6. Discussion

Section 1708 of the Pub. Util. Code permits the Commission to "rescind, alter, or amend any order or decision made by it," after notice to all the parties and with an opportunity to be heard. The statutory language of Section 1708 provides no limitations on the Commission's authority to reopen and reverse its decisions. As stated by the California Supreme Court: "That section...permits the commission at any time to reopen proceedings even after a decision has become final."¹³

However, we have long recognized that this broad authority should be exercised with great care and justified only by extraordinary circumstances:

"By its very nature, Section 1708 provides the possibility of an extraordinary remedy. Res judicata principles are among the most fundamental in our legal system, protecting parties from endless relitigation of the same issues. Section 1708 represents a departure from the standard that settled expectations should be allowed to stand undisturbed. Our past decisions recognize that the authority to reopen proceedings under Section 1708 must be exercised with great care and justified by extraordinary circumstances..."¹⁴

"...[O]nly a persuasive indication of new facts or a major change in material circumstances, which would create a strong expectation that we would make a different decision based on these facts or circumstances, would cause us to reopen the proceedings."¹⁵

We have also articulated specific parameters for this authority, stating in several decisions that we "may only modify or rescind a decision if (1) new facts are brought to the attention of the Commission, (2) conditions have undergone a material change, or (3) the Commission proceeded on a basic misconception of law or fact."¹⁶ It is within this legal context that we consider whether we should reopen R.91-08-003/I.91-08-002 for the purpose of rescinding or modifying the shared-savings incentive mechanism adopted in D.94-10-059.

We first turn to WEM's position. Although WEM does not refer to the legal standards discussed above, WEM argues that the Commission reached its determinations in D.94-10-059 under misconceptions of fact because it rejected TURN's position in that proceeding on several issues. We disagree. The Commission considered TURN's testimony during both the threshold and implementation phases of the proceeding, and concluded that it was not persuasive.¹⁷ The fact that the Commission did not adopt a position that WEM apparently prefers is not a legitimate basis for reopening the proceeding.

The March 13, 2002 ruling and TURN's comments point to the lack of penalties under the adopted shared-savings mechanism as an indication that the Commission proceeded under a factual misconception in adopting that mechanism. They also suggest that the sheer magnitude of earnings under this mechanism warrants Commission reconsideration of the underlying incentive mechanism. However, our review of D.94-10-059 indicates that the utilities' claims for profits are not out of line with the monetary value that we expected in adopting the shared-savings mechanism, assuming that the utilities deployed cost-effective programs based on verified savings. In other words, they are not unexpected.

Table 1 in Attachment 2 presents the range of potential profits at various levels of energy savings that we estimated in D.94-10-059. In the following discussion, we use the term "net benefits" to reflect the performance earnings basis for the shared-savings mechanism. The performance earnings basis, or "PEB" represents a calculation of energy savings minus costs (in dollar terms) associated with the utility's energy efficiency programs. The PEB is multiplied by the 30% shared-savings rate to produce the level of utility profits awarded under the mechanism.¹⁸

Our best estimate at the time we issued D.94-10-059 was that the utilities could earn profits of $89 million (collectively) for a typical program year if they met their savings targets. These profits would be paid over the ten-year measurement period in four equal installments. We estimated that those savings targets would yield $295 million in net benefits on a statewide basis. At 200% of target savings, we estimated that profits could be as high as $177 million under the incentive mechanism. If energy efficiency activities were not cost-effective, the penalties could be as large as $215 million, or the total ratepayer cost of the programs.

Attachment 4 presents the utilities' claims for profits for pre-1998 shared-savings programs, by program year, along with forecasted and actual performance. Although the shared-savings mechanism was discontinued for all new program initiatives effective January 1, 1998, it still applied to pre-1998 program commitments that could not be fully implemented until later program years, such as new construction activities and the competitive bidding pilots. Accordingly, there are some program activities implemented during PY1998-PY2000 that are subject to the shared-savings mechanism. (See Attachment 4.)

