4. Nature of the RA Obligation

4.1. Generator Obligations

As we noted earlier, the adopted RAR framework establishes an LSE-centered obligation under which the regulatory requirements apply to LSEs that fall under the Commission's jurisdiction. The obligation of suppliers to be available and perform is established indirectly through their contracts with LSEs. D.04-10-035 outlined (at p. 41) certain broad aspects of the contractual obligations to be imposed on generators. These include a sequential generator obligation to (1) be scheduled by the LSE, (2) bid into the forthcoming day-ahead market if not already scheduled, or (3) be subject to the CAISO's Residual Unit Commitment (RUC) process if the bid is not accepted. The Commission determined that in order to count for RAR purposes, contracts executed after the Phase 2 decision should include such provisions.

The record reflects three sets of concerns regarding these determinations outlined in D04-10-035: (1) a concern expressed by the CAISO that generators' obligation to be scheduled or bid should extend into real-time absent physical constraints on the unit, including special availability provisions for short start units; (2) an additional CAISO concern that we clarify our position on RUC availability payments to RA resources; and (3) LSE concerns that generators share in the RA obligation and any sanctions for failure to perform. These concerns are addressed in turn.

Regarding the first concern, the CAISO submitted comments on the Workshop Report requesting the Commission to "extend the requirement for RA resources to make themselves available into real-time where the resource is physically capable of performing on short notice" and to define specific availability obligations for short-start units. Short-start units are those with a start up and minimum run time of less than five hours.

As set forth throughout our decisions on Resource Adequacy, including this one, a key purpose of our RAR is to ensure that resources are made available to the CAISO when and where they are needed. Thus, we clarify here that an RA resource's obligations do not end with the RUC process. RA resources must be made available to the CAISO on a real-time basis to the extent they are physically able to perform. As a practical matter, this means that units that are already running and that have unscheduled RA capacity shall make that unscheduled RA capacity available to the CAISO, if requested. Additionally, short start RA units must self-schedule or offer into the CAISO's hour-ahead market and real time market for each hour of the operating day, subject to use limitation and contingency designations, even if not scheduled in the day-ahead market or committed by RUC, unless otherwise released from this obligation by the CAISO. LSE contracts with RA resources should reflect these obligations.

Regarding the second concern, the CAISO notes that the Commission, in Resolution E-3955, committed to clarify its position on RUC availability payments in this decision. We hereby reiterate that an RA resource must submit a zero dollar ($0) bid for RA capacity bid into RUC and that an RA resource will not be eligible for any RUC availability payment or revenue. As we said in Resolution E-3955: "It is not the intention of this Commission to simply provide needless revenue streams, or the ability to double-recover costs, to generators. It is the Commission's position that an RA resource that receives an RA payment should not also receive a RUC availability payment through the CAISO." Accordingly, LSE contracts with RA resources should reflect these policy determinations.

Regarding the third concern, the Phase 2 workshop discussions revealed a concern that the Commission's LSE-centered approach to RAR obligations could jeopardize RAR security objectives and burden LSEs alone with the RA obligation. LSEs noted in particular that they will not be in a position to know whether resources actually perform according to the contract terms, and they objected to the possibility that they would be subject to sanctions if the supplier failed to perform.

In response to these concerns, Commission staff and the CAISO developed a working proposal that would refine how the RA obligation is split between generators and LSEs. The LSE obligation would essentially be that outlined in D.04-01-050 and D.04-10-035, although the LSEs' RAR showings would be made to the CAISO as well as the Commission, and the proposal suggests the use of contract reference numbers. The key component of the Staff/CAISO proposal is the promulgation of CAISO tariffs with availability and performance requirements applicable to generators. For its capacity to be qualified for RA purposes, a generator would be obligated to (1) be available for testing by the CAISO to determine qualifying capacity, (2) have its qualifying capacity linked to performance (i.e., forced outages impact the next year's or period's qualification), (3) be included on the CAISO's "listing" of qualified capacity so the CAISO can make an accounting against the LSEs' submittals, (4) bid into the CAISO's forthcoming day-ahead market, and (5) be subject to pay the CAISO sanctions for non-performance, such as Uninstructed Deviation Penalties.

Staff reports that workshop participants agreed that this approach would resolve many of the implementation and fairness issues and provide better incentives than the LSE-centered approach. Notwithstanding widespread support for symmetrical LSE/generator obligations, however, comments on the workshop report show that most if not all parties believe that further consideration is required before the proposal can be implemented.

