4. Contract Allocation Options and Positions of the Parties
The contract allocation options described below distribute each of the DWR contracts, or specific product components of each contract, among the resource portfolios of PG&E, SCE and SDG&E.
The contracts vary significantly not only in pricing terms, but also in terms of the amount of dispatch flexibility provided. Many contracts specify must-take quantities, and some contracts allow the seller to "put" additional volumes to the buyer. They also differ with respect to delivery points for the purchased power. Some of the contracts require deliveries exclusively to PG&E's service territory north of Path 15, the major transmission tie between northern and southern California. These are referred to contracts having a "NP-15" delivery point. Others specify deliveries to points south of Path 15 within the service territories of SCE and SDG&E ("SP-15" delivery point). Several contracts specify separate NP-15 and SP-15 must-take contract quantities. One of the contracts (Sunrise) specifies a delivery point in ZP-26, the third major ISO zone, which is located within PG&E's service territory between NP-15 and SP-15.
Many of the contracts provide for multiple products. Products include baseload "7x24" products, also referred to as "clock hour quantities,"20 that are designed to deliver energy continuously over a twenty-four hour period, seven days a week. Some contracts include "6 x 16"peaking (or "peak hour") products, which are designed to deliver energy six days per week during the hours from 6 a.m. to 10 p.m. In addition, individual products can be either "must-take" or "dispatchable." A dispatchable product allows the buyer to determine whether or not to dispatch the contract, depending upon need and market prices, while a must-take product does not.
Three of the contracts have delivery optionality, i.e., the seller can specify the delivery points of contract quantities. For the PacificCorp contract, which represents 300 megawatts (MW) of dispatchable capacity, the seller can specify either an NP-15 deliver point or a delivery point at the California-Oregon Border.
The Coral and Sempra must-take contracts are the two largest volume contracts that have delivery optionality, and both continue into the 2011-2012 timeframe. For 2004, the Coral contract specifies that 520 MW out of the total (950 MW) possible quantities delivered under the contract must be delivered into NP-15. This minimum requirement is broken down into "base quantities" (a combination of clock-hour and peak-hour products) and "additional quantities" that are peak-hour products. Under the contract, all the peak-hour quantities can be increased or decreased by 10% at the seller's option. The seller can specify a delivery point at any of the three ISO zones (NP-15, SP-15 or ZP-26) for those quantities where a NP-15 delivery point is not specified.21
The Sempra contract allows for the seller to specify any of the ISO delivery points for the entire must-take quantities, which total close to 1500 MW at present. Like the Coral contract, it contains a combination of products and contract complexities.22
Finally, three of the DWR contracts are between DWR and affiliates of the utilities. The Sempra contract is affiliated with SDG&E, the Sunrise contract is affiliated with SCE, and the PG&E Energy Trading contract is affiliated with PG&E.
In the following sections, we summarize the various options presented in this proceeding for allocating the DWR contracts, and the recommendations of the parties on which option should be adopted.23
4.1 DWR's Presentation of Allocation Options
At the request of Commission staff, DWR developed several contract allocation options to serve as the starting point for dialog with the utilities and interested parties during the discussions contemplated in the May 15 Ruling. The discussions started with DWR presenting a "straw man" allocation that followed the basic allocation guidelines in the April 2 Scoping Memo, and two additional alternatives. This led to further refinements of the options and an update of DWR's analysis of the options using its most recent modeling run, PROSYM Run35.24 DWR's July 19th filing presents the straw man and four contract allocation options, each representing a different permutation of (1) whether contract splitting is allowed and (2) whether the utility can be allocated a contract with its affiliate. By contract splitting, we refer to situations where a must-take contract includes more than one delivery point for specified contract quantities, and the allocation can be split along those lines for operational purposes.25 Dispatchable contracts were not considered candidates for splitting.
At the workshop, DWR explained that it developed the specific contract allocations under each option by attempting to balance the energy allocation with the net short energy need for each utility. As a measure of that need, DWR calculated a 7-year average of each utility's net short position. DWR used the PROSYMRun35 for the years 2003 to 2009 that included an estimate of direct access migration.26 The following average net short for each utility, as a percentage of total net short, result from DWR's calculations: 41% for PG&E, 40% for SCE and 19% for SDG&E.