As indicated in that Attachment, the net benefits achieved for pre-1998 programs have exceeded the forecasted performance at that time. Overall, the forecast of net benefits for the period totaled $972 million. However, the utilities have actually achieved $1.7 billion in net benefits based on ex post measurement verification to date, or $425 million per year on average over the 1994-1997 period. This has exceeded the Commission's expectations of approximately $295 million in net benefits on an annual basis, per the table presented in D.94-10-059.

Attachment 4 also shows that the utilities' claims for profits have been in line with profit levels discussed in D.94-10-059. Collectively, the utilities request approximately $315 million in profits for programs associated with the shared savings incentive mechanism adopted in D.94-10-059. This translates to an average of $105 million in claims for profits per program year when the incentive mechanism was in effect. This level of profits is within the range of profit levels presented in D.94-10-059 for performance at or above target.¹⁹

The fact that the utilities' claims have never reached the bottom of that range, in the form of "negative incentives" or penalties, does not in and of itself justify reopening the proceeding, as TURN suggests. All this indicates is that the utilities implemented energy efficiency programs during 1995-1997 that are cost-effective to ratepayers. As ORA observes:

"ORA is surprised to see the absence of penalties being imposed in the past being put forward as evidence of failure of the incentive mechanism. Yes, the utilities were able to meet or exceed the minimum performance levels that form the threshold for penalties. That was the Commission's desire. We don't think the Commission was looking to penalize the utilities. Not only are the utilities glad that they did not incur penalties, ratepayers should be too since this is evidence that the utilities were able to generate energy savings."²⁰

While none of the utilities incurred penalties under the mechanism's cost-effectiveness guarantee, we note that other penalty mechanisms and risks under the shared-savings mechanism have come into play. For example, there have been instances where a utility has not met the MPS for a program and had to forego any earnings associated with that program.²¹

The March 13, 2002 ruling also suggests that the proceeding should be reopened based on a "rate-of-return" calculation where the lifecycle earnings are divided by program costs. Attachment 4 (Table 2) includes rate of return calculations similar to the ones presented in the March 13, 2002 ruling, but uses updated data that is specific to the shared-savings program.

The comments raise several valid concerns about the March 13 ruling. First, the incentive mechanism adopted in D.94-10-059 was designed to encourage energy savings, thereby allowing the State to reduce investments in generation, transmission, and distribution facilities. Accordingly, the profits should be compared to how much ratepayers would have had to pay if those savings had not been realized.

Second, even if one does compare the level of incentives relative to ratepayer-funded program budgets, one should take into account the fact that the lifetime of the program (both the benefits of the energy savings and the incentive payment period) is considerably longer than one year. Energy efficiency program delivers energy savings for several years, and the earnings are spread out over a ten-year period. The returns calculated in the ALJ ruling (and updated in Attachment 4) do not represent annual returns on the utilities' programs, as indicated in the March 13, 2002 ruling. Rather, they represent a ten-year return.

Third, as NRDC points out, a retrospective evaluation of profits from a rate-of-return perspective is counterproductive because it suggests that the programs cost too little: "For example, if the utilities had spent twice as much to achieve the same result, the earnings `rate' would probably have been so low as to escape notice. In effect, the Commission would be sending the message that it would prefer utilities to spend more ratepayer money than less."²² In D.94-10-059, we determined that the effective earnings rate associated with supply-side resources deferred or avoided by energy efficiency investments range from 26% to 52%.²³ As indicated in Attachment 4, even under a rate-of-return approach, the 44% rate achieved by the utilities for pre-1998 program activities is within the range of effective earnings rates allowed for supply-side investments during that same period.

