5. Discussion

5.1. Summary Findings Regarding the True-Up of Incentive Earnings

In finalizing the 2006-2008 period, we are guided by the following fundamental principles:

1. The program should promote the Commission's energy efficiency goals;

2. Incentive methodologies should be applied in a fair, transparent, and conceptually consistent manner; and

3. The utilities should receive incentive rewards or face penalties based on their effective administration of the energy efficiency portfolios given the information they had access to at the time the portfolios were being implemented.

Accordingly, we evaluate the parties' disputes in terms of these goals and principles. Our task is to true-up the interim calculations of incentive earnings for the 2006-2008 cycle, and thereby determine whether additional earnings are due, or whether penalties apply. As a basis for evaluating whether the interim RRIM earnings awards warrant further adjustment in the final true-up phase, we must identify: (a) the magnitude of energy savings accomplishments subject to incentive rewards; and (b) an appropriate percentage allocation of the identified net benefits between ratepayers and IOU shareholders.

Because parties could not reach consensus on a reasonable basis to simplify the calculation of energy savings achievements, we look to the record and use our independent judgment to assess an acceptable outcome

As discussed below, we rely upon the ex ante assumptions from the 2005 DEER, as the basis for the true-up of energy efficiency incentives for the 2006-2008 program cycle. The EM&V process was the vehicle established by the Commission in D.07-09-043 for measuring success (or failure) in achieving energy efficiency accomplishments and cost savings for various purposes. We have used some of the information developed pursuant to the EM&V process, as reflected in Scenario 3, to arrive at the incentive amounts authorized by this decision; however, for reasons discussed herein, we do not rely on the entirety of the results. Unlike expenditures for energy resources that are measured through arms-length transactions, energy savings cannot always be as easily quantified. To calculate cost savings and net benefits associated with energy efficiency measures, it is necessary to develop assumptions as to relevant parameters based on surveys, sampling, and extrapolation of estimates over extremely large volumes of data points. As we have noted here and in earlier decisions, the EM&V process has been extremely contentious, resulting in considerable disagreement over estimates of energy savings achievements, and the resulting incentive payments due.

5.2. Role of the Energy Division Evaluation
in the True-Up

The Final Evaluation Report of 2006-2008 energy efficiency savings performance was finalized by non-party staff and their consultants in our Energy Division in accordance with adopted Commission processes. The Report shows that California ratepayers' $2.1 billion investment in energy efficiency resulted in over 6,000 Gigawatt hours (GWh), 80 million therms, and over 1100 MW in annual energy savings over the 2006-2008 cycle.¹⁶ These accumulated savings represent approximately 3.2% of electricity and 1% of the natural gas sold in California in 2008. The reported savings were evaluated through field work to verify energy efficient technologies installed and the related savings attributable to the programs. In total, the evaluations for any given parameter directly assessed the majority of the ex ante claimed savings. Evaluations of measure installations accounted for 77% of kilowatt-hour (kWh) savings. Evaluations of unit energy savings accounted for 86% of kWh savings. Evaluations of load shapes covered 80% of kW savings and evaluations of NTG ratios covered 90% of kWh savings.

Energy Division focused evaluation resources on measuring gross savings from the end-use measures or technologies that dominated portfolio savings, i.e., high-impact measures (HIM), and on estimating net savings attributable to programs with the highest savings from installed technologies. The IOUs argue that HIM methodology developed point estimates for certain measures and then applied them to similar measures across the portfolio. They further contend that the shift in methodology to evaluation of HIMs represents an untested divergence from longstanding and commonly accepted EM&V protocols without the opportunity for public review. In addition, the IOUs argue that the evaluated results were not properly translated into earnings projections, and that the ERT itself was systematically flawed such that it produced earnings estimates with no statistical confidence.

The IOUs allege that the findings in the Energy Division Report are unreliable, lack transparency, and have not been subject to an adequate public review process. PG&E, for example, argues that given the breadth of the evaluation, the time provided for review and comment on EM&V evaluations was too short. PG&E contends that critical data needed to conduct a comprehensive review was not made available in a timely fashion, which foreclosed the possibility of robust analysis. Consequently, PG&E asserts that the process did not provide for the free exchange among stakeholders, as contemplated by the Commission in D.07-09-043.

