4. New Service List for the 2006 Update

Per D.05-09-043, all reports, notices of availability, notices of workshops or other filings related to the 2006 Update are to be distributed to the service list in this avoided cost rulemaking (R.04-04-025), the energy efficiency rulemaking (R.01-08-028), the distributed generation rulemaking (R.04-03-017), the procurement proceeding (R.04-04-003), including any separate service list established in that proceeding that is specific to resource adequacy issues, and the demand response rulemaking (R.02-06-001). The temporary service list in this proceeding will consist of these services lists, as well as the service list in the renewable portfolio standard rulemaking (R.04-04-026), until further notice. However, because these service lists are very lengthy and duplicative, I am establishing today a process for creating a new service list for this phase of the proceeding.

For this purpose, all interested parties are required to notify ALJ Meg Gottstein, who has been co-assigned to this proceeding, by no later than December 16, 2005 in order to be placed on the new service list for the 2006 Update phase of this proceeding. Notification should include all of the following:

· The name of the individual and affiliated organization

· Mailing address and phone number

· Electronic mail address.

All those who wish to be placed on the permanent service list should mail a hard copy of this notification ALJ Gottstein at 505 Van Ness Avenue, Room 5044, San Francisco, California, 95689 and also send an electronic copy to meg@cpuc.ca.gov.

IT IS RULED that:

1. Interested parties shall file and serve comments on the proposed scope and schedule for the 2006 Update phase of this proceeding, including the discussion of the data collection and longer term updating issues presented in the November 1, 2005 Joint Report. Concurrent comments are due by December 15, 2005.

2. Until further notice, the service list for all submittals related to the 2006 Update phase of this proceeding is comprised of the service lists in this proceeding (Rulemaking (R.) 04-04-025) and the following resource proceedings: the energy efficiency rulemaking (R.01-08-028), the distributed generation rulemaking (R.04-03-017), the procurement proceeding (R.04-04-003), including any separate service list established in that proceeding that is specific to resource adequacy issues, the renewable portfolio standard rulemaking (R.04-04-026) and the demand response rulemaking (R.02-06-001).

3. A new, permanent service list shall be established for the 2006 Update phase of this proceeding as described in today's ruling.

4. All comments and other submittals in this phase of the proceeding shall be served pursuant to the Electronic Service Protocols attached to R.04-04-025 and consistent with Rule 2.3 and 2.31.

5. This ruling shall be served on the service lists in this proceeding and in R.01-08-028, R.04-03-017, R.04-04-003, R.04-04-026 and R.02-06-001.

Dated December 7, 2005, 2005, at San Francisco, California.

Pages 108-114

[T]he comments raise important questions concerning the appropriateness of using the daily average peak reduction metric of performance in the broader context of how we should value energy efficiency across proceedings: Is this definition of peak load reductions appropriate in the context of resource planning and resource adequacy counting rules? Is there another definition that is more appropriate that we should work towards incorporating into the E3 calculator? Do we need to have identical definitions of peak demand reductions for all purposes (e.g., energy efficiency cost-effectiveness evaluation, establishment of energy efficiency peak reduction goals and evaluating achievement of those goals and resource adequacy counting), or do we just need to ensure that there are clear and consistent crosswalks between them to meet both program and resource planners' needs? These are fundamental issues that we should consider before adopting a common definition of peak for energy efficiency planning and evaluation purposes. As discussed further below, we will address these and other related issues in conjunction with the process we establish in today's decision for updating avoided costs and making necessary refinements to the E3 calculator. (See Section 8.8 below.)

We recognize that until these longer-term definitional and methodology issues are fully addressed, we will need to move forward with calculations of peak demand reductions during the compliance phase that are subject to modification when we resolve these issues in 2006. However, we prefer this situation to one where we attempt to impose a common definition of peak load reductions now that will also be subject to change, and in doing so, cause potentially significant delays in roll out of the 2006 program plans as we sort through the issues outlined above. Moreover, as described in this decision, we will be updating other inputs for our assessment of the performance basis for the 2006-2008 program cycle after the bid solicitation cycle is complete, i.e., avoided costs and EUL assumptions. (See Sections 8.2 and 8.8.) We will also be making corrections/refinements to the E3 calculator model and consider improvements to the underlying load shape data, as part of this updating process.

