3. The RAR Policy Framework
3.1. Introduction
The Commission's RAR policy framework was established in prior decisions, and the task for Phase 2 was to be, in large part, implementation of adopted policy determinations. It is for that reason that we ordered workshops on the myriad technical details that must be considered in developing a comprehensive RAR program. However, the Phase 2 workshop discussions revealed a need for us to clarify and amplify our underlying policies for RAR. We do so here before turning to the resolution of Phase 2 issues.3
3.2. The Adopted RAR Policy Framework
We begin by reiterating our adopted concept of resource adequacy as we expressed it in D.04-01-050:
"Resource procurement traditionally involves the Commission developing appropriate frameworks so that the entities it regulates will provide reliable service at least cost. This involves determining an appropriate demand forecast and then ensuring that the utility either controls, or can reasonably be expected to acquire, the resources necessary to meet that demand, even under stressed conditions such as hot weather [footnote omitted] or unexpected plant outages. `Resource adequacy' seeks to address these same issues. In developing our policies to guide resource procurement, the Commission is providing a framework to ensure resource adequacy by laying a foundation for the required infrastructure investment and assuring that capacity is available when and where it is needed." (D.04-01-050, pp. 10-11, emphasis added.)
The Commission envisions the resource adequacy program as the means by which the function of reliably matching resources to demand at least cost will be accomplished in the current industry environment. Historically, this function was the responsibility of integrated utilities that provided bundled service to retail customers, and the regulatory compact provided clear standards for utility accountability along with the opportunity for the utility's investors to earn a reasonable return on the investment they devoted to public service. Procurement and reliability responsibilities that were once the IOUs' are now diffused among various industry participants and oversight agencies, and both accountability mechanisms and the opportunities for investment returns are less well defined. Through RAR, the Commission is taking steps to (1) identify and assign these responsibilities in a manner that is effective in achieving reliability, cost-efficient, and fair for all stakeholders; and (2) foster an environment that is more conducive to investment.4
Several points from this foundational decision are worthy of emphasis here. First, the Commission seeks through RAR to ensure that the infrastructure investment required for reliability actually occurs. Second, the Commission seeks to ensure that the generation capacity made possible through that investment is available to the grid at the times and at the locations it is needed. Third, the Commission intends that capacity must be sufficient for stressed conditions, i.e., sufficient generation should be available under peak demand conditions even when there are unexpected outages. Finally, the Commission noted that the traditional utility role in procurement included the responsibility to provide reliable service at least cost, and that this is one of the "same issues" of traditional resource procurement that RAR seeks to address. Thus, the concept embodied in the phrase "reliability at any cost" is not a policy option. Ultimately, measures that are proposed to promote greater grid reliability should be evaluated by weighing their expected costs against the value of their expected contribution to reliability.
3.3. Revenue Adequacy
TURN is concerned that the topic of revenue adequacy for generation assets received little attention in either Phase 1 or Phase 2, and it urges that revenue adequacy should not be used as grounds for adopting a physical RAR. As TURN puts it, "the primary rationale for RAR up to now has been system reliability, not generator economics." (TURN Opening Comments, p. 6.) However, even though "revenue adequacy" may not have been explicitly discussed in workshops under that rubric, there is no impediment to considering it here. This is because reliability and generator economics cannot reasonably be de-linked. It is axiomatic that those who risk funds to develop the generation capacity California needs should have an opportunity to recover their investment costs and a reasonable return commensurate with the risk. A discussion of the means to achieve reliability necessarily encompasses a discussion of revenue adequacy. We believe that Constellation appropriately recognizes the linkage of RAR, investment, and reliability as follows:
"The fundamental premise for developing an RA requirement is to ensure that California maintains a reliable electric system by putting in place a resource adequacy construct that serves, first and foremost, to provide the appropriate incentives for new investment in infrastructure when and where it is needed." (Constellation Opening Comments, p. 2.)