The IOUs also allege the Energy Division results utilized values without references to sources, and that methodologies lacked actual documentation. The IOUs further allege various technical errors in the processes utilized by the Energy Division in evaluations of savings.

The claimed errors involve various technical details often involving minute and arcane details such as how the Energy Division consultants conducted surveys, extrapolated samples, and used data in calculating the various savings measures. We recognize that there is room for debate about judgments made in conducting surveys and extrapolating results to estimate ex post measures.

In D.05-01-055, we mandated that the Energy Division take responsibility for managing and contracting for all EM&V studies. ¹⁷ This mandate marked a shift in the responsibility from the utilities to Commission's non-party staff and helped ensure unbiased results by having a neutral entity overseeing the EM&V process. In addition, Energy Division has contracted with experienced expert evaluation contractors, whom they have used throughout the processes for developing the research and data to estimate interim and final earnings claims.

The Energy Division Report necessarily encompasses review of a large number of records that reflect considerable technical complexity and detail. The Commission established a process by which evaluation studies must be posted for public comment prior to finalizing the results. Energy Division followed the correct protocols for vetting that were adopted in D.07-09-043, which required:

...a specific and adequate process by which parties can submit questions, concerns and comments to both Energy Division and evaluation contractors. Conferences and the submission of written comments based on conferences, allow parties to participate in the process by raising and discussing issues. This takes place in formulating the several reports before they are finalized: the draft Verification Report, the draft final evaluation reports, and the draft Final Performance Basis Report. Our belief is that any concerns the parties may have can be resolved through such a process.

(D.07-09-043 at p. 129.)

Energy Division circulated requests for technical participation from parties, provided draft materials, held several meetings to discuss technical issues, provided opportunities for comments, and responded in writing, explaining how assumptions were applied in developing and measuring performance results.¹⁸ Energy Division changed or updated numbers where comments were found to have merit.

The Energy Division contractors provided updates to installation rates (how many technologies were installed and operating), unit energy savings (savings for any given technology), and NTG ratios (a factor used to adjust savings to account for the influence of the program) where evaluation updates were available. Several parameters, primarily cost data, were part of the data set but were not updated with evaluation results.¹⁹

The Energy Division adhered to strict timelines and rigorous public review process. Stakeholders were provided opportunities to comment on the evaluation plans. Consultant reports were published at different times in 2007 and 2008, and the Energy Division's final report was released for public comment in December 2009. Results from the impact evaluations were posted for public review and comment in December 2009 in detailed technical reports, and were also presented in public webinars. The Energy Division Report included voluminous and detailed point-by-point responses to stakeholders' questions and claimed errors. The public comment period generated approximately 1,700 comments, all of which were addressed by the Energy Division and its evaluation contractors. The reports were finalized in February 2010. Summaries of these report findings are included in the Energy Division report, and the final reports were posted on the California Measurement and Advisory Council (CALMAC) website.

The IOUs claim the Energy Division results are non-transparent and utilize values without references to sources, and that methodologies lack actual documentation. The IOUs claim various technical errors in the processes utilized by the Energy Division in evaluations of savings.

The claimed errors involve various technical details often involving minute and arcane details as to how the Energy Division consultants conducted surveys, extrapolated samples, and used data in calculating the various savings measures. We recognize that there is room for debate about judgments made in conducting surveys and extrapolating results to estimate ex post measures.

The Energy Division managed a budget of $97 million, representing one of the largest energy efficiency impact evaluations in the world, which was implemented by leading evaluation professionals. The focus of its studies was to verify IOU self-reported energy savings and identify energy savings that would not have likely occurred in the absence of the program. The Energy Division report adopts the findings of numerous individual EM&V studies of the performance of various individual energy efficiency programs in the IOUs' portfolios for the 2006-2008 cycle. The studies form the foundation for updates to the utility ex ante savings assumptions used to estimate portfolio and program savings and cost effectiveness, and provide information for program improvements and future estimates.

The Energy Division Report synthesizes three years of program implementation and evaluation and presents the final outcomes of multiple billions of dollars in ratepayer investments. The Energy Division Report incorporates multiple attachments of data and tools that allowed for detailed review by stakeholders. Most pieces (i.e., Contractor Reports, Decision Framework and ERT) have been introduced to the public in advance of the Energy Division report release. The largest and most complex portion of the data (over 4 million tracking records) was provided by the IOUs and standardized in collaboration with Energy Division consultants over the course of a three-year period.

