The most controversial issue in this phase of the proceeding is the process for planning a CCA's cut-over and how to determine at what point the serving utility's power purchase liabilities will no longer be included in the CRS. The primary objectives of this process are to mitigate costs incurred by CCAs and the serving utilities and to provide a mechanism for coordinating a CCA's cutover. Our Phase 1 order, D.04-12-046, found that an open season would limit the CCA's liabilities for utility power purchases and provide information upon which the utility could rely about when to stop purchasing power for future CCA customers. The dispute is over the details of this commitment.
The utilities propose that in order to be relieved of prospective power purchase liabilities, the CCA must make a binding commitment to a five-year forecast of the CCA's load as of a date certain. This commitment would be made during an open season period between January 1 and February 15 of each year. The utilities would also require a Commission decision directing the utility to stop procuring load for the CCA's customer, arguing that this decision is needed to provide assurance to the utility that its procurement decisions would not be second-guessed at a later time. The CCA's five-year forecast could be modified during each year's open season without penalty. CCAs would be forgiven forecast errors within ten percent of the forecasted amounts to recognize the difficulty of estimating the load of customers deciding to remain with the utility. Forecasting errors outside of this deadband would be subject to penalties.
CCAs object to the utilities' proposal because it imposes forecasting liability on the CCA for utility load. The CCAs argue the resource adequacy forecasting process adopted in D.04-10-035 anticipates a forecasting and power purchase planning process that obviates the need for the duplicative and punitive process the utilities propose as part of the open season. CCAs believe the utility proposal would make CCAs liable for forecasting errors even in cases where the utility actions were the cause of the CCAs' nonperformance.
TURN and CCSF proposed an open season process that provides (1) a specific cut over date for commencement of CCA service; (2) a list of the customer classes the CCA will offer service to; and (3) a cooperative load forecasting process. CCAs would be subject to fees only in the event that they failed to accept the transfer of customers on the specified date. The fees would be equal to the incremental cost to the utility of continuing to serve the customers.
Local Power objects to the concept of an open season generally. It argues that an open season unreasonably requires the CCA to assume risk. It argues that AB 117 requires the CCA to initiate service within 30 days of the date it signs a contract with the utility and that the open season creates a circumstance which is almost impossible for the CCA to accommodate.
Discussion. As a preamble to our discussion, we refer to D.04-12-048, which we issued in the long-term procurement proceeding the same day we issued the Phase 1 decision in this proceeding, and which states:
A CCA may execute a binding notice of intent with a commitment to a target date, at which the CCA is responsible for its own energy procurement and resource adequacy. If the CCA does so, its customer will not be responsible for stranded costs of any utility commitments entered into after the agreed upon date. However, if the CCA does not meet the target date, it will be liable for any incremental costs that the utility incurs in excess of its average portfolio cost to serve the load that the CCA is not able to serve. (Finding of Fact 29, D.04-12-048.)
The objective of a binding notice of intent is to transfer liability for customer power purchases from the utility to the CCA according to a specified date and in so doing minimize the liabilities of all customers for stranded costs associated with power purchase commitments. While this latter objective sounds simple, its accomplishment may not be, given the many variables and contingencies inherent in the open season process.
The utilities propose a voluntary annual open season at which time CCAs would make that binding commitment. It would occur between January 1 and March 1 (or February 15, depending on the available date of the California Energy Commission's resource adequacy load forecast) and appears to require a standard format that would apply to all CCAs equally. We agree with the utilities that the open season provides a reasonable procedure for getting CCA binding commitments and we agree that they are necessary in order for the CCA to be assured of limiting its liability for utility power purchases.
We do not agree with Local Power and CCSF that the filing of an implementation plan or the creation of the CCA must automatically trigger changes in utility procurement practices. In some cases, the utility may be able to modify procurement strategies without imposing additional cost or risk on utility customers. As the utilities observe, however, if the CCA never initiates service, changes in procurement in other cases may ultimately be costly to utility customers. To the extent the CCA is willing to make a commitment, even tentatively, there may be ways to mitigate procurement costs. For example the CCA and the utility may enter into a preliminary agreement whereby the CCA assumes some liability for changes in power purchase strategies in exchange for relief from other risks. In all cases, the utility must reasonably manage procurement consistent with Section 366.2, which provides that CCAs must assume only the "net unavoidable costs" of utility power procurement. While we recognize the uncertainties the utilities face in trying to forecast load loss prior to receiving a CCA's binding commitment, we also believe the utility should take reasonable steps to plan for that contingency, for example, by reducing long term commitments until a CCA's plans are assured. In any case, the uncertainties of procurement planning are not novel and the utilities remain liable for unreasonable power purchases. We will undoubtedly address this issue in proceedings addressing utility procurement plans.
We do not believe that AB 117 limits the Commission's discretion to adopt a process for a binding commitment between the utility and the CCA, as Local Power suggests. The statute certainly anticipates the Commission's adoption of the administrative elements of the program that are responsible and practical. An open season and the requirement for a binding commitment are reasonable tools for implementing the statute and recognize several elements of the statute designed to facilitate CCA operations while protecting utility customers. The binding commitment is a reasonable way to balance the interests of the utilities and the CCAs with regard to the parties' mutual understanding and the limitations on their respective liabilities. If a utility and a CCA are able to develop an agreement that is tailored to the more specific circumstances at hand, so much the better. With all of this in mind, we concur with the utilities and TURN that a CCA that declines to participate in the open season is liable for any power commitments made on behalf of its customers up to the date the CCA begins operations. An exception to this rule would be where the utility and the CCA can craft a binding commitment outside the open season that is tailored to the CCA's circumstances.