Although DWR considered the energy allocation relative to the utility's net short position in developing each of its contract allocation options, DWR did not present any recommendations regarding which allocation should be adopted by the Commission. Instead, DWR provided for the Commission's consideration several comparison metrics in tabular and graph format for the options it analyzed, as well as for other options presented in this proceeding. These include: (1) allocated energy as a percentage of total contract energy, (2) allocated capacity as a percentage of total contract capacity, (3) residual net short as a percentage of utility load, (4) must-take surplus as a percentage of utility load, (5) contract costs, on a dollar and per MWh basis if costs follow the contract and (6) contract costs, on a dollar and per MWh basis if costs are allocated on a fixed $/MWh basis.
The utilities have used the DWR allocation options as a starting point and refined the "no affiliates" options to develop their preferred scenarios. They modify the specific contract allocations under these options based on delivery point and operational considerations, as well as additional cost metrics they take into consideration. Because DWR's options have met their intended purpose -- to initiate the dialog among parties and present comparison information for our consideration -- we need not consider the specific contract allocations presented in those options any further. Rather, we focus on the refinements discussed below and the metrics that DWR has developed to facilitate a comparison among proposals.
SCE and SDG&E have reached consensus on a proposal for allocating the DWR contracts. Their preferred allocation of contracts is presented in Attachment 2. As described below, SCE and SDG&E reach the same result for somewhat different reasons.
SCE takes the position that contract allocation should be based on sharing above-market costs in proportion to the net-short position of each utility at the time DWR entered into the contracts, i.e., during the 2001-2002 timeframe. In SCE's view, current data indicates that PG&E is responsible for 46% to 48%, SCE for 35% to 38% and SDG&E for 14% to 19% of the net short during that period.27
SCE developed its preferred allocation of contracts in four general steps. First, SCE looked at the forward market prices ("forward curve") for each product in each DWR contract, over the term of the contract. Second, SCE calculated the difference between the contract price and the forecasted market price for each product (and contract) to derive the above- market cost (or "burden") of all DWR contracts. Third, SCE applied the net-short percentages described above to determine the above-market burden that should be allocated to each utility. Finally, SCE "mixed and matched" the various contracts assigned to each utility, taking into consideration certain operational factors such as the delivery points specified in the contract, until it arrived at a solution that approached this allocation of above-market burden.28
As a starting point, SDG&E takes the position that the allocation of contracts should (1) reflect circumstances that existed at the time original DWR commitments were undertaken and (2) be consistent with the allocation of costs that are embedded in present rates to customers. To SDG&E, this means that an approximate 16% share of contract energy (as a percentage of the total DWR portfolio) should be allocated to SDG&E. SDG&E further argues that a historical framework for determining what percent of "the pie" that each utility gets is particularly appropriate for SDG&E, since it has no hydro and the contracts were purchased in a low hydro year.29
In addition to taking into account the overall energy allocation as well as specific operational characteristics of the contracts, SDG&E evaluated the allocation options in terms of the "all-in" costs of the contracts to its ratepayers. SDG&E's definition of all-in costs includes DWR contract power costs, residual net short procurement costs and the impact of excess must-take power sales.30 "Residual net short" refers to the power that the utility still needs to procure to meet loads after DWR contract quantities are allocated. It other words, it is the "net short," less those quantities.
To develop the all-in cost metric, SDG&E first calculated an average contract price based on the energy and capacity costs of the contract, similar to the contract cost metric prepared by DWR.31 SDG&E then adjusted this number to reflect the cost of buying the residual net short on its system and the impact of selling excess must-take power sales. The all-in cost rate in $/MWh is determined by dividing the contract costs and residual net short costs (net of surplus sales revenues) by the total net short energy requirement.32 SDG&E's preferred contract allocation, which is identical to SCE's proposal, minimizes the all-in costs to its ratepayers.
4.3 PG&E's Preferred Allocation
PG&E recommends that if the Commission pursues contract-by-contract allocation, the Commission should take a "rough justice" approach and consider a zonal allocation of power based on delivery point. Under this approach, PG&E would take scheduling and dispatch responsibility for those contracts which have NP-15 as their primary delivery point, and contracts with SP-15 delivery points would be allocated to the southern utilities.