Nevertheless, significant controversy remains with respect to the Energy Division evaluation reports. The information in the Energy Division Report may be valuable and useful for a variety of purposes, including in the planning of future energy efficiency portfolio design. We have utilized certain aspects of Energy Division's Report, as reflected in Scenario 3, for purposes of calculating the incentives adopted herein. However, for reasons explained below, we conclude that for purposes of the 2006-2008 true-up, we shall not rely solely on the results contained in the Energy Division report.

5.3. Use of Ex Ante versus Ex Post Measures for Measuring Savings

As noted by Energy Division, the energy efficiency savings goals for the last two program cycles (i.e., 2004-2005 and 2006-2008), were developed from analyses conducted in 2002-2004. As a result, significant variances exist between the savings estimates from the Energy Division ex post evaluation and the assumptions underlying the original ex ante assumptions used to develop the Commission's efficiency goals. This is not because those initial assumptions were necessarily inaccurate when they were adopted, but because market dynamics are likely to have changed in the intervening years.

The 2006-2008 energy efficiency net benefits used to determine final incentive earnings varies significantly depending on how key parameters are quantified. Parties disagree, in particular, on the appropriate values for the NTG ratio, expected useful lives, and in-service installation rates. The IOUs and NRDC advocate using ex ante values from the 2005 DEER. The Energy Division Evaluation Report calculated updated ex post values for these measures. A key factor contributing to the differences between ex ante and ex post savings found in the Evaluation Report is the much lower than expected impact attributed to interior screw-in lighting measures. Because they made up a significant portions of the portfolio, adjustments to NTG ratios, installation rates, and unit energy savings based on the Energy Division evaluation all contributed to these impacts.

In the aggregate, utility self-reported energy savings during 2006-2008 were claimed at the level of 151% of the adopted goals. By contrast, the Energy Division evaluation shows that energy savings equal to just 62% of the adopted goals. Similarly, utility self-reported demand savings for 2006-2008 were claimed to be 122% of the savings goals, but the Energy Division evaluation showed demand savings amounting to 55% of the savings goals.

In D.05-04-051, the Commission adopted principles requiring ex post updates "as a general policy" in the true-up of energy efficiency savings for programs implemented in 2006 and beyond, requiring:

A true-up of ex ante (pre-installation) assumptions for program participation (e.g., types and number of measures or equipment) with actual participation verified on an ex post basis, i.e., during and after program implementation.

A true-up of ex ante program costs assumptions with actual expenditure levels.

As a general policy, ex post evaluation of per unit kWh, kW, and therm savings through load impact studies. An exception to the general policy may be appropriate for measures and/or programs for which there are well-established ex ante values with a high degree of confidence, and low external sources of variability that could influence the energy savings.

Persistence studies will not be tied to the performance basis, but shall still be performed to inform future planning. This policy shall be revisited and revised, as appropriate, if there is evidence at a future date that the results of persistence studies are significantly different from the ex ante estimates.

(D.05-04-051 at pp. 92-93, Ordering Paragraph No. 8 a-d.)

In accordance with these policies, and through evaluations and other research conducted since the original goals were developed, Energy Division developed updated, end user adoption rates, and per unit savings levels. One of the key principles underlying the original design of the RRIM, as adopted in D.07-09-043, was that key parameters were to be trued up based on updating of net energy savings that are based on actual ex post load impact studies, and subject to independent verification. In D.07-09-043, we expressly stated:

...[P]otential earnings for the 2006-2008 program cycle start at $176 million if all four utilities achieve the minimum performance threshold of 85%, which in turn would deliver approximately $1.9 billion in net benefits. That is, if the utilities actually produce net benefits of $1.9 billion (based on verified costs and resource savings) when they reach 85% of the savings goals, then their shareholders will receive $175 million of those net benefits under the shared-savings structure we adopt today. (D.07-09-043 at p. 10, emphasis added.)

In D.07-09-043, we acknowledged DRA's concern that failure to update may create a perverse incentive, because:

... an approach that fails to true-up savings and net benefits (PEB) accomplishments based on the results of final load impact studies creates a perverse incentive for utility managers to promote exaggerated savings assumptions during the planning process. This is because the utility knows that it can get progress payments based on these inflated estimates that are not returnable when the final true-up reveals lower load impacts. (D.07-09-043 at p. 123.)