Given the considerations outlined above with respect to the definition of peak, we believe it is more prudent to include this issue in the post-compliance phase updating process as well. In this way, we can develop the performance basis for this next three-year program cycle that incorporates all the updating discussed in this decision, based on a careful and coordinated consideration of the issues. This will enable us to establish a performance basis for the 2006-2008 program cycle that provides a solid foundation for performance incentive mechanism discussions.

We plan to complete this updating process by mid-2006. As discussed in Section 8.8 below, the updated performance basis parameters and definition of peak savings that result from this process will be used to evaluate performance for the 2006-2008 program cycle.

The utilities may need to rebalance some of their program offerings and budget allocations based on these updates, using the funding shifting rules adopted by this decision. We recognize that this introduces some uncertainty with respect to program planning and budgeting during the upcoming compliance phase competitive solicitations. However, this is unavoidable unless we completely delay the solicitations until we have completed our updates to performance basis inputs (including avoided costs), refinements to the E3 calculator and consideration of peak demand definition issues. These efforts will take several months, even on an expedited schedule.

We do not believe that it is in the public interest to forgo the savings that can be achieved with the completion of the compliance phase and roll out of the portfolio plans in early 2006, while we undertake necessary refinements to the performance basis that will require more time to complete. As discussed in this decision, we expect that the portfolio plans (including the measures offered) will be adjusted continually throughout the program cycle in response to market feedback and other information. It is therefore unrealistic on the part of third-party bidders and other stakeholders to expect that once the compliance phase is complete, there will be no changes to the program offerings or the budgets allocated to them. Instead, those program offerings and budget allocations will change overtime, and in this instance, some of those changes may be necessitated by improvements in our valuation of avoided costs, in our definition of peak savings and the other refinements we discuss in this decision.

In the meantime, the utilities should meet with interested parties to discuss all the cost-effectiveness inputs in the E3 calculators, as suggested in their comments. This meeting should be held by the utilities, led by the E3 consultant that developed the calculators under contract to them, within 15 days from the effective date of this decision. It should be structured similarly to the April 18, 2005 workshop in our avoided cost proceeding, where all the energy efficiency avoided costs and cost-effectiveness calculator details were discussed. However, in anticipation of the level of detail that will be of interest to participants to this proceeding, each utility is directed to make available the underlying load shape data used to develop the inputs to their respective E3 calculator model to all interested parties several days prior to the workshop. The E3 consultant should be prepared to describe in the workshop how the 8760 hours of adopted avoided costs were mapped to that load shape data, particularly for the summer peak hours.

We believe that there is considerable value in further information exchange at this juncture, so that interested parties become more familiar with how the calculator produces peak savings estimates for the portfolio as a whole, as well as for specific types of measures, as the utilities move into their compliance phase solicitations and filings. There will clearly be continued disagreements over what elements of the E3 calculator model, underlying load shape data and avoided cost "mapping" approaches (in addition to the peak demand definitional issues) need to be revised for the future. This workshop is not the forum for debating or resolving these disagreements. Rather, its primary purpose is informational. However, we expect that the discussions will also help Joint Staff and interested parties begin to identify what issues should be addressed during the post-compliance phase updating process, described further in Section 8.8 below.

Another purpose of the workshop discussion will be to identify any E3 calculator (model or input) "fixes" that are relatively easy to implement and where there is general consensus that such modifications are appropriate. For example, the CMS document indicates (based on the TecMarket Works report) that there are existing counting period inconsistencies with respect to how the E3 calculator accounts for peak load reductions. There were also anomalies identified with respect to how the E3 calculator produces the Standard Practice Manual cost-effectiveness results. These may be areas where the utilities and their E3 consultant, after further input from workshop participants, can easily resolve the inconsistencies in time for the upcoming competitive bid solicitations. There may be other examples that emerge from this informal process of information exchange.

After the informational portion of the workshop is concluded, workshop participants should engage in discussions on what improvements can be made relatively quickly to the E3 calculator model. The utilities are authorized to make further refinements to the E3 calculators based on the feedback that they receive during the workshop, and are directed to describe those changes in the November 1 filing discussed below. However, we will hold over to the updating process described in Section 8.8 the longer-term improvements/refinements that need to be considered with respect to the calculation of energy efficiency peak load reductions.