While we agree that further consideration is required before the Staff/CAISO proposal can be implemented, that further consideration must occur before the CAISO. The elements of the proposal that are not resolved, namely, how qualifying capacity will be measured, specific generator performance obligations, and appropriate penalties for generator non-compliance with those obligations, are within the province of the CAISO. It is our understanding that these issues will be addressed in the CAISO tariff and/or protocols implementing its new market redesign. Accordingly, we hereby adopt those portions of the Staff/CAISO proposal that we have control over - the LSE RAR obligations and penalties for non-compliance. Further, we approve, in principle, the concept of a dual RAR obligation including CAISO enforcement of specific generator availability and performance duties, and Commission enforcement of LSE RAR obligations. We encourage the CAISO to move quickly to adopt performance criteria and sanctions to meet the proposal's goal of a balance of obligations between LSEs and generators, and to arrange a sharing of information with LSEs so that they can seek appropriate enforcement of their contracts with generators.

In the interim, before such tariff provisions or protocols are in place, while we recognize the LSEs' concern that they might be held accountable by this Commission for generator performance over which they have little or no control (or even knowledge), we believe this concern can be reasonably mitigated. In particular, we would expect contracting parties to formulate terms and conditions that appropriately allocate any risks of generator nonperformance that accrue to the LSE. Additionally, we would expect RA contracts to require generators to comply with all CAISO tariff provisions, including those to be developed addressing RA resource performance obligations and penalties.

We support a coordinated CPUC/CAISO RAR program that includes CAISO-enforced generator performance obligations. Such a program holds considerable promise for a more effective approach to achieving RAR goals. The approach is consistent with the fact that the CAISO is the only entity with the ability to know whether a generator has met its obligation and showed up in the market. A CAISO registry listing the qualifying capacity of units could provide an incentive for plant investments/upgrades and a reduction in forced outage rates, and CAISO-based enforcement will be more effective than an LSE-only obligation. At the same time, LSEs will have better information regarding the value of the capacity when making purchases. Finally, staff notes that the approach is more consistent with the pricing and transmission service rules established in FERC's Standardized Interconnection Rules, and that it builds on the experience in the Eastern markets.

We therefore direct our staff to work closely with the CAISO towards meeting the goals of the proposal by developing and implementing the necessary CAISO tariff provisions and/or protocols.

4.2. Treatment of De-Rated Resources

Some workshop participants argued that the CAISO-listed qualifying capacity should count for RAR purposes for the life of a multi-year agreement, even if that capacity is de-rated based on the resource's performance. According to these parties, doing otherwise would undermine longer-term contracting. Other parties believe that fixing the level of qualifying capacity without regard to actual performance would create disincentives for the contracted resources to perform and for LSEs to make good choices from among available resources. While this issue arose in connection with Staff/CAISO proposal for rebalancing the RA obligation between LSEs and generators (see Section 4.1 above), and implementation of a de-rating requirement would be a part of such rebalancing, we state our policy preference here because of the importance of this issue to the RAR program.

Because we are implementing a physical capacity-based RAR program, it is our policy that a resource should only count to the extent that the capacity of that resource can be relied upon to perform. For example, if it is known that a resource with a multi-year LSE contract that once qualified at 100 MW can only be relied upon to provide 80 MW in the future, it would run counter to our RAR objectives to allow the LSE to continually count the full 100 MW for its RAR showings for the duration of the contract. If the LSE were not required to replace the missing 20 MW, it would in effect be able to socialize the costs of that missing capacity. As noted earlier, we seek to adopt RAR program elements that minimize or eliminate such free ridership.

The principle argument against de-rating the capacity that counts in RAR showings is that it might undermine long-term contracting. We support the use of longer-term forward contracts for capacity to foster a stable planning and investment environment. However, we see no justification for promoting long-term contracting through an artifice that (1) ignores the actual capacity availability of a particular resource, and (2) does so at the expense of reduced reliability, cost shifting, or both.

Moreover, we are not persuaded that a de-rating policy will unduly discourage long-term contracting. Several parties have observed that LSEs can protect their interests by negotiating appropriate contract terms. For example, Constellation notes that contracts could provide that if the capacity rating of the resource has diminished, it will be the obligation of the resource's owner to

replace the amount of the capacity reduction. We also note that such provisions would provide incentives for suppliers of capacity to operate and maintain their assets efficiently.

Thus, if the CAISO determines that the effective capacity of a resource is to be de-rated based upon the resource's actual performance, then only the adjusted amount should be counted as qualifying capacity in subsequent RAR showings. Similarly, if a resource makes investments to increase capacity, its capacity contribution should be recalculated to incorporate the upgrades.