Although we adopt the utility open season proposal in concept, some of its details require modification. Specifically, we reject that portion of the proposal would require CCAs to assume the risk of utility load forecasts for five years, which would be required under the utilities' proposal for the CCA to pay for any variations from forecasts for the departing load of CCA customers. This proposal effectively requires the CCA assume liability for utility forecasts, a risk that is properly the utility's. Currently, the utilities have information about the number and type of customers they serve and they forecast demand accordingly and, by all accounts, well. The utilities are adept at forecasting customer load because they have historic customer information and technical resources. The utilities routinely modify system load forecasts when they gain or lose customers, in light of changing usage patterns and other circumstances. The departure of CCA customers from a utility's system is no different from any other change in system load the utility may experience except that the lost load may be larger. When a group of customers transfers from the utility to the CCA, the utility still has the information and expertise it requires to forecast the change in its system load. The utility should assume responsibility for the final forecast of its total load, just as it assumes that responsibility today. Under this policy, the CCA retains responsibility to forecast its own load and assumes all risk and costs where the forecast and demand vary for its own customers.
Although we reject the utilities' proposed requirement that CCA's assume risk for 5 years of forecasting departed load, we recognize the need and opportunity to minimize the risk of forecasting that is imposed on utility customers. TURN's proposal for a collaborative forecasting process appears to be a reasonable compromise with some modifications. TURN would require the CCA and the utility to work together to determine the load that would be transferred to the CCA. We understand the utilities' concerns that the collaborative forecasting process may not provide strong incentives for the CCA to work with the utility on developing a reasonable forecast. For that reason, the open season rules should require the CCA to disclose which portion of each class of customers would be subject to a cut-over. As the utilities propose, the open season rules should also require the CCA to provide all relevant information about the number of customers to be cut-over, the rates, rate design and special contracts to facilitate forecasting. All of this information would be provided to the utility confidentially and subject to a nondisclosure agreement, at the option of the CCA. As the utilities propose, where a CCA initiates service before or after the date of its commitment to the utility, the CCA will be responsible for all incremental costs incurred by the utility.
To account for the possible failure of the collaborative forecasting process, we also adopt TURN's proposal to establish default opt-out figures for the first year of the CCA's operation, which will spread the forecast risk of this type of uncertainty. The utilities should propose such default opt-out percentages in their tariffs. The purpose of this collaborative forecasting exercise is to limit the risks of the utilities and the CCAs. If we find in the future that it fails to accomplish that goal, we will reconsider it.
Of course, utilities' bundled customers should not assume the risk for a CCA's failure to transfer customers on the date to which the CCA commits for any reason. Where the CCA is at fault, it should pay the utility's incremental cost of the delay that is related to power purchases. Where the utility is at fault, it should credit the CCA with the incremental cost of its power purchase losses. If the CCA believes the utility is at fault and the utility does not agree with its culpability, the CCA should file a complaint seeking credit for power purchase costs. If the CCA seeks damages, it must file suit in a court of law. In any of these cases, we encourage the CCA and the utility to seek assistance with alternative dispute resolution. As TURN observes, the CCA and the utility may and should mitigate risk associated with forecasts by agreeing in advance that the party who has too much power will sell the power to the party who is short on power. We will direct the utility to include such a provision in its tariffs.
We agree with TURN that a CCA submitting a formal notice of intent under the open season tariff has automatically provided a "self-executing" notice that relieves the utility of its power supply obligations. Requiring a Commission order, which the utilities suggest, would unnecessarily add cost and delay to this process.
The CCA's proposal for the utility and other parties to keep a CCA's open season information confidential is reasonable. If knowledge of a CCA's cut-over date and costs were to become public, the CCA could be vulnerable to price gouging by potential power suppliers.
In sum, we find that a voluntary open season in which the utility and the CCA work cooperatively to reduce risk and stranded costs is reasonable. We encourage the utilities and CCAs to tailor their agreements to minimize costs and promote cooperation wherever possible. In the open season tariffs we adopt today, the utilities are responsible for their load forecasts once the CCA has begun operations just as the CCA is responsible for its load forecasts, although we adopt measures designed to facilitate good utility forecasts. The CCA's open season agreement is a binding commitment under which it assumes all liability for delays in implementation except those attributable to utility actions. While voluntary, if the CCA chooses not to participate in the open season or to sign a more tailored binding agreement with the utility, the CCA must assume the risk for all utility power purchased up to the CCA's initiation of service. The CCA's binding commitment for an in-service date, in combination with the approaches we adopt today to mitigate risk, represent a reasonable middle ground that is sensible and fair to bundled customers, utilities, CCAs and their procurement customers. Attachment B, proposed by TURN and adopted herein with the minor exceptions we discuss, describes the open season and should be included in each utility's tariffs.