We previously denied a request by the IOUs to remove the requirement for updates of key parameters in assessing RRIM earnings. (See e.g., D.08-12-059.) In denying the IOUs' earlier request to reconsider the requirement we adopted in D.08-01-042 for updating of parameters, we explained in D.08-12-059:

At this point we do not think it would be reasonable to remove, in part or in whole, the requirement that the ex ante assumptions used to assess interim claims be updated. This updating is part and parcel of the balance that was struck in D.08-01-042 between providing utilities the ability to book interim rewards without the uncertainty that they would have to return these interim amounts after the fact, and limiting the risk to ratepayers of overpayment. (D.08-12-059 at p. 19.)

We reiterated the importance of this principle in D.09-12-045 where we relied upon updated assumptions in the Energy Division verification studies as the basis for the net benefits used to allocate incentive awards. By not updating ex ante assumptions, to the degree more accurate assumptions are available, we feared we could risk being left with an outdated basis for measuring cost savings and associated incentive payments. We have also previously stated that the earnings true-up would reflect updated assumptions in the DEER, as noted in D.08-01-042:

Updating measure load impacts using the DEER database prior to the payout of interim claims in 2008 and 2009 should help to mitigate the risk of extremely large swings in earnings (positive or negative) at the final earnings true-up, which serves the interests of both utility shareholders and ratepayers. (D.08-01-042 at p. 17.)

For purposes of determining the actual impacts of energy efficiency programs in reducing demand and obviating the need for supply side resources, it is clearly incumbent on the Commission to update the assumptions used to quantify the impacts of the utilities' efforts. Because the actual impacts of energy efficiency play a key role in determinations of supply side resource need, it would be inappropriate to assess savings achieved from energy efficiency based on outdated assumptions in this context. If, for example, a given variable, like a measure expected useful life, is found to be less than what was assumed when the utility portfolios were adopted, it would make no sense to ignore that update and rely on an exaggerated estimate of energy savings since this will have real world impacts in terms of Commission determinations of supply-side resource need once energy efficiency has been accounted for. Similarly, in the case of variables like NTG ratios, which do not, in of themselves, impact the gross savings of a given measure or program, it is incumbent on the Commission to evaluate these as it does drive determinations of energy efficiency measure and program cost effectiveness, thus impacting program portfolio design. However, in the context of the incentive framework, as explained below, our experience with the mechanism over the past three years, establishes that the reliance on ex post updating creates an unreasonable amount of risk for the utilities because it results in an imposition of unrealistic expectations regarding their ability to anticipate and respond to changes in thousands of parameters that influence program performance.

In D.08-01-042, the Commission endorsed the idea that failure to update the ex ante assumptions may create a perverse incentive for utility program managers to exaggerate savings assumptions during the portfolio planning process. While such an incentive may exist absent updating, on further reflection this theory failed to account for the fact that the utility portfolios are submitted for review and approval by the Commission with extensive opportunity for feedback from stakeholders. Consequently, any claims by the utilities regarding the cost effectiveness or savings potential of their portfolios are expressly subject to Commission review. During that review, the Commission may receive information from sources other than the IOUs as well. In conducting that review, the Commission must make a determination regarding the cost-effectiveness of the utility portfolios and their ability to meet the energy efficiency savings goals we have adopted for the particular period. Nothing in that process prevents the Commission, or parties, from contesting the assumptions made by the utilities based on other, more objective and/or reasonable sources of information.

Over the 2006-2008 period there has been profound disagreement on the appropriateness of the various adjustments to many of the underlying assumptions and parameters driving the estimated performance of the utility programs. In our view, however, these disputes raise a fundamental question regarding the fairness in how the mechanism, originally adopted in D.07-09-043, actually operates. In particular, the intense debate over factors like the net to gross ratios, measure expected useful life, and the residential/non-residential installations of incented lighting products under the utilities' upstream lighting programs, has caused us to consider whether the incentive mechanism appropriately rewards or penalizes the utilities for things that could be reasonably anticipated or are within their control. Or, instead, does the mechanism reward or punish the utilities for a variety of factors over which they have only limited control or ability to anticipate and respond to? The efficacy and legitimacy of the incentive mechanism hinges fundamentally on the ability of the utilities to modify their programs and portfolios over the course of the 2006-2008 program cycle in response to changes in the various parameters that influence measure savings and attribution.