In today's decision, we ask whether the incentive mechanism resulted in the deployment of cost-effective energy efficiency programs and rewarded the utilities for that deployment as envisioned in D.94-10-059. If, for example, the utilities earned profits on programs that were not cost-effective due to some unintended flaw in the incentive mechanism that had not been anticipated when it was authorized, then we might have a basis for reopening the proceeding. However, none of the data submitted on the performance of the programs indicate that such a flaw exists. These energy efficiency programs have been cost-effective based on our ex post verification to date.²⁴ Accordingly, the incentive mechanism has produced profits in the form of shared-savings to the utilities. We also note that the effective shared-savings rate has not exceeded the 30% authorized level. (See Attachment 4.)

The March 13, 2002 ruling also characterizes the shared-savings incentive mechanism as awarding profits for "events unrelated to any utility actions, such as technical degradation levels of customers' equipment," and suggests that this is one reason to reconsider D.94-10-059. This characterization implies that the incentive mechanism should be neutral with respect to factors that the utilities cannot directly control, such as actual equipment failure or early equipment replacements. However, this view is at odds with the mechanism we adopted-not through misconception of fact or law-but through deliberate consideration of how shareholder incentives should be structured for both the upside and downside earnings potential:

"As described in previous sections, the next generation of DSM incentive mechanisms will have a risk/reward profile different from any of the individual supply-side options discussed above, as well as from the DSM incentive mechanisms we have authorized in the past. Although ratepayers continue to put up the investment capital for DSM programs, shareholders will now be at risk for 100% of any losses to that capital. Unlike a rate-based plant, shareholder earnings will vary in direct proportion to performance, i.e., realized net benefits, even when factors entirely beyond the utility's management control affect that performance. And unlike any of the DSM shared-savings incentives in the past, DSM performance will be measured over a 7 to 10-year period for the purpose of calculating both earnings and penalties, and earnings for each program year will be distributed in four equal installments over that timeframe."²⁵

In sum, based on our review of D.94-10-059, the record in R.91-08-003/I.91-08-002 and the actual performance of the incentive mechanism adopted therein, we find no basis for reopening the proceeding due to a basic misconception of fact or law.

We now turn to the issue of whether D.94-10-059 should be reopened based on new facts or a major change in material circumstances. Clearly, the electric industry has undergone major changes since 1994. However, as discussed above, the relevant question is whether we would have made a different decision in R.91-08-003/I.91-08-002 given those changes.

TURN answers this question in the affirmative. According to TURN, the circumstances surrounding electric restructuring "moved faster than anticipated" and "[b]y 1996 at the latest the utilities knew they would be exiting the field of new plant construction. Nevertheless, the sharing mechanism adopted in 1994 was left unchanged."²⁶ As a result, TURN argues: "by December 1995, at the latest, and probably as early as April 1994, it was clear shareholder risk for DSM versus supply-side investments was substantially different than portrayed during testimonies and pleadings submitted in 1992 and 1993."²⁷

TURN's statements imply that we adopted the shared-savings mechanism without anticipating potential changes in the electric industry, and then failed to make timely modifications to that mechanism when such changes became apparent. To the contrary, we proceeded to adopt the shared-savings mechanism in full recognition that such changes could be imminent. In fact, we knew the nature of those changes, since both Commission proposals being debated at the time shared an identical vision for energy efficiency in a restructured electric industry.²⁸

Accordingly, in D.94-10-059 we specifically acknowledged that the adopted mechanism would need to be reevaluated sometime in the near future, and directed that such a review occur no later than the 1997 AEAP. We also provided interested parties the opportunity to petition for an earlier review if a final decision on electric restructuring fundamentally altered the role of utilities in energy efficiency or the regulatory disincentives to energy efficiency. We issued our electric restructuring policy decision on December 20, 1995, and the Governor signed Assembly Bill (AB) 1890 into law on September 23, 1996. No parties filed petitions to reconsider D.94-10-059 in response to these actions. Nonetheless, shortly after the issuance of AB 1890, we solicited comment on reevaluating the shared-savings mechanism in light of changes to energy efficiency brought about with restructuring.

