4. Parties' Comments on the Interim Application of the E3 Avoided Cost Methodology for use in the Energy Efficiency 2006-2008 Program Cycle

The parties have had several opportunities to comment on the E3 draft report, including pre-workshop opening and reply comments and post-workshop opening and reply comments. These filings provide the Commission with an adequate record on which to consider the avoided cost forecast contained in the E3 Report for use on an interim basis in the generation of avoided cost energy forecasts to be used in the evaluation of energy efficiency programs for the 2006 program year.

Although this rulemaking contemplates using E3's proposed methodology in all calculations and forecasts of avoided costs across all Commission proceedings, we do not address the comments raised by parties with respect to the applicability of the E3 avoided cost methodology for purposes of calculating long-run avoided costs for use in valuing Distributed Generation (DG), Demand Response (DR), Qualifying Facility (QF) pricing or other resource options and programs at this time. Pursuant to the February 18, 2005 ACR Consolidating R.04-04-003 with this Rulemaking for purposes of addressing QF issues, QF policy and pricing issues will be addressed in Phase 2 and all other long-run avoided cost issues, including the development of a common methodology, including the potential development of a common methodology input assumptions and updating procedures will be addressed in Phase 3 of this rulemaking.

The majority of parties recommend that the Commission issue an interim order approving time differentiated avoided costs for energy efficiency based on the E3 report. The only exception is SCE. PG&E notes that "SCE alone... takes exception to using the E3 avoided costs for energy efficiency planning purposes."²¹ However, even SCE acknowledges that the avoided costs currently in use for demand side management and energy efficiency (DSM/EE) program evaluation, which were originally adopted in 1999, need to be replaced. SCE also agrees that the development of a revised forecast of avoided costs for DSM/EE program design and ex post analysis is a critical issue that needs to be resolved in early 2005. While SCE's comments focus primarily QF pricing and other resource areas, SCE suggests that the E3 report (and associated methodology) "if suitably revised," may be useful for DSM/EE applications.²²

All other parties either support or are not opposed to the adoption of the E3 avoided cost methodology for use on an interim basis for the purposes of evaluating energy efficiency proposals for program year 2006. In particular, PG&E, SDG&E/SoCalGas, TURN, NRDC, and UCS each recommend that the Commission approve the E3 avoided cost methodology immediately for generating avoided cost forecasts to be used in the evaluation of energy efficiency programs for 2006, although SDG&E/SoCalGas each recommend that the Commission adopt certain technical changes to the model. PG&E and TURN also recommend certain changes to the model, but state that the Commission should adopt the E3 methodology and forecast for use in evaluating the 2006 energy efficiency programs immediately, and consider any potential changes at a later date.

Issues that were raised regarding the applicability of the E3 avoided cost methodology for evaluating energy efficiency programs related to both the avoided cost methodology and the data inputs used in the avoided cost methodology. As stated above, this decision is limited to evaluating the applicability of the E3 Methodology and forecast for purposes of the program year 2006 energy efficiency proposals. We do not repeat or discuss the parties' comments regarding the applicability of the E3 methodology to other resource options.

4.1. All-in Hourly vs. Separate Capacity and
Energy Costs

The issue of whether avoided costs need to be separated into capacity and energy components arose in the pre-workshop and post workshop comments of several parties. SCE strongly supports the separation of capacity and energy components of avoided costs. PG&E also supports this separation, but does not feel it is necessary immediately or in all applications.

For energy efficiency purposes beyond the 2005 and 2006 program years, PG&E recommends that the adopted methodology separate energy from capacity, in order to provide for a separate capacity value that could be used for dispatchable resources or as a replacement for the CCGT power plant that is used for the long-run avoided cost proxy. PG&E also suggests that the separation of capacity and energy avoided costs is needed to be able to give proper credit for avoided capacity to only those resources that reliably count for purposes of resource adequacy. PG&E further recommends that the estimation of avoided costs for capacity be stated for three categories of electric generation resources: peaking, intermediate, and baseload. PG&E states that this estimate can be accomplished within the E3 methodology by using the costs for the three categories to reshape the E3 load duration curves without changing the area under the curve.

