by the utility in its long-term procurement are incurred solely for the benefit of bundled customers, and since customers that are currently eligible for direct access do not create stranded costs when they move to direct access, imposing the stranded cost NBCs on such customers would be inconsistent with the principle that costs should be allocated on the basis of causation.
In response TURN argues the potential for stranded costs resulting from load departing from the bundled portfolio is exactly what the relevant portions of D.04-12-048 were all about. Rather than "not seeing a need" to address the circumstances described in D.03-05-034, the Commission saw a need and addressed it by adopting the stranded cost NBC. TURN adds that AReM's further statement that: "any costs incurred by the utility in its long-term procurement are incurred solely for the benefit of bundled customers" proves TURN's point. Currently bundled customers who happen to be eligible for direct access are just that - bundled customers, the very people for whom the utility is incurring costs.
We do not adopt AReM's request to confirm that bundled service customers who are eligible to return to DA should be exempted from the NBC associated with D.04-12-048. We generally agree with the responses by the IOUs and TURN as detailed above. None of the decisions cited by AReM specifically exclude these customers from the charge.44 In D.04-12-048, we found that the stranded costs should be recovered from all customers and did not indicate any exceptions. By our decision today, we have addressed the implementation of the D.04-12-048 NBC by employing the previously used principles of bundled customer indifference and customer responsibility for costs incurred on their behalf. We consider this to be logical and fair, and consistent with the principle of these customers paying their fair share for costs incurred on their behalf, and of preventing cost-shifting. We do not see such logic or fairness in AReM's request.
As described by AReM, there is a detailed process by which certain customers can return to DA service. However, until these customers return to DA, they are no different from the other bundled customers on whose behalf the IOUs are making procurement related decisions. Until the proper notice is given, the IOUs have no way of knowing if and when such customers will depart. The IOUs therefore properly include the related loads of the potential DA customers in their load forecasts. By doing so, the IOUs are procuring and making procurement commitments on behalf of these customers. As is the case with all other customers, these customers should be subject to the D.04-12-048 NBC for procurement commitments made on their behalf up until the date they provide notice to the IOUs of their intent to return to DA.
4.3. Above-Market Standard Offers for New QF Contracts
In D.07-09-040, dated September 20, 2007, the Commission ordered that the utilities make standard offer contracts available to existing qualifying facilities (QFs) with expiring PPAs or to new QFs. PG&E argues that this requirement, similar to RPS and RA requirements, impacts utility procurement and creates uncertainty in resource planning, and to the extent the prices in the new QF standard offer contracts are above-market prices, bundled customers may incur additional stranded costs. In its opening brief, PG&E requested that the Commission, in this decision, affirm that stranded costs associated with these contracts can be recovered under D.04-12-048 or D.06-07-029.45 In reply briefs, SCE agreed with PG&E's request. No other party replied on this topic.
We agree that the IOUs should be able to impose NBCs for the above market costs of these new QF contracts. This can be accomplished through the D.04-12-048 NBC, and we will authorize that NBC for this purpose. However, there has been no demonstration of need for cost recovery of these new QF contracts through the CAM that was authorized by D.06-07-029, and we will not do so. The CAM was designed to get new system reliability resources built and the resigning of QF contracts does not accomplish that. Even for contracts with new QFs, cost recovery under the CAM may not make sense due to the requirements and costs associated with the energy auction process.
4.4. Other Applicability Related Issues that Will Not Be Addressed in this Proceeding
D.04-12-048 and D.06-07-029 established the NBCs at issue here in Track 3 of this proceeding. In general, Track 3 was intended to address implementation issues related to NBCs. That scope was modified slightly to include the issue of determining the fair share of DL liability of the new generation NBCs. Our obligation is to reconcile issues properly within the established scope. TURN's BNI proposal directly relates to this issue and we felt it necessary to address AReM's request for clarification. We also felt a need to address PG&E's request regarding the inclusion of new QF contracts, since it is relevant to the applicability of the NBCs at issue and came about because of our recent decision on the matter. However, there were also other issues that related somewhat to the applicability of NBCs which were identified and addressed by certain parties in the Track 3 briefing process. They include such things as:
· There is a lack of statutory basis for NBCs;
· Utilities should not be able to recover NBCs for procurement costs arising in the normal course of business;
· NBCs will "chill" combined heat and power and CGDL development;
· The benefits of CGDL justify an exclusion to the NBCs; and
· Imposition of the stranded cost NBCs on customers currently eligible for direct access would hamper retail competition.
While many of these issues may have been rendered moot by our resolution of the applicability of the NBCs as they relate to DL, they are also outside the scope of this track of the proceeding and will not be addressed in this decision. Such issues should be, or should have been, pursued in the proceedings that established the charges, not in this proceeding which was principally designed to implement the charges. To fully address such issues now would not be fair to the parties that did not fully address the related arguments in briefs. Those parties, with good reason, assumed the issues were beyond the scope of the proceeding and treated them accordingly, and so shall we.
18 The LTPP Phase II, Track 2 decision in this proceeding.
19 Addressing this issue now is consistent with D.07-11-051 wherein the Commission, in modifying D.06-07-029, stated, "Our definition of benefiting customers subject to the cost allocation mechanism does not include current POU customers, and departing customers who take POU service will not be able to avoid cost responsibility pursuant to D.04-12-048, as modified by D.05-12-022. As noted in D.04-12-048, Ordering Paragraph 9, IOUs are required to forecast and plan for departing load as they file their biennial long-term procurement plans which establish each IOU's long-term resource needs. Further, we will consider issues of need in a subsequent phase of this proceeding and POUs may address whether specific facts suggest refining our approach to the allocation of costs to municipal departing load." (Ordering Paragraph 1(h), emphasis added.)