Regardless of the final definition of peak savings we choose to adopt (e.g., daily average, coincident, non-coincident), the Commission will need the E3 calculator and cost-effectiveness calculations in general to be based on the best available data related to the shape or pattern of energy savings over at least the four to seven hours of the peak period. This type of data is also needed to establish resource adequacy and for resource planning in general. In particular, as we move to refine our accounting of energy efficiency savings for resource planning purposes, including resource adequacy, it will not be sufficient to simply multiply annual savings by one factor (e.g., the 0.217 conversion factor used to translate the Commission's GWh savings goals to MW peak load reduction goals) without any knowledge of what is happening during the hours of the peak period.

Therefore, Joint Staff and the utilities, with input from interested parties, should also use this workshop process to begin to identify for which measures/programs additional or better quality hourly data needs to be collected. We expect such improvements to be reflected in ongoing data collection activities throughout the program cycle, and reflected in specific evaluation and measurement projects under the EM&V plans.

By November 1, 2005, the utilities shall file a report summarizing the workshop discussion and reporting the E3 calculator refinements that they have made in response. Based on the workshop discussion, the report should also present a preliminary list of issues that participants recommend be addressed during the updating process described in Section 8.8. The report should also present the workshop discussion on the data collection needs discussed above. The utilities are encouraged to hold additional workshops in October, as time permits, to further discuss the data collection and longer term updating issues with their PRGs and interested parties before preparing their report. The Assigned Commissioner or ALJ will solicit written comments on the final report to assist in scoping the issues for the 2006 updating process.

In addition to any other refinements to the E3 model that results from these workshops, the utilities should incorporate a correction to the erroneous demand reduction estimate for lighting currently contained in DEER that was identified during the course of this proceeding. In particular, SDG&E acknowledges that it needs to reduce residential CFL impacts by a factor of 2.34 in upstream lighting because DEER erroneously incorporated the wrong demand reduction.7 If this error is applicable to lighting measures in the other utilities' portfolio plans, they are also required to make the appropriate adjustments for the compliance phase filings.

In response to concerns over our current avoided cost valuation of peak demand reductions, 8 in particular for those hours that are considered "critical peak," we take immediate steps today to evaluate the issues raised in this proceeding as part of the avoided cost updating process anticipated by D.05-04-024. The proper valuation of peak load reductions, however we may define those hours, is needed whether such reductions are achieved through energy efficiency measures, distributed generation or demand response. As we observed in D.05-04-051, it is far from clear how critical peak avoided costs should be used in the context of energy efficiency measures that are not fully dispatchable. This issue will need to be explored during the updating process. We describe that process in Section 8.8 below.

Pages 137-143 (Section 8.8 "Avoided Costs/E3 Calculator Related Issues")

During the course of this proceeding, the following issues were raised with respect to current avoided costs and the E3 calculator model used to calculate cost-effectiveness:

· The E3 calculator presents cost-effectiveness results that are inconsistent with the California Standard Practice Manual. For example, when an incentive equals the full cost of the measure, such as when a refrigerator is given away at no cost to the participant or when a program is offering incentives above the incremental cost of the measure.9

· Each of the utility E3 calculator models uses a different "counting period" with respect to the calculation of peak demand savings, whereby the calculator for PG&E only counts kW savings for programs with a useful life of five years or greater. For SDG&E and SCE, this counting period is three years and two years, respectively.10

· The E3 calculator does not easily display the underlying load shapes being used to estimate the peak savings.11

· The current avoided costs do not value savings during critical peak periods for each utility (top 100 hours of peak demand each summer, typically occurring for a few hours a day on 8 to 12 days per year) differently from saving energy during the summer peak period.12

Parties disagree on how to address these issues, particularly with respect to the valuation of critical peak load reductions. SDG&E, for example, contends that the current avoided cost methodology appropriately values avoided costs during critical peak periods, and the problem lies solely with the manner in which the E3 calculator needs to be modified when the full 8760 hour load shape for an energy efficiency measure is not available.13 In contrast, the comments of TURN and Proctor Engineering imply that current avoided costs do not adequately reflect the demand reduction value during the top 100 hours of demand, i.e., they are too low. PG&E, on the other hand, suggests that there are more fundamental changes to avoided cost valuation (and the definition of peak or critical peak) that should be considered in order to properly value capacity consistently across all resource options, and in the context of the resource adequacy counting rules that are being developed in our procurement proceeding.14

The debate over the E3 calculator and associated avoided cost valuation also raises the following corollary issue: What load shape data currently underlies the E3 calculations, and how can we establish a more uniform set of assumptions/methods that are appropriate for translating annual energy savings from energy efficiency measures into peak savings? The first part of this question will be addressed in the informational workshop we discuss in Section 8.3. The second part will be addressed as part of the updating process described below. As part of this process we intend to develop a common E3 calculator for use by all implementers, in order to facilitate an apples-to-apples comparison of projected savings and cost-effectiveness calculations. As ORA points out, a common calculator ensures consistency in assumptions (e.g., end-use load shapes, expected useful lives, net to gross values) while alleviating program implementers from the burden of carrying out data-intensive calculations involving hourly avoided costs and end-use load shapes.