SCE asserts that the Commission's implementation of a resource adequacy requirement has returned the utilities to a structure that requires the clear separation and identification of capacity and energy costs. SCE maintains that the full capital cost of a CCGT cannot be used as a proxy for capacity value alone because a portion of the capital costs are motivated by a desire to achieve fuel savings.²³ SCE recommends that the Commission use the deferral value of a CT for the avoided cost of capacity and system incremental cost (as determined using a production cost model) for estimating the avoided cost of energy instead of the CCGT approach proposed in the E3 report. SCE further recommends that the Commission use CT deferral costs and system incremental costs that are consistent with each utility's long-term procurement plan.

TURN strongly disagrees with SCE that capacity value is the cost of a CT. TURN asserts that the economic theory that established that the capacity value is based on the cost of a CT was established at a time when CTs were far less efficient than they are today. TURN argues that technology has changed, and that modern CTs are more efficient and offer more flexible operations than steam plants, therefore, the Commission can no longer simply add the full cost of a CT to market prices to calculate marginal generation costs, because the result would be a significant overvaluation of capacity. TURN states that this issue also relates to the dispatchability issues raised by PG&E, suggesting that because a CT has few limits on its dispatchability, it is clearly worth more than a program that can only be called upon to save energy for a limited number of hours per year.

GPI agrees, stating that "for across-the-board programs such as energy efficiency, as well as renewables and QFs, an all-in, properly profiled avoided cost is a better approach to use than the traditional method of separate energy and capacity prices based on unrepresentative TOU periods."²⁴

Discussion

Our primary goal in this phase of the proceeding is to identify whether the E3 avoided cost methodology and associated forecasts are appropriate for use in evaluating energy efficiency programs for 2006. One of the criticisms of the current avoided cost values is that they are outdated statewide average values that do not reflect on-peak vs. off-peak reductions, as well as utility-specific cost differences. E3 has presented us with a methodology that estimates current avoided costs by hour, time and location, a significant improvement over the current annual average method.

Forecasting avoided costs on hourly and location-specific bases better allows for the proper valuation of programs that target peak hours and particular locations.

The E3 report sets forth a clear example on this point with a comparison between existing avoided cost values verses new avoided cost values based on the E3 methodology:

"In Figure 1, we compare the results for three example electricity efficiency measures.²⁵ The three efficiency measures are air conditioning, outdoor lighting and refrigeration programs. The air conditioning measure (upgrade of a residential A/C [air conditioning] unit from 12 to 13 SEER) has a total avoided cost savings of $133/MWh with the new avoided costs compared to a savings of approximately $80/MWh under the existing avoided costs (this is equivalent to $0.133/kWh and $0.080/kWh as shown on the right-hand y-axis). The large differential in avoided costs under the two forecasts exists because the majority of the savings in an A/C upgrade occurs during the summer peak period when both the generation avoided cost and T&D avoided costs are highest. In contrast, the value for outdoor lighting efficiency drops under the new avoided costs from $80/MWh to approximately $61/MWh. For outdoor lighting, there is no T&D avoided cost benefit because the savings occur at times when the T&D system has excess capacity. Refrigeration, which is assumed to have a flat energy savings profile has a closer comparison under both sets of avoided costs ($75/MWh for the new avoided costs and $80/MWh in the existing avoided costs), and similar proportions of generation, T&D, and environmental avoided cost." (E3 Report, pp. 11-12.)

Figure 1

In response to the parties' comments, E3 states that capacity costs in $/kW-yr. form are not needed if analyses are performed using the hourly costs because hourly costs arrive at the same capacity value for a DSM/CEE measure as using $/kW-yr capacity costs with a weighted-average kW impact. (This weighted-average kW impact approach would be similar to the PCAF-weighted load factors that PG&E and SCE have been using for their cost of service analyses).

E3 further notes that, although it is not required for the evaluation of energy efficiency programs, the annual stream of total generation avoided costs can be separated into an annual value of generation capacity and residual energy. For example, prior to the resource balance year, the level of market-based returns that a CT owner would earn by selling energy into the spot market is a reasonable measure of the value of capacity. This can be estimated as the difference between the estimated market prices (avoided costs) and the variable costs of operating a CT summed over an entire year. As California approaches resource balance, the CT owner's earnings should increase until it reached the full cost of owning a CT in the resource balance year.