As a practical matter, however, the ability of the utilities to reasonably anticipate, much less respond to, such changes is limited. Assessing the changing dynamics of the energy efficiency environment and market is a profoundly difficult task. This fact is reflected in the tens of millions of dollars in ratepayer monies that are allocated to EM&V activities²⁰ each year to develop the very same estimates that the mechanism, as implemented, to date, all but requires the utilities to anticipate. An argument has been made that because of the ongoing EM&V activities of Energy Division, the utilities had ample information available to them regarding changes in some of the key underlying assumptions. In light of that information, some parties argue, the utilities could and should have modified their portfolios accordingly. As an example, prior to the incorporation of formal updates to DEER in October of 2008, draft EM&V studies of the 2004-2005 energy efficiency programs were made available to parties. Those draft EM&V studies indicated, among other things, that NTG values for lighting were declining. A legitimate argument may be made that these results could be reasonably deemed final, and actionable, in October of 2007, when the 2004/2005 Statewide Residential Retrofit Single-Family Energy Efficiency Rebate Evaluation (Itron Report) was published. However, an equally valid point is that prior to that date, these updated assumptions were merely preliminary and subject to additional review by parties and the Energy Division, they were not final and, thus, not actionable.

The IOUs argue that the NTG updates in the Energy Division Verification Report are fundamentally flawed, and, even if correct, occurred too late in the 2006-2008 cycle to enable the IOUs to make meaningful mid-course adjustments in program funding in response to the updated NTG ratio. By way of example, for PG&E's programs, allocations of incentives to upstream lighting manufacturers/distributors must be made at least 120 days prior to the movement of the products into the marketplace. Therefore, the IOUs argue, the October 2007 report, even were they to accept them as accurate, allowed little time for adjustments to program delivery and implementation to take hold during the 2006-2008. They further argue that it is inappropriate to apply these NTG values to the entire 2006-2008 program cycle for purposes of awarding incentives. We agree.

Until the review process has run its course and numbers are adopted as final, we do not think it is reasonable to, in effect, require the utilities to modify their portfolios as if preliminary assessments are, in fact, final. To do so undermines the purpose of the review, and it essentially prejudges the outcome of that process. A more reasonable approach and expectation is for the utilities to modify their portfolios based on assumptions available to them at the time they are developing and implementing their portfolios. We do not believe the changes to the parameters, and the magnitude thereof, that result in the dramatic swing in earnings under the incentive mechanism, as adopted, were available in a manner that would have allowed the utilities to react in a timely manner.

We also note that controversy over key parameters, most notably NTG ratios, was discussed in D.05-09-043, which authorized 2006-2008 programs. D.05-09-043 cautioned that "[s]pecific sensitivities around the NTG ratio assumptions indicate that the proposed portfolios may not meet the cumulative 2006-2008 energy (GWh) savings targets."²¹ The Commission found some risk that the portfolio plans may not meet the Commission-adopted GWh and therm energy savings goals, due to uncertainties over free ridership assumptions and the useful life estimates associated with certain lighting measures, among others.

The Commission directed that NTG ratios used for planning purposes would be "further addressed through ex post true-up of these ratios in performance basis evaluation, consistent with our direction in D.05-04-051."²² In recognition of the uncertainty regarding whether the assumptions underlying the achievement of savings goals were realistic, the Commission did not direct that those assumptions remain frozen throughout the 2006-2008 program cycle for incentive purposes. Instead, we stated:

Our decision today on how best to bound the uncertainty associated with this key savings parameter for planning purposes is predicated on the expectation that NTGs will in fact be adjusted (trued-up) on an ex post basis when we evaluate actual portfolio performance. We believe that this is entirely consistent with the resolution of threshold EM&V issues in D.05-04-051. (D.05-09-043 at p. 97.)

Looking back, we see that our expression was indicative of concerns regarding the uncertainty around the NTG ratios, and the possibility that the NTG ratios used in developing the portfolios were too high. In our view, because these concerns are expressed only in qualitative terms and based on preliminary results, this information provided an insufficient basis for the utilities to act. Given the preliminary nature of the information available to the utilities over the 2006-2008 period regarding changes to key parameters, the expectation that they should have dramatically modified their portfolios in a manner sufficient to avoid the adverse consequences under the incentive framework is unreasonable.