We have already noted that under traditional regulation of integrated utilities, providing an opportunity for a reasonable return on investment was at the core of the regulatory compact. In significant part, generation is now provided outside of the traditional regulatory regime. Still, even in a market-based regime, revenues must be adequate so that investors who provide needed capacity can earn a return over time. An RAR program that does not address the need for a return on investment would fail in "laying a foundation for the required infrastructure investment."
Therefore, as we evaluate individual RAR program elements that have been proposed in Phase 2, we will, all other things being equal, give preference to those that promote appropriate investment needed for system reliability over those that do not do so. In particular, because capped energy pricing limits the revenues available for recovery of investment costs, which is particularly problematic for resources that are only needed for a few peak hours, we will look favorably to mechanisms that promote the recovery of investment costs through payments for capacity. It is for this reason that we view RAR as a physical, capacity-based program where a significant portion of the capacity is committed beforehand.5
3.4. Resource Planning vs. Operational
Requirements
In California's restructured electric industry, the CAISO is the designated agent for determining when and where generation capacity is needed in its control area on an operational basis. The Commission's policy that RAR should ensure that capacity is available when and where it is needed means that the RAR program design must be consistent with the CAISO's operational needs. Some parties have implied that because RAR is a resource planning exercise, it need not attempt to meet CAISO's system operational needs. Notwithstanding the distinction between planning and operational concerns, however, it is pointless to design a regulatory system that encourages investment in order to create capacity unless that capacity is actually available to the grid operator to serve load where it exists in day-ahead, hour-ahead, and real-time circumstances. Because our resource adequacy policy includes this availability dimension, we will not attempt to draw a bright line between planning and operational concerns. We will instead take a pragmatic approach to translating resource adequacy and availability into the operational needs of the CAISO. We note that it is not our intention to replace the operating reserve requirements of the CAISO with a more burdensome 15% reserve requirement extending into real time.
3.5. LSE-Based Procurement
D.04-01-050 adopted an LSE-based RAR program wherein each LSE is responsible for acquiring the resources needed for its own forecasted load and a reserve margin. This is consistent with the established regulatory principle of establishing prices on the basis of cost causation. Ultimately load will be served through the CAISO, and an LSE that does not provide resources in proportion to the load of its retail customers could effectively be subsidized by others. Through LSE-based RAR, we seek to eliminate "free ridership" and to minimize CAISO procurement where the costs of such procurement are socialized without reference to cost causation. Therefore, to the extent possible, we will favor RAR design elements that promote the LSEs' procurement responsibility over those that rely on CAISO procurement.
3.6. A New Paradigm for LSEs and
Their Suppliers
We make one additional point in reviewing D.04-01-050's provisions for a resource adequacy program. In recent years, California has made significant progress in building its generation infrastructure, but by most accounts that progress has not been sufficient to assure adequate generation availability in the coming years.6 Stated differently, it is apparent that, the status quo has not yielded a condition of resource adequacy in the CAISO control area, and cannot be relied upon to do so going forward. We are adopting RAR in order to spur infrastructure development and assure that capacity is available to the CAISO for dispatch. In so doing, we are rejecting business as usual and instead favoring more robust LSE procurement practices.
This almost certainly means that LSEs and their suppliers will need to change their procurement strategies. We will seek to avoid imposing unnecessary disruptions and costs on market participants, and we recognize that transitional mechanisms will be required to avoid unduly impairing existing business arrangements. On the other hand, as we move forward to give effect to D.04-01-050, we will not refrain from implementing those RAR program elements we determine to be necessary for reliability simply because those requirements may require changes in the operations of market participants.
3.7. Current Objectives for RAR
It has been suggested that this Phase 2 RAR decision should specify the "end state" for California's electric industry design. For example, Mirant/WCP support a resource adequacy construct that includes, among other things, central market-clearing mechanisms for uncommitted capacity and a capacity pricing mechanism that employs demand curve pricing. IEP similarly urges that we focus on an end-state that, in IEP's view, should include (among other things) an active market for trading capacity. While such ideas may well have merit, and we will explore many of then in the near future, we are not ready to adopt them here. As we determined in D.04-10-035, topics such as a multi-year RAR and the development of a capacity tagging and centralized trading regime are second generation issues that will be considered in other proceedings.7 The only end state that we specify at this time is, as indicated in the Phase 2 workshop report (at p. 19), a capacity-based resource adequacy program.