Attachment 2 describes in detail the steps we took to develop the shared-savings mechanism and, after the issuance of D.94-10-059, to consider changes to that mechanism in response to changes in the electric industry. As discussed in Attachment 2, we concluded that it would not be productive to reassess the issue of shareholder incentives under a restructured electric industry until the fundamental issues of energy efficiency administrative oversight and governing policies were resolved. Our goal was to have these issues resolved by January 1, 1998. By that date, we expected to have fully transitioned away from utility-administered programs that focused on resource savings to those that focused on market transformation under an independent administrator. In the meantime, we directed that: "During this transition, utilities should retain their stewardship of demand-side management programs funded in prior years and continue to implement the adopted measurement and evaluation protocols. During this transition, the existing shareholder incentive mechanisms should continue to apply to utility DSM programs."²⁹

We also addressed the changes in risks and rewards under a restructured electric industry. Notably, we concluded that the new AB 1890 regulatory structure created greater disincentives than in the past for utility development of energy efficiency.³⁰ Our decision to retain the existing shared-savings incentive mechanism for utility programs during the transition was consistent with this conclusion. The decision language on this issue is presented in Attachment 2, Appendix 2.

In sum, we adopted a shared-savings mechanism in D.94-10-059 fully expecting that it would need to be reevaluated in light of industry restructuring, and established a procedural vehicle for interested parties to petition for such a change. Once electric restructuring was underway with the passage of AB 1890, we developed a transition plan for the associated shift in energy efficiency program focus and administration. In developing that plan, we considered the changed risks and rewards for utilities under electric restructuring and determined that the shared-savings mechanism adopted in D.94-10-059 should be continued until January 1, 1998. On that date, we discontinued the shared-savings mechanism and replaced it with a mechanism based on market transformation milestones. Given this chronology, we find no merit to the argument that we would have made a different decision in R.91-08-003/I.91-08-002 due to changed circumstances, either by adopting a materially different shared-savings mechanism for the 1994-1997 period, rescinding or modifying it earlier, or authorizing no incentive mechanism at all.

TURN also suggests that a valid criterion for modifying the shareholder incentive mechanism would be if substantially changed circumstances "would result in gross injustice or unfairness to ratepayers." That is not the standard we have adopted and this is not the proceeding in which to change that standard.³¹ Pointing to the drop in program spending in 1995 relative to previous years, TURN argues that the shared-savings mechanism did not provide incentives to the utilities to aggressively pursue cost-effective energy efficiency, despite the continuation of substantial shareholder incentives.³² However, as even TURN acknowledges, the reasons for the reduction in program spending are certainly debatable.³³ TURN fails to point out one very plausible factor to explain this reduction, namely, that we authorized reductions in DSM expenditures in order to continue an electric rate freeze that eventually became the basis for the electric rate freeze codified in AB 1890.³⁴

Moreover, no parties assert in this proceeding that net benefits to ratepayers from the utilities' 1995-1997 program activities, even with the payment of utility profits, have not materialized. In general, the estimates of net benefits have already been verified in prior AEAPs with respect to program participation, program costs and first-year load impacts.³⁵ We will be verifying the persistence of program savings over time in pending (and future) AEAPs, as Energy Division completes its independent verification of the utilities' retention and persistence studies.³⁶ Shareholders will earn profits under the shared-savings mechanism only if (1) these savings are found to be "real" through ex post verification, (2) program benefits are greater than costs, and (3) the level of program savings surpasses the required performance thresholds across portfolios. If those requirements are met, ratepayers will have benefited by the deployment of pre-1998 energy efficiency measures that continue to save energy over the life of the equipment.