We point out that when the RRIM was adopted in D.07-09-043 it allowed for interim claims subject to a holdback of 30%. That we realized within a short time that a 30% holdback was smaller than necessary, is illustrative of the difficulty in anticipating the magnitude of the changes Energy Division has observed in the various parameters that affect the energy savings estimated to result from the utilities' energy efficiency programs. As stated in Ordering Paragraph No. 4(c): "Thirty (30) percent of the earnings calculated for each interim claim shall be "held back" until the final true-up claim, in order to minimize the risk of overpaying the utilities in their interim claims."(D.07-09-042 at p. 221, emphasis added.)²³ In order to minimize the risk of overpayment, the Commission initially thought that the possibility of the incentives changing by more than 30%, based on ex post review, to be relatively remote. However, the results of Energy Division's Verification Reports strongly indicate that this assumption was incorrect. For the 2006-2008 portfolios, the estimated incentive earnings the utilities would have earned if their programs were evaluated on the basis of ex ante assumptions would have been $307 million.²⁴ Yet changes in the underlying parameters result in collective earnings declining to minus $45 million, a swing of $353 million in incentives. This represents a reduction of more than 100%.²⁵ This enormous swing is entirely due to changes in the underlying parameters, over which considerable dispute remains. Clearly the magnitude of the shift in the incentive amounts driven by these changes far exceeds the relatively substantial 30% holdback that the Commission adopted as a buffer in D.07-09-043, to minimize the risk of overpayment. The Commission itself failed to reasonably anticipate the magnitude of the dramatic changes to the parameters underlying its assessment of energy efficiency program performance and the huge swings this would cause in the incentive calculations. It is telling that the timing of D.07-09-043 predated by only one month the issuance of the 2004/2005 Statewide Residential Retrofit Single-Family Energy Efficiency Rebate Evaluation. Unfortunately, due to the timing of the two different events, information regarding the potentially significant reduction in the NTG ratios was not available when D.07-09-043 issued. It is now obvious that the 30% holdback adopted by D.07-09-043 to "minimize" the risk of overpayment, has since been proven to be too low, in light of the dramatic changes in the estimated savings and incentives based on the updates to the ex ante assumptions.

The forgoing review establishes that one of the fundamental premises on which the incentive mechanism adopted in D.07-09-043 was based was fundamentally flawed. Specifically, it was/is unreasonable to expect the utilities to anticipate the very substantial changes in a number of the key parameters over the three year cycle that drive their energy efficiency program results. Furthermore, given the after-the-fact timing of Energy Division's updates to these parameters, we find that the IOUs did not have the opportunity to modify their portfolios on the basis of this updated information in a way that would allow them to substantially avoid the adverse impacts of those updated assumptions on estimated program performance. Irrespective of the accuracy of the updates adopted by Energy Division, we find that the incentive mechanism as implemented was/is unfair to the utilities, in that it bases its results on assumptions the utilities cannot be reasonably expected to anticipate; and further, when those changed assumptions come to light, cannot be reasonably expected to respond to in a way that enables them to substantially avoid the adverse impacts on the estimated performance of their programs.

A more reasonable approach to assessing the 2006-2008 period for purposes of determining utilities' energy efficiency program performance and the associated incentive earnings is to rely on ex ante assumptions. These were the assumptions the utilities used in developing the portfolios that the Commission approved in D.05-09-043 for the 2006-2008 cycle.

Notably they are also the assumptions that align with the goals against which the utilities performance is being measured as noted in D.09-12-045.²⁶ D.09-12-045 recognized the inconsistency of holding the utilities, on one hand, to the achievement of energy savings goals which are not adjusted to reflect market transformation (as reflected in a much reduced NTG ratios), while discounting the efficacy of their programs in realizing energy savings based on the assumption that such market transformation had occurred. Since higher per-unit energy savings estimates were expected when the Commission developed the goals against which the utilities' programs are compared, significant downward adjustment in per unit savings, without a commensurate reduction in the goals could make it virtually impossible for utilities to achieve those goals.²⁷)

