IV. Proposed Settlement Agreement
Given the broad-based support among the parties for the proposed settlement agreement, we must give it serious consideration. At the same time, we acknowledge SDG&E's questioning the legitimacy of a settlement entered into by two parties (who already largely agreed with each other) in which those two parties agree to shift costs to a third, non-settling party. (SDG&E Comments re Proposed Settlement, p. 27.)
The proposed settlement divides DWR's revenue requirement into three categories, referred to as: (1) as DWR's contract costs; (2) DWR's other power costs; and (3) planned changes in power charge accounts. (Motion of Settling Parties, p. 4.) Each category receives its own allocation approach.
The proposed settlement would start by allocating DWR's contract costs on the same basis that the contracts were allocated for operational purposes in D.02-09-053. This method is generally referred to as the "cost-follows-contracts" or "CFC" methodology, and was also generally advocated in the litigation positions of PG&E and SCE.
This initial CFC-based allocation would then be adjusted by "Fixed Annual Adjustment Amounts." (Brief of Settling Parties, pp. 8, 14.) According to the Settling Parties, using these fixed annual adjustment amounts results in the projected "above market costs" of DWR's long-term contracts being allocated to the customers of PG&E, SCE and SDG&E based on utility allocation percentages of 43.6%, 42.6%, and 13.8%, respectively.7 These percentages "[A]re designed primarily to constitute a reasonable reflection of the relative net-short positions of the three utilities that DWR initially sought to serve when it entered into its contracts, and the other equity "yardsticks" advanced by the parties to this proceeding." (Id., p. 8.)
The same utility allocation percentages would be applied to DWR's other power costs, which include all of DWR's administrative, general, and extraordinary item expenses, and any new cost categories specified by DWR not directly related to a specific contract. (Id., p. 7.)
For the third category, planned changes in power charge accounts, which reflects the planned annual changes in the operating reserves maintained by DWR, different percentages would be applied, of 44.4% for PG&E, 45% for SCE, and 10.6% for SDG&E. These percentages represent the currently adopted allocation percentages for DWR's bond charge revenue requirement. (Id., p. 7.)
The proposed settlement agreement is in fact an odd hybrid. It starts from a CFC approach, which in its pure form has the advantages of simplicity, ease of administration, and not requiring the use of confidential information. A pure CFC approach does, however, have one serious problem - it is simply not equitable, as even its advocates will admit. (SCE Opening Brief, p. 29.)
The benefits and flaws of a CFC approach are well summarized by SCE's witness:
As Edison's testimony states, CFC allocation methodology provides certain operational and administrative benefits that none of the other allocation methodologies provide, but we were also very clear in our testimony that it is not cost-based, but neither are the other allocation proposals; it is not equitable, and neither are most of the other allocation proposals; but that in -- in large, when you look at all the factors considered, it's a reasonable way to allocate these costs for what is a very difficult decision the Commission has to make. (SCE witness Cushnie, Transcript p. 7237.)
The allocation of fixed costs resulting from a CFC approach is somewhat arbitrary, as we noted in D.02-12-045:
Since DWR signed contracts for a statewide need, allocating the fixed costs of contracts to utility service territories based upon geographic location does not match how or why those contracts were obtained. It would be arbitrary and unfair for one or more service territories to end up with a disproportionate number of high-priced contracts when DWR was not trying to balance costs among service territories. (Id., p. 11.)
In order to remedy this problem with the CFC approach, the proposed settlement adjusts the CFC allocation through the fixed annual adjustment amounts. The logic behind this approach is explained:
The customers of the utilities must take power from DWR, and must bear the costs of that power. To the extent that the costs are equal to the market costs for the power, there is no burden associated with the requirement that customers take DWR's power. It is only to the extent that the costs of the power exceed its market value that a burden is imposed on customers. Therefore, in allocating the DWR Annual PCRR, the Settlement Agreement focuses on achieving a result that fairly allocates the above-market component of the DWR contract costs to the customers of the three utilities. Under the Settlement Agreement, a CFC allocation is adjusted, through the use of Fixed Annual Adjustment Amounts, so that each utility's customers are expected to bear a market price for the power they receive from DWR, plus a fair share of the above-market component of DWR's costs. (Brief of Settling Parties, p. 15.)
While the underlying logic - attempting to fairly allocate the above-market costs of the DWR contracts - is sound, the percentages that the proposed settlement uses to allocate above-market costs do not yield a fair result. This is because they do not spread the costs equally among the ratepayers who will be paying these costs, so the impact on similarly situated ratepayers will vary, depending on that ratepayer's location. As described below, our adopted allocation methodology avoids this flawed result.
Another problem with the proposed settlement is its fundamental reliance upon forecasts of the relative net-short positions of the three utilities. The "Utility Allocation Percentages" that form the basic yardstick for the proposed settlement "[A]re designed to constitute a reasonable reflection of the relative net-short positions of the IOUs that DWR initially sought to serve when it entered into its contracts." (Motion of Settling Parties, p. 6.)8
The propriety of allocating future revenue requirements on the basis of forecasts of the utilities' net-short positions was actively litigated. (See, e.g., PG&E Opening Brief, pp. 14-19, re 2001-2002 net-short.) The basic idea behind the use of the net-short forecasts is that DWR was procuring power to fill the net-short positions of the utilities, creating a causal link between the net-short positions and the size and cost of the contracts themselves. The net-short forecasts are in essence treated as a proxy for the state of mind of DWR at the time it entered into the contracts that are at issue here.
The main problem with use of the individual net-short forecasts for allocating the costs of the contracts is the fact that DWR's purchases and contracts were made to cover the aggregate net short position of all three utilities, not the individual net short of each utility. (See, e.g., D.02-12-045, p. 12.) Contracts were signed to meet statewide needs, not the needs of individual utilities. (Id., p. 11.) We must find an allocation approach that reflects this fact. Second, a forecast of the net-short for only 2001 and 2002 does not actually reflect what DWR may have expected each utility's needs (and other sources of electricity, such as hydro) to be over the life of the contracts. (See, D.02-09-053, p. 30; SDG&E Reply Comments, pp. 5-6.) Third, the longer-term forecasts presented in this proceeding as reflecting the information available to DWR back when it was signing the contracts (the Nichols Declaration and Prosym Run 19g) are of uncertain value. (See, e.g. PG&E Opening Brief, pp. 16-19, SDG&E Reply Brief, pp. 11-12.) Also, as we noted in D.02-12-045, the amount of energy actually delivered to each utility's customers by the remaining DWR contracts does not necessarily match each utility's net short. (D.02-12-045, pp. 11-12.) While superficially appealing, the net short forecasts do not presently provide a principled basis for allocating the costs of the DWR contracts.
Accordingly, because of the flaws described above, we do not approve the proposed settlement agreement.
7 These annual adjustments are reduced to zero for the years 2012 and 2013.
8 The Motion of the Settling Parties states that it used "three principle net short allocation percentages presented in the proceeding:" SCE's proposed net short percentages based on D.02-02-052, PG&E's percentages based on DWR's Prosym Run 19g, and percentage shares derived from the Nichol Declaration. (Id., p. 14.) The first source addresses the 2001-2002 net short, while the other two contain forecasts for both 2001-2002 and for future years.