Our best estimate to date is that the energy efficiency programs implemented (or initiated) during 1995-1997 have paid for themselves and, in fact, will yield net benefits after the payout of utility profits of approximately $795 million to ratepayers over the life of the measures. This represents benefits to ratepayers over and above program costs of $531 million and utility profits of $315 million. When PY1994 shared-savings programs are added to this calculation, net benefits to ratepayers (after program costs of $780 million and utility profits of $346 million) increase to approximately $1.4 billion.³⁷

At the same time, we recognize that electric restructuring did alter the nature of the avoided costs associated with energy efficiency investments. For the years 1994-1997 the utilities still owned a majority of the generation facilities used to produce the power they distributed. For the period 1998 to 2000, the utilities were procuring all of their marginal energy in the California Power Exchange and Independent System Operator (ISO) spot markets. For the period 2000 to 2002, they were using power purchased through the ISO day ahead market to cover any system imbalances. (See Attachment 3.)

However, these changes are not reflected in the forecasts of avoided costs used to value program savings under the 1995-1997 shared-savings incentive mechanism. The avoided cost forecasts adopted before 1998 were based on the incremental production costs, expected fuel purchases and energy contracts associated with the utility's long-term resource plan, i.e., a "pre-AB 1890" market structure.³⁸ Once adopted for a program year, the forecasts remained in place for the duration of the incentive recovery period (i.e., over all four earnings claims). Therefore, the net benefit (and shareholder incentive) calculations associated with the 1995-1997 program years do not reflect any of the changes in market conditions described above.

This raises the issue of whether ratepayers have been disadvantaged by such changes, in terms of the calculations of net benefits (and shareholder earnings) under the 1995-1997 shared-savings mechanism. At the direction of Judge Gottstein, the utilities presented a comparison between the energy avoided cost forecasts used to estimate the benefits of their pre-1998 energy efficiency programs, and the actual cost of energy in the years following program implementation. The information submitted is summarized in Attachment 3, and accompanying charts present the percentage differences between forecasted avoided costs and actual costs of energy over the 1994-2002 period. It is important to note that the actual costs of energy (associated with the last amount of power actually purchased or generated), will underestimate the actual avoided cost of the energy saved. This is because avoided costs are not directly measurable after the fact: One only observes data related to the costs that were not avoided.

With this caveat in mind, we observe the results of this comparison. (See Attachment 3.) As predicted by NRDC and others, the actual costs of procuring energy in a restructured industry for the years 2000 and 2001 were significantly higher than the utilities' forecasted avoided costs for those periods. For the years prior to 2000 and for the year 2002, the difference between the actual cost of energy and the forecasted avoided costs varies by utility, but in no case was the difference as significant as during 2000 and 2001. On the whole, it appears that the costs avoided by the pre-1998 energy efficiency programs under a restructured industry are higher than expected when these programs were initiated, to the benefit of ratepayers.

For the reasons discussed above, we conclude that R.91-08-003/I.91-08-002 should not be reopened to reconsider the shared-savings incentive mechanism adopted in D.94-10-059. However, nothing in today's decision is intended to preclude us from disapproving or modifying the requests for profits associated with pre-1998 programs that the utilities submit in the pending and future AEAPs. All claims for profits are subject to verification, consistent with our adopted measurement and evaluation protocols.

Moreover, in reaching today's determinations we recognize that the shared-savings incentive mechanism at issue in this proceeding was in place for a relatively short time, relative to the time period over which utilities have administered energy efficiency programs both before and after 1994-1997. As the energy industry has changed in California, we have revised our approach and thinking about energy efficiency program goals, implementation approaches and the utilities' role in administering energy efficiency activities. Since 2002, we have directed the implementation of successful energy efficiency efforts in California without the payment of profits to utilities. We continue to address policy and program implementation issues in our ongoing energy efficiency rulemaking, R.01-08-028. Accordingly, we emphasize that nothing in today's decision portends where we may be headed with energy efficiency in the future. Rather, we have addressed only the specific issue of whether actions taken by the Commission to encourage energy efficiency in the mid-1990s should be revisited at this time.