5.4. Shared Savings Percentage Rate
for the True-Up

While evaluating the utilities programs on an ex ante basis is straightforward, we also find that changing the incentive framework in this way substantially changes the risk profile of the incentive mechanism. By removing the updating provisions, the risk to ratepayers of overpayment (defined here as incentive payments that are justified on the basis of the assumptions relied on at the time the portfolios were adopted, but that with updated assumptions would not be earned), increases substantially. Conversely, the risk of penalties is greatly reduced to the utilities because not updating the ex ante parameters removes a key source of uncertainty in the mechanism, namely changes to key underlying parameters that occur over the three year program cycle. This change conflicts with a number of the criteria adopted in D.07-09-043. Many of those criteria were adopted in the hopes of eliminating all risk to ratepayers in the event that the portfolios turn out to have delivered fewer savings than had been anticipated when those portfolios were adopted, and to place that risk on the utilities. For reasons explained above, we find the mechanism adopted in D.07-09-043 did not provide a correct balance and was unfair in view of what we now are able to see are more reasonable expectations regarding the ability of the utilities to anticipate and respond to changing assumptions. While our modifications to the mechanism for the 2006-2008 true-up that reflects this change substantially reduces the risk to the utilities of incurring penalties, it does not wholly eliminate it. Even holding ex ante assumptions constant, the utilities may only receive incentive earnings in relation to actual portfolio measures verified as installed. This limitation will help to ensure that ratepayers fund incentives for programs and measures actually deployed. However, because the use of ex ante assumptions does alter the risk profile of the incentive mechanism by shifting additional risk to ratepayers, some additional modifications to the mechanism are necessary.

There are a number of levers available to the Commission for rebalancing the incentive mechanism in light of this shift to an ex ante approach. These "levers" are represented by the various factors or elements of the incentive mechanism, including the minimum performance standard and the placement of the various inflection points on the earnings/penalty curve, the caps applied to the penalties and rewards, and the shared savings rate. In our view, the most straightforward, reasonable and fair option is to reduce the shared savings rate, i.e., the percentage share of net benefits the IOUs may earn as incentive payments, to reflect the substantially reduced risk the utilities face under an ex ante approach. By using a reduced shared savings rate, the IOUs' potential earnings under the incentive mechanism shall be reduced relative to the mechanism adopted in D.07-09-043. This approach is consistent with the views expressed by DRA and TURN in the context of proposed reforms to the RRIM. Both DRA and TURN have argued that should the Commission modify the RRIM in a way that reduces the risk to the utilities and increases the risk born by ratepayers, that corresponding changes should be made to the shared savings rate and incentive cap.²⁸ We agree with the thrust of these arguments and find they are equally applicable in the context of modifications to the incentive mechanism as it applies to the 2006-2008 period.

We now turn to the issue of the magnitude of the reduction in the shared savings rate that is appropriate under the ex ante approach we adopt in this decision relative to the shared savings rate incorporated into the incentive mechanism adopted in D.07-09-043. As noted in that decision, "establishing the level of earnings opportunity for a shareholder risk/reward mechanism is ultimately a judgment call the Commission must make, and not a precise science."²⁹ The range of proposed shared savings rate was fairly broad, as indicated in Attachment 3 of D.07-09-043. DRA, TURN, and the Community Environmental Council (CEC) proposed shared savings rates generally below 5% of the Performance Earning Basis. In contrast, the utilities argued for much higher shared savings rates, ranging from 10% and increasing to as much as 30%, depending on the performance of the utility programs relative to the energy efficiency savings goals. NRDC supported a middle-ground level, closer to TURN's, CEC's and DRA's proposals, ranging from 6% to 12%. D.07-09-043 ultimately adopted a shared savings rate of 9% of the PEB in cases where the utilities' programs achieved between 85% and 100% of the energy efficiency goals, and a 12% shared savings rate if the utilities' programs exceeded 100% of the energy savings goals.³⁰ Because of the much reduced risk associated with an ex ante approach, we believe it is just, reasonable, appropriate and necessary to reduce the shared savings rate. In our judgment a shared savings rate of 7% provides the appropriate level of risk re-balancing to offset the effects of relying on ex ante assumptions to derive final PEB-related incentive levels. This relatively lower shared savings rate of 7% will be applied to the PEB calculated using ex ante assumptions instead of the 9% and/or 12% shared savings rates that were initially adopted in D.07-09-043 for the 2006-2008 true-up. By applying a 7% share of savings as incentive earnings of the 2006-2008 period, we preserve the remaining majority of energy efficiency savings as a ratepayer benefit.