Under this method, the resulting avoided costs would have the following characteristics. First, the capacity and energy costs would be equal to the full market price for firm delivered power forward contracts prior to resource balance. After resource balance, the capacity and energy costs would be sufficient to run and pay a reasonable return of and on a new CCGT. Second, the marginal energy costs would be capped at the running cost of a CCGT. Third, the capacity costs could be expressed in $/KW-year form. And finally, the value of capacity would be explicitly tied to the operating characteristics of the resource being evaluated. More efficient dispatchable resources with few operating restrictions would provide higher values of capacity. Conversely, resources that had limited ability to provide cost effective energy would have lower values of capacity.

We find that it is reasonable for the utilities to use the E3 avoided cost methodology and forecast, without modification to separate energy and capacity costs, for purposes of evaluating energy efficiency programs for program year 2006.

4.2. Environmental Adders

The E3 avoided cost methodology incorporated proposed environmental adders for NOx, PM-10, and CO2. SCE and SDG&E/SoCalGas expressed concern that the draft avoided cost methodology `double counts' the cost of NOx and PM10 emissions. The E3 report states that NOx and PM10 costs are internalized in the forward market prices used up until the resource balance year in 2008 (referred to as `Period 1' in the E3 Report). Therefore, NOx and PM10 abatement costs are only applied to the Long-Run Marginal Cost (LRMC) estimates after the resource balance year, and only on the residual emissions of new plants with the required abatement technology installed. E3 and SCE disagree as to whether abatement and permitting costs are included in the CEC plant cost numbers. The E3 report states that abatement and permitting costs are not included in the CEC plant costs, while SCE believes that abatement and permitting costs are included in the CEC plant cost numbers.

With regard to CO2, SCE and SDG&E/SoCalGas argue that it is inappropriate to include a separate adder for non-regulated pollutants because future regulation is speculative. Other parties, such as TURN and NRDC disagree, as do we.²⁶

Unlike criteria pollutants such as NOx and PM-10, which are regulated under the federal Clean Air Act and corresponding state legislation, CO₂ is not consistently regulated at either the federal or state levels. We recognize that CO₂ costs are not included in the marginal cost of producing electricity or thermal energy from natural gas today, and that CO₂ is strictly an unpriced externality. However, as discussed in the body of the E3 Report, there is a precedent going back to at least 1994 for including emissions costs in the avoided cost calculation for comparing efficiency measures in California (California Energy Commission Energy Report 1994 - ER94). The current avoided costs used for program evaluation adopted in D.01-11-066 in the Energy Efficiency Policy Manual include externality benefits for reduced electric and gas consumption including CO2. In addition, there is significant likelihood that the US and California will be a participant in the CO2 market over the life of the efficiency measures (as long as 20 years). The States of Oregon and Washington already regulate CO2 emissions from new power plants, and California has enacted legislation to limit CO2 emissions from automobiles, making state-level limitations on stationary sources more likely.

The California Energy Commission's finding that "Global climate change is real... and matters to California,"²⁷ also suggests that state level limitations on stationary sources are likely. Given the 20-year time frame of the E3 avoided cost analysis, it is highly likely that CO2 will be regulated and become part of the cost of producing electricity.

If we were to ignore the future cost of CO2 emissions, we would be assuming that the probability of CO2 emission costs being significant during the entire 20-year period would be zero, i.e., that the probability of the cost being zero is 100% for 20 years. While there is some finite probability that this cost will remain at or near zero during that time, it seems far more reasonable to assume that the future CO2 emission cost has a probability distribution over a range of values from zero to the high values forecasted by some analysts (see page 97 of the E3 Report for reference of forecasts up to $69/ton under some Kyoto compliance scenarios). This approach supports an intermediate price trajectory such as the $8/ton value used in the E3 Report.