The workshop report described general agreement of the participants that a policy of aligning a resource's qualifying capacity with the CAISO's capacity rating for that resource implies that the required 15-17% reserve margin should be evaluated and possibly adjusted. This is because if average forced outage rates decline as a result of tying RAR eligibility to performance, then presumably the overall reserve requirement could be safely reduced. Conversely, if average forced outage rates are high, then a higher reserve requirement may be justified. After we have gained experience with the operation of the RAR program, it will be appropriate to revisit the 15%-17% reserve margin and consider possible adjustment.

4.3. The Must-Offer Obligation (MOO)

The MOO is a FERC-approved, CAISO-administered mechanism under which certain generation units not otherwise scheduled are obligated to operate and bid into the CAISO's real time market. The mechanism includes a process under which the CAISO grants or denies MOO waiver requests. The CAISO provides some compensation to units that are denied waivers and thus required to operate in real time. FERC has indicated its intent that the MOO will be terminated when our RAR program becomes operative. Generators, in particular, are eager to have the MOO eliminated as soon as possible, and they recommend that the mechanism be terminated when the RAR program is implemented. Other parties believe that the mechanism should be retained until both the RAR and the CAISO's Market Redesign and Technology Upgrade (MRTU) programs are operating. MRTU is slated to commence operation in February 2007.

In connection with the SVLG's proposal for standard contract language (see Section 4.4 below), the Phase 2 workshop report invited comments on whether the MOO and associated waiver process should be extended until the MRTU process is implemented (Topic 4). Also, in connection with interagency coordination issues (see Section 5 below), the workshop report invited comments on (1) the proposition that the Commission, the CAISO, and the FERC must coordinate to determine both replacement requirements and the schedule for eliminating the CAISO's MOO authority (Topic 13); and (2) the proposition that RAR will replace the FERC-imposed MOO (Topic 14). We take up these three related topics here.

The CAISO, LSEs, and their customers generally supported extending the MOO, including the waiver denial process, until the CAISO's MRTU program is implemented. As described in the workshop report, there is concern that if the MOO and the associated waiver process are eliminated earlier, the CAISO will not have a means to commit RA resources for the next day. However, such a means will be available with implementation of the day-ahead market as part of the MRTU. Extending the MOO until MRTU implementation would provide an interim mechanism to assure dispatch of needed resources. SCE believes that continuing the MOO will provide needed market power mitigation until MRTU is implemented.

Joint Parties believe that the MOO will remain necessary until the RAR program has been proven to meet California's energy needs, the CAISO has implemented the MRTU and its day-ahead market, and the CAISO has authority to enter into backstop local capacity contracts.8 They also propose that the CAISO track waiver denials for non-RA resources and report them to the Commission. They believe that this would indicate whether the CAISO is relying on excess reserve levels, needed local resources have been identified, and the RAR program design is missing any needed element.

Suppliers of generation and others opposing continuation of the MOO beyond June 2006 contend that it would undermine the incentive to contract for long term capacity, and discourage investment by continuing short-term procurement and inadequate compensation. They also believe that the MOO will not be necessary when the RAR program is operative, and in particular they dispute the workshop report's conclusion that the MOO mechanism is necessary as an interim measure for the CAISO to commit resources a day ahead. For example, IEP contends that there is no need to keep the MOO since RAR contracts will provide the CAISO with the commitments and resource availability is needs.

It appears that the MOO and associated waiver mechanism may discourage contracting, provide inadequate compensation, and fail to foster a stable investment environment. For these reasons, the mechanism is not aligned with our RAR goals and should be terminated.9 Nevertheless, we conclude that it should be retained at least until the MRTU mechanism and the day-ahead market are operative. As discussed later in this decision, we are permitting the use of certain non-unit specific contracts for RAR showings on a transitional basis. These contracts may not provide the CAISO with the level of commitment that unit-specific contracts should provide.

In light of this, the lack of a mechanism for scheduling units in the day-ahead time frame prior to MRTU implementation, and the recognition that any major new program such as RAR may have unanticipated initial implementation issues, it is prudent to proceed with caution. For the early stages of the RAR program, we do not have the same level of confidence as the generator parties that RAR contracts will provide the CAISO with the commitments and resource availability that it needs.

While we recognize that the MOO may act as a disincentive for LSEs to enter into forward contracts, one of our purposes in adopting RAR is to provide an incentive that should lead to that very result. Eventually, adding a multi-year forward commitment dimension to the RAR program may enhance this effect. In any event, we note that nothing prohibits multi-year forward capacity commitments from qualifying for year-ahead RAR showings, and we encourage such commitments.