For purposes of calculating incentives under the mechanism adopted herein, we rely primarily on the PEB assumptions set forth in Scenario 3 (S3), Template 1 (T1) included among the various scenarios that were developed and presented in the Energy Divisions 2006-2008 Scenario Analysis Report. This scenario reflects the use of ex ante assumptions (including those in the DEER at the time the utilities energy efficiency portfolios were adopted in 2005) adjusted to reflect verified installations. However, we do make some adjustments to this scenario to reflect certain modifications proposed by SCE in its comments on Commissioner Peevey's alternate proposed decision. In its comments, SCE proposed modifications to some of the ex ante parameters. The specific parameters that SCE proposed be changed in Scenario 3 were provided in Attachment B to SCE's comments on the alternate decision of Commissioner Peevey. During the three year period, neither PG&E, SDG&E, nor SCG modified their assumptions. SCE however, made some modifications in response to updated information. As a result, SCE argues, under Scenario 3, relative to SCE, PG&E, SDG&E and SCG would be rewarded for not making any changes because they are attributed more savings than they would have been had they modified these parameters. SCE argues that reliance on Scenario 3, absent its proposed changes, would effectively punish SCE for trying to "do the right thing" because many of the parameters reflected in Scenario 3 had been modified in a way that reduced the energy savings attributable to SCE's programs.

We asked the Energy Division to review SCE's recalculation of the incentive amount incorporating the parameter changes it proposes. After consulting with Energy Division, we agree to modify Scenario 3 to reflect SCE's proposed modifications. The parameters and changes to those parameters identified by SCE are provided in Table 5 below.

Table 5: Ex Ante Parameter Modifications Proposed by SCE and Reflected In Scenario 3

Utilizing the adjustments reflecting SCE's proposed modifications, under this scenario-template combination, both the utilities' program performance against the minimum performance standard and the performance earnings basis are calculated using ex ante assumptions. Applying a 7% shared savings rate yields the results provided in Table 6 below. This scenario-template combination includes the 2004-2005 cumulative savings goals used in determining the relevant minimum performance standard, as well as attribution of 50% of the savings attributed to codes and standards development to the utilities, for purposes of determining program performance relative to the minimum performance standard. With regard to interactive effects, S3-T1, as modified, does not reflect any specific updates, nor does it account for interactive effects, either positive or negative.

Table 6: Scenario 3, Template 1 RRIM Results Modified to SCE compliance ex ante numbers and a 7% Shared Savings Rate

¹⁶ The Energy Division Final Report used an updated E3 calculator that corrected the error for natural gas therm savings that was identified by SDG&E in its Petition to Modify D.09-12-045, filed on February 19, 2010. Because we incorporate this correction into our true-up, the referenced Petition to Modify D.09-12-045 is rendered moot, and therefore, we shall dismiss it.

¹⁷ See also, D.07-09-043 at p. 4.

¹⁸ See e.g., Evaluation Report, Appendix O for a compilation of comments and responses.

¹⁹ The updates applied, the source of the update, and the justification of the values were provided by each group, and presented in Appendix C of the Energy Division Report.

²⁰ D.05-11-011 authorized $162,794,829 for EM&V activities over the 2006-2008 energy efficiency program cycle.

²¹ D.05-09-043 at p. 56. See, generally, the discussion concerning the Case Management Statement at pp. 53-56.

²² Id. at p. 167, Finding of Fact No. 7.

²³ Ordering Paragraph No. 4 of D.07-09-043 was modified by D.08-01-042, where, among other things, we increased the holdback to 35%. (D.08-01-042 at p. 14.)

²⁴ 2006-2008 Energy Division Scenario Analysis Report at p. 39.

²⁵ 2006-2008 Energy Division Scenario Analysis Report at p. 30, Table 9.

²⁶ D.09-12-045 at pp. 68 and 79 Finding of Fact No. 19.

²⁷ See D.09-12-045 at p. 68.

²⁸ DRA Post Workshop Comments and Further Recommendations for the 2009-2011 Shareholder Incentive Mechanism at p. 13; TURN Post Workshop Reply Comments on Energy Efficiency Incentive Mechanisms at p. 4.

²⁹ D.07-09-043 at p. 104.

³⁰ D.07-09-043 at p. 8.