20 This position is advocated by CMUA, Merced ID, Modesto ID, and CCDC, each on behalf of its specific interests.
21 http://www.energy.ca.gov/2007publications/CEC-200-2007-015/CEC-200-2007-015-SF2.PDF
22 The 2007 IEPR Demand Forecast, p. 35 indicates that the individual LSE forecasts were also adjusted to account for load migration (customers migrating from one service provider to another).
23 2007 IEPR Forecast, p. 35.
24 The referenced Chapter 1 states in part, "To forecast future self-generation load, staff used the IOU reports on completed new interconnections and pending applications to develop projections of capacity additions of new interconnections. (Footnote omitted.) The interconnection reports provide a detailed picture of capacity addition trends."
25 Regarding the load forecasts prepared by the IOUs, the effects of CGDL were reflected, as indicated in each IOU's LTPP which included a section on forecasting DG (PG&E's 2006 LTPP, Volume 1, pp. IV-20 - IV-25, SCE's 2006 LTPP, Volume 1B, pp. 16-24, and SDG&E's 2006 LTPP, pp. 194-197). It was also established that the CEC demand forecast reflects embedded amounts of DG (CCDC, Wong, 8 Tr. 1046-1047). MDL is implicitly reflected in SCE's load forecast as a decline in SCE's bundled load growth through the extrapolation of historical data. (See Exhibit 37, p. 37 and SCE, Canning 2 RT, pp. 216-218.) PG&E similarly takes projected POU departing load into account in its load forecast. (See PG&E, Aslin, 5 RT, pp. 647-660.)
26 D.07-12-052, pp. 34-35.
27 Id. at pp. 39 and 42.
28 In its Opening Comments on the Proposed Decision, p. 5, AReM indicates that SDG&E's approved load forecasts reflect reasonably foreseeable DA load migration and that it follows that based on the rationale for excluding MDL and CGDL customers from new generation NBC cost responsibility, DA-eligible customers located in SDG&E's service territory should also be exempt from the new NBCs. The record indicates otherwise. Form 1.4b of the 2007 IEPR Demand Forecast shows a constant forecast of DA demand over time for SDG&E. Also, in its Reply Comments on the Proposed Decision, SDG&E clarified that while DA load shown in its compliance capacity tables increase slightly over the planning period, this increase is driven by the assumption that usage per customer increases slightly over time not and not a forecast of DA customer increases. Therefore, it is evident that DA load migration has not been forecasted for SDG&E, and DA eleigible customers in SDG&E's service territory should not be excluded from the new generation NBCs..
29 Exhibit 211, Response to Question II.2.
30 CGDL has been excluded from procurement related charges based on load forecasts in previous Commission decisions. (See D.03-04-030, p. 54, D.04-12-048, Ordering Paragraph 11, and D.04-10-035, p. 20.) Regarding MDL, the Commission excepted transferred MDL indentified in the Bypass Report in D.04-11-014 (see pp. 4 and 40), Findings of Fact 3 and 18, Conclusion of Law 5, and Ordering Paragraph 4); the Commission excepted new MDL associated with the transferred MDL identified in the Bypass Report in D.04-11-014 (see pp. 4-5, 21, Findings of Fact 10 and 11, and Ordering Paragraph 2) and the Commission granted an exception to new MDL of "existing" POUs in D.03-07-028 (see p. 61, Findings of Fact 12, 13, and 16, Conclusions of Law 9 and, 10, and Ordering Paragraph 6) and extended it to other new MDL on the basis that new MDL was implicitly accounted for in the utility forecasts (see D.04-11-014, pp. 10-13, Findings of Fact 2 and 4, and Conclusions of Law 1 and 3, and Ordering Paragraphs 1 and 2).
31 The fact that system needs are not impacted by possible load shifting due to DA and CCA means that the load forecasts are not reduced to reflect DA and CCA. It is therefore unnecessary to examine the implications of forecasted load reductions in this context, and no party has recommended that we do so.
32 Exhibit 212, Response to Question 2.
33 Exhibit 211, Response to Question II.2.
34 Exhibit 213, Response to Question 8.
35 CCAs and large municipalizations are similar in that there is potential for significant load migration and neither is reflected in the LTPP load forecasts.
36 If the large municipalization entity is of the belief that, at some point in time, the IOU should have known the load would be departing by a certain date, the municipal entity should explain why a BNI commitment could not have been made by the municipal entity to reflect that.
37 New WAPA departing load is additional customer load of certain so-called `new allottee" customers who, for example, were served by PG&E under its retail tariffs prior to expiration of Contract 2948-A with WAPA but are now served by WAPA.
38 Split wheeling departing load is that portion of the load of certain so-called "split-wheeling" customers which, for example, was served by PG&E under its retail tariffs prior to the expiration of PG&E's Contract 2948-A with WAPA but is now served by WAPA.
39 PG&E Initial Comments on the Proposed Decision, p. 12. There were no replies to this comment.
40 See Exhibit 7, p. II-5.
41 D.04-12-048, pp. 60 and 63.
42 D.05-09-022, p. 15.
43 See D.03-05-034, Finding of Fact 13.
44 AReM is correct that D.03-12-059 finds: "Although Edison established a need for Mountainview, in order to not over-burden ratepayers in the early years of the contract, we adopt TURN's proposal that all customers of Edison that are currently ineligible for direct access be obligated to pay for stranded costs for the first 10 years of Mountainview's life." (Finding of Fact 22.) However, at the time of the decision, dated December 18, 2003 the relevant former DA customers who had returned to bundled service would have been "currently" (as of the decision date) ineligible for DA, because of the three-year commitment obligation established by D.03-05-034.
45 Evidentiary hearings in Track 3 concluded on September 21, 2007. The opening brief was the first real opportunity for PG&E to raise this issue in this proceeding.