At this time, we will not approve the other pre-conditions for termination of the MOO that were suggested by Joint Parties. In particular, we will not require that RAR be proven (at least in the context of a formal proceeding) to have met California's energy needs, as that strikes us as an unnecessarily high standard for elimination of a mechanism that appears to be at odds with our RAR goals.10 The proposal that the CAISO track waiver denials for non-RA resources and report them to the Commission appears reasonable as an early means of monitoring the effectiveness of the RAR program. We request that the CAISO periodically provide such reports to our Energy Division during the transitional period between the commencement of the RAR program and the termination of the MOO.

The workshop report states that continuation of the MOO mechanism on an interim basis may require that supplier cost information be provided to the CAISO so that it can efficiently select necessary resources. It also notes that existing must-offer compensation may duplicate payments under RA contractual arrangements, and suggests that appropriate adjustments to must-offer compensation for RA resources should be considered. After reviewing all of the comments and replies, we are persuaded that these measures are not necessary. In particular, the possibility of duplicate payments seems somewhat unlikely.

Based on the foregoing, we determine that an extension of the MOO and associated waiver process is necessary to facilitate commitment of RA resources until CAISO's MRTU process is implemented.

4.4. SVLG's Standard Contract Proposal

In connection with the resource availability dimension of the RAR program, D.04-10-035 observed (in Section 3.8.2, pp. 41-43) that standard contract terms and conditions would be important to the development of readily transferable capacity contracts. While it was not ready to endorse the creation of a mandatory, centralized capacity market, the Commission noted that a readily traded capacity contract that parties can voluntarily exchange would be useful. The Commission included this topic in the scope of Phase 2 workshops, and it raised the possibility of approving specific contract language.11

For the Phase 2 workshops, SVLG developed a set of proposed "essential elements" of a capacity contract along with illustrative contract language. According to the Phase 2 workshop report, the SVLG proposal is viewed as a replacement to the MOO and would only be effective until implementation of the CAISO's MRTU. The report notes that SVLG's proposal would (1) allow the buyer to count capacity towards RAR, (2) allow the seller to retain ownership and/or control of the capacity, (3) qualify the contract capacity, (4) ensure that capacity is not double-counted, (5) require the seller to make its resource available to the CAISO for all hours of the delivery period in the contract.

Most of the parties addressing this question oppose our adopting specific, mandatory contract language, and SVLG itself does not propose that we do so. We are persuaded that it is neither necessary nor desirable to require that specific language be adopted as a mandatory component of qualifying contracts. As AReM point out, contract language is sometimes modified on a company-by-company basis due to internal legal requirements or preferences. We agree that the focus should be on essential contract elements.

Development of a readily transferable capacity contract may warrant further study. At this time, we believe it is important to provide guidance to LSEs and suppliers while furthering the availability objectives of the RAR program. Accordingly, we determine that an RA capacity contract that includes the following minimum elements, drawn from the SVLG proposal as well as Resolution E-3995 dated September 22, 2005, shall qualify as an eligible contract that LSE's can rely upon to meet their RA procurement obligations:

The workshop report suggested that the Commission should consider how any changes to standard contracting elements should be incorporated into the Renewable Procurement Standard (RPS) contracting process. We concur with a number of commenters who indicated that this topic was not developed in workshops and that such consideration is not necessary at this time.

Before leaving this topic, we note that TURN has proposed that there should be a capped energy strike price for contracts where suppliers are recovering fixed costs through capacity payments. Without commenting on the merits of such a cap, we observe here that this would seem to be a major addition to the RAR program, yet one that has received scant attention to date. Such a proposal would require full vetting before it could be adopted as a mandatory requirement. However, nothing in this decision precludes parties from voluntarily pursing such contract terms.

8 CMTA states that it no longer joins the other parties with which it submitted joint comments (CLECA and Joint Parties) with respect to their positions on the MOO. CMTA now characterizes the MOO as a "vestige of the energy crisis which should be eliminated as soon as possible." (CMTA supplemental comments, p. 2.)

9 We note that on August 25, 2005, IEP filed a complaint with FERC, seeking to replace the MOO with an alternative tariffed payment structure. The Commission will be participating in that proceeding, Independent Energy Producers Assoc. v. California Independent System Operator Corp., FERC Docket No. 05-146 (IEP Complaint).

10 Of course, mid-course corrections to the RAR program may prove necessary. Also, as noted elsewhere in this decision, we are planning to conduct further proceedings to upgrade the RAR program and to consider developing a centralized capacity market.

11 D.04-10-035 proposed the following workshop discussion topic (among others):

"What specific standard language, if any, should be included in future contracts between LSEs and generators that will sufficiently obligate generators to bid into Day-Ahead markets and be subject to RUC and other appropriate processes?" (D.04-10-035, p. 42.)

Previous PageTop Of PageNext PageGo To First Page