The key consideration for PG&E, however, is that its customers should be required to take no more than their pro rata share of DWR power, based on future need. PG&E calculates each utility's pro rata share as an average of the net short projected under DWR's PROSYMRun35 for the years 2003 to 2009. In all but one respect, PG&E's calculation of the 7-year average net short is the same as DWR's calculation. While DWR takes into account expected direct access migration in the future, PG&E uses the PROSYMRun35 that ignores such migration in deriving its pro rata share. PG&E contends that this approach is the most consistent with the principle that DWR contracted power for the future needs of all utility customers, but at the time it did so, could not have anticipated the level of migration from utility loads due to direct access.
PG&E's calculations produce an average pro rata share of future net short of 40% for PG&E, 44% for SCE and 16% for SDG&E. PG&E's preferred contract allocation, which results in an allocation of 38% of contract energy to PG&E, is presented in Attachment 2.
As discussed at some length during the July 26 workshops, the contract allocations preferred by PG&E and jointly by SCE/SDG&E differ only with respect to two contracts: Sunrise and Coral. At the close of the workshops, the ALJ directed the utilities to present permutations to their proposed Sunrise and Coral allocations, as follows:
"(1) Allocate all of Coral to PG&E and all of Sunrise to SDG&E/SCE. In doing so, SDG&E and SCE may rebalance the mix of the SP-15 contracts they allocated between them in their July 19 proposals in making this modification. However, dispatchable contracts (such as Sunrise) should not be split. If the rebalancing between SDG&E and SCE would now involve allocation of an affiliates contract or portion thereof to either of them-they should discuss why this exception is warranted. Do not modify any NP-15 allocations to PG&E.
"(2) [L]ook at an alternative that instead of Sunrise being allocated to SCE/SDG&E, a comparable shift of contract quantities was accomplished by an alternate split of the Coral contract. This additional permutation is optional. No others should be submitted."33
SCE and SDG&E jointly submitted the permutation described in (1) above (referred to as the "ALJ alternate"), along with the comparison cost metrics they presented for their July 19, 2002 filings. They mutually agreed to a rebalancing of the allocation of contracts between them from their July 19 filings in presenting this permutation. PG&E also evaluated the ALJ alternate. All three utilities continue to argue in favor of their preferred allocations described above.
In particular, SDG&E argues that it should not be allocated Sunrise because (1) the contract specifies a ZP-26 delivery point that is not in SDG&E's service territory and (2) the transmission route to deliver power from ZP-26 to SP-15 has historically been congested. In addition, SDG&E objects to this permutation on the grounds that it produces a distinctly worse result for its customers in terms of cost burden and energy allocation than its preferred option. SCE objects for similar reasons.
PG&E argues against the ALJ alternate on different grounds: PG&E believes that the allocation of Sunrise to SDG&E/SCE and all of the Coral contract to PG&E results in a disproportional amount of energy being allocated to PG&E's customers in 2003 (i.e., 45%).34 Consistent with the ALJ's directions, PG&E presents an additional permutation whereby Sunrise energy remains with PG&E, and the Coral contract power is split by moving the clock-hour and peak-hour base quantities to SDG&E and SCE. Under this scenario, PG&E's customers would receive about 40 percent of the energy available under the DWR contracts between 2003 and 2009, which PG&E argues is more equitable based on its pro rata share of future net short.
SCE objects to this permutation on the grounds that it shifts proportionately too much above-market costs to SCE's ratepayers and would impose, combined with the allocation of the Sempra contract to SCE's customers, too much delivery risk. SDG&E objects for similar reasons.
4.5 ORA's Position
In concurrence with the utilities, ORA supports the principle that contracts should not be assigned to a utility that is an affiliate of the contracting party. ORA also supports PG&E's use of a multi-year forecast of net short, rather than a recent base period, as the benchmark for allocating contracts. In ORA's view, allocation based on system needs going forward will minimize costs.
In terms of the allocation of specific contracts, ORA supports the allocation of Sunrise to SP-15 to SDG&E under the ALJ alternate. ORA prefers this allocation to splitting the Coral contract because, in ORA's view, doing so would violate delivery point principles for contract quantities that are designed for NP-15 delivery.