Some parties contend that there are significant practical impediments to implementing a comprehensive RAR program for 2006, and that certain proposed program elements are not ready for adoption. For example, CLECA/CMTA find gaps in the framework despite the extensive workshop discussions, and urge that we proceed by implementing RAR for 2006 on a trial basis with compliance penalties waived for that first year. SDG&E recommends adoption of a minimalist, non-precedential RAR program for 2006-07 while important implementation issues are further considered. TURN believes that several critical implementation elements are not "ready for prime time," and suggests a "keep it simple" approach for the first year of RAR implementation. Joint Parties urge adoption of only those measures that have a realistic chance of being implemented between the Fall of 2005 and June 2006. In its reply comments, PG&E argues for a streamlined RAR program for Summer 2006. Several parties urge that we postpone implementation of the local capacity requirement until that element of RAR is more fully developed.
In the remainder of this decision we will consider the myriad RAR implementation issues laid out in the workshop report. As we do so, we will give careful consideration to the state of readiness of the various program elements that have been proposed. Substantial and immediate progress toward the achievement of our goals for RAR through the adoption of a program at this time is vital to the development of the infrastructure needed for reliability. However, we will consider postponement of those elements that, despite their merit, require further consideration. The alternative of delaying the start of any RAR program until the details of all possible program elements are more fully vetted is simply unacceptable given the fragility of California's grid reliability. The other alternative--implementing program elements that have not been fully and fairly considered--is equally unacceptable given both due process requirements and the possibility of adopting unnecessarily costly RAR schemes.
3 Beginning with Section 4, this decision follows the general organizational approach of the Phase 2 workshop report. Thus, Section 4 corresponds to Chapter 2 of the report, Section 5 corresponds to Chapter 3, etc. Accordingly, we address many of the 87 topics identified for comment in the same sequence as in the report. We combine some topics due to their overlapping nature. For example, Topic 80 (split RA obligation) is essentially the same as Topic 2 and is not separately addressed. Topics 4, 13, and 14 all address the must-offer obligation and are therefore considered together. Other topics are similarly combined for discussion.
4 Throughout this section we refer to the need for generation investment and generation capacity. We believe this is appropriate for the general nature of this policy discussion. We recognize that the adopted RAR program includes dispatchable demand response as a countable resource. Also, some parties draw a distinction between investment in new generation units and investment in existing assets. For purposes of our discussion here, we make no such distinction.
5 See D.04-10-035 at p. 44. After stating that the purpose of RAR is "inducing forward commitments" the Commission stated:
Prospective limitations on liquidated damage contracts, eligibility thresholds that exclude energy limited resources that cannot be available for a minimum number of hours in a month, and other means by which capacity qualifies to cover loads and a 15-17% planning reserve margin are all part of creating a capacity-oriented resource adequacy requirement.
6 As noted in the Phase 2 workshop report, new power plant capacity totaling 6,700 MW has entered service in California since the end of 2001. Approximately 4,950 MW are under construction, with much of that total scheduled to enter service this year. (Workshop Report, p. 17.) However, the report went on to note that while an additional 8,500 MW of capacity has been permitted by the CEC, it is not under construction and has either been suspended or cancelled. (Id.)
7 Parties were notified that certain topics would not be taken up in Phase 2, and it would contravene due process to take up such topics here. We note that the February 28, 2005 ruling of Assigned Commissioner Peevey advised parties that while evaluation of a capacity market approach is not being carried out in Phase 2, it was his "expectation that the issues being addressed in Phase 2 will be resolved in a way that would not foreclose our movement toward a capacity market in the near future." We affirm the Assigned Commissioner's approach as the Commission's. We also note that the Energy Division published a staff white paper regarding capacity markets on August 25, 2005.