The interim E3 avoided cost methodology adopted in D.05-04-024 clearly represents a vast improvement over the prior use of statewide average values that did not reflect on-peak vs. off-peak reductions, or utility-specific cost differences. At the same time, we fully anticipated that we would "continue to refine the E3 methodology and forecast" in Phase 3 of that proceeding:15

"As discussed in this decision, we intend to consider the permanent adoption of the E3 methodology for generating avoided cost energy forecasts for use in [Standard Practice Manual] cost-effectiveness tests used to evaluate energy efficiency programs. We will also consider any potential revisions to the E3 methodology in Phase 3 of this rulemaking.16

Based on the record in this proceeding, we believe that further consideration of the E3 methodology with respect to peak valuation, as well as the E3 calculator model-related issues outlined above, should be undertaken without delay. We recognize that of the tasks outlined above, refining avoided costs with respect to the value of savings during peak hours on the utility system is likely to be the most difficult and controversial. However, this clearly needs to be undertaken in order to more accurately evaluate the relative cost-effectiveness of various energy efficiency measures, as well as demand-response and distributed generation options in the future. How further refinements to avoided cost values will be used in the context of energy efficiency measures that are not fully dispatchable, should also be addressed.

Commissioner Kennedy is assigned to both our generic energy efficiency rulemaking (R.01-08-028) and our avoided cost rulemaking (R.04-04-025). Therefore, she is in the best position to coordinate the development of these avoided cost/E3 calculator refinements in consultation with the assigned ALJs. For this purpose, we believe that the most cost-effective and expeditious approach is to build upon the E3 work conducted in the avoided cost rulemaking. Consistent with the approach we have taken in that proceeding, we direct the utilities to contract with the appropriate expertise in consultation with Energy Division staff. The costs of the contract(s) will be paid for out of the utilities' portion of EM&V budgets for the 2006-2008 program cycle.

The contractor(s) will be tasked with developing a draft report with specific recommendations on (1) the definition of peak (and critical peak or other terms, as appropriate) demand reductions to use in evaluating energy efficiency resources, (2) refinements to avoided cost methodology/E3 calculator, and (3) improvements to the consistency in underlying load shape data and the methods by which that data is translated into peak savings estimates. In addressing these issues, the contractor(s) should take into consideration the specific issues and concerns raised in comments in this phase of the proceeding and during the informational workshops. The contractor(s) draft report will be due by February 20, 2006. Energy Division will hold public workshops on the draft report. The contractor(s) will be present to respond to feedback and questions concerning the proposed refinements. Based on that feedback, the contractor(s) will develop a final report addressing the issues discussed above.

Energy Division will then develop recommendations on these issues for Commission consideration. For this purpose, Energy Division may solicit pre- and post-workshop written comments from interested parties, obtain input from additional technical experts and/or take other steps as necessary to consider the recommended avoided costs/E3 calculator refinements. In consultation with Energy Division, the Assigned Commissioner or assigned ALJ will establish the schedule for the submission of Energy Division's recommendations for comments on those recommendations that will enable us to issue a decision on these issues during the first half of 2006 or as soon thereafter as practicable. Nothing in today's decision precludes the Assigned Commissioner or ALJ from taking additional steps to address these issues, including soliciting further input from technical experts or scheduling additional workshops, as they deem appropriate.

All reports, notices of availability, notices of workshops or other filings related to the avoided cost/E3 calculator refinements discussed above should be distributed to the service list in this proceeding, the energy efficiency rulemaking (R.01-08-028), the distributed generation rulemaking (R.04-03-017), the avoided cost rulemaking (R.04-04-025), the procurement proceeding (R.04-04-003), including any separate service list established in that proceeding that is specific to resource adequacy issues, and the demand response rulemaking (R.02-06-001.) Our draft decision will be issued for comment in our avoided cost proceeding. All those who are not currently parties to R.04-04-025 (i.e., listed as an appearance on the service list) and wish to reserve the right to comment on that draft decision should file a motion to intervene with the assigned ALJ in R.04-04-025 as soon as possible.

Even under an expedited schedule for this effort, we will not be able to consider Energy Division's recommendations and parties' comments in time to make our final determinations on them before we complete the compliance phase and program roll-out for 2006 begins. We note that strict application of our performance basis "true-up" procedures would require that the results of these efforts be used only on a prospective basis, and not to evaluate the performance results of activities undertaken during a prior program cycle. However, as explained below, we believe that the unique circumstances facing us as we embark on the 2006-2008 program cycle warrant a limited exception to this requirement.

In particular, the practice of using the same set of avoided cost assumptions for both planning and for performance evaluation makes sense when an established avoided cost methodology is in place, where updates generally reflect new forecasts of what generation resources are on the margin and their associated fuel costs. The risk of these types of forecasting errors is applicable to any resource decision made using the planning assumptions, and these errors generally move in either direction (over-estimation and under-estimation) without systematic bias over time. Therefore, we have ruled in the past that we would not adjust projections of avoided costs on a retrospective basis, to reflect these forecasting errors.

In contrast, the updates we are considering to avoided costs at this juncture relate to fundamental aspects of the interim avoided cost methodology that need to be addressed, i.e., whether that methodology appropriately values savings during critical peak periods and related issues that have been raised with respect to the appropriate definition of peak for energy efficiency across all proceedings. It would be unreasonable to ignore the resolution of these and the E3 calculation issues just because the timing for completion of this update, relative to the upcoming three-year program cycle, is off by a few months. Moreover, it is important that program administrators know that these improvements are in the making, and that they will be incorporated into the evaluation of 2006-2008 portfolio performance as they finalize their program selections during the compliance phase of this proceeding.

Accordingly, we put the utilities and all interested parties on notice that we will use the common definition of peak load reductions, improvements to avoided cost methodology and refinements to the E3 calculator that are developed through the process described above to assess the performance basis of the 2006-2008 portfolio and programs. We will also incorporate adopted improvements to the consistency in underlying load shape data and the methods by which that data is translated into peak savings estimates into the E3 calculators. The EM&V protocols being developed in a separate phase of this proceeding, will identify how and when this load impact data should be trued up to calculate performance basis for the 2006-2008 program cycle, per our direction in D.05-04-051.

(END OF ATTACHMENT)

CERTIFICATE OF SERVICE

I certify that I have by mail this day served the original attached Assigned Commissioner's Ruling Soliciting Comments on Scope and Schedule for Avoided Cost Updating Process Directed by Decision 05-09-043 and Establishing a New Service List for the 2006 Update Phase of the Proceeding on all parties of record in this proceeding and in Rulemaking (R.) 01-08-028, R.04-03-017, R.04-04-026 and R.02-06-001 or their attorneys of record.

Dated December 7, 2005, at San Francisco, California.

NOTICE

Parties should notify the Process Office, Public Utilities Commission, 505 Van Ness Avenue, Room 2000, San Francisco, CA 94102, of any change of address to insure that they continue to receive documents. You must indicate the proceeding number on the service list on which your name appears.

7 See Joint Reply Comments of SDG&E and SoCalGas on Parties' Comments, July 21, 2005, pp. 2-3; CMS, p. 11.

8 "Current avoided costs" are those avoided costs calculated using the E3 avoided cost methodology, as specified in D.05-04-024, and as set forth in the associated May 2005 compliance Advice Letter filings by PG&E, SCE, and SDG&E.

9 TecMarket Works Report, p. 9. See also CMS, p. 1. Our policy rules direct the utilities and implementers to perform cost-effectiveness analyses that are consistent with the indicators and methodologies included in the Standard Practice Manual. (See Rule IV.1.)

10 TecMarket Works Report, pp. 24-25; CMS, p. 13.

11 Id.

12 Id.

13 Joint Reply Comments of SDG&E and SoCalGas on Parties' Comments, July 21, 2005, pp. 3-4; Comments of SDG&E and SoCalGas on Interim Opinion, September 6, 2005, pp. 2-3.

14 Comments of PG&E on the Draft Decision, September 6, 2005, pp. 7-8, 10-11.

15 D.05-04-024, p. 37

16 Ibid., p. 3.

Previous PageTop Of PageGo To First Page