A key goal of the Commission's energy efficiency effort is to reduce per capita energy use and peak demand through energy efficiency and a reduced reliance on fossil fuels. Assuming zero emission cost, or zero probability of significant emission costs for 20 years, would encourage utility investments that ignore the potential financial risk of CO2 emission costs. While CO2 limitation and regulation is controversial and uncertain, there is a wide range of potential cost levels, and an assumption of zero future does not adequately reflect the potential risk. Rather, a value that reflects the full range of reasonably possible outcomes would be more responsible.

We agree with the NRDC that it would be illogical to conclude that carbon emissions costs will be zero over the timeframe of the E3 report, as suggested by the SCE. Multiple scenarios do not suggest that the value should be zero, but rather than the value should fall within a reasonable range. It is reasonable to adopt a policy that requires utilities to calculate avoided costs using a methodology that incorporates a CO2 adder. The E3 report examines a range of carbon values from $5 to $69 per ton of CO2 and uses $8 per ton as a levelized cost in its analysis, based on a trend of $5 per ton in the near term, $12.50 per ton by 2008, and higher values thereafter. Adopting the E3 forecast of CO2 values as an adder in the avoided cost calculation and forecast reasonably reflects the cost to California of carbon emissions. The CO2 values adopted here will be used as an analytic tool in the evaluation of energy efficiency programs.

4.3. Generation Avoided Costs

SCE argues that the use of the full capital cost of a CCGT as a proxy for the avoided cost of capacity will misstate avoided costs for high-usage periods when CTs would be operating as the marginal units and low-usage periods when baseload units may be operating on the margin and CCGTs would not be in operation.

Both TURN and SDG&E/SoCalGas recommend modifications to the energy price forecast. TURN recommends that the energy price duration curve be modified at the bottom end because it expects few zero cost hours over the next 20 years, while SDG&E/SoCalGas recommends that the top end of the price duration curve be modified to contain the explicit cost of a CT if the commission intends to use the methodology outside of EE program evaluation.²⁸

Discussion

E3 states that it used three years of electricity forward price data published by Platts to estimate the market-based generation avoided cost for the period 2004-2006. The expected electricity price level for 2007 is calculated by escalating the 2006 electricity forward price using the NYMEX natural gas futures price. The avoided cost for the load-resource balance year of 2008 and beyond is the LRMC based on the all-in cost of a CCGT. E3 recommends this approach on the basis that the forward price data represent the best publicly available source for the four-year (2004-2007) sub-period for EE/DSM evaluation over a 20-year planning period.

E3 states that using forward price data reflects market prices, including capacity, because a forward contract obligates the seller to sell and the buyer to buy at a specific price for a specific quantity delivered to a specific location. Therefore, the energy delivery under a forward contract is firm. As part of forward price determination, the market assigns a value to the capacity used to ensure firm delivery of the contracted energy. This value does not necessarily track the historic fixed cost of capacity. The value is small (large) to reflect the expected surplus (shortage) in the capacity used for firm delivery. In the years prior to resource balance, the forward prices do not cover the full cost of a new entrant.

E3 is not providing a cost shape that assumes that CCGTs are the marginal plant for all 8,760 hours in the year. Rather, the CCGT is used to set the average annual market price. When this average price is applied to the hourly market shape, the result is that some hours will have costs higher than the CCGT annual average cost (when CTs would be on the margin) and some hours would have lower prices (when other baseload units would be on the margin).

The proposed generation costing methodology relies on California PX hourly NP 15 and SP 15 zonal prices from April 1998 to April 2000 to develop hourly market price values. The historical market prices incorporate bids from a variety of resources including CTs and CCGTs during both high and low-usage periods. The relative differences in the historical market prices over high and low usage periods is maintained throughout the forecast period by proportional scaling to reflect future market price quotes prior to resource balance, or the all-in cost of a CCGT for the resource balance year and beyond. E3 states that the historical hourly market prices over the 25-month period prior to the energy crisis provide reasonable price variations over time that are reflective of variations in both the level of energy usage by time period and the characteristics of different generating resources that might be the most cost effective resources by time period.

On balance, we conclude that the forecasting methodology in the E3 report should be approved, but with updated electric forward prices (through 2007). The parties claim that the Commission should use either a CT on a CCGT to represent avoided generation costs run contrary to our objective of developing a methodology that attempts to reflect hourly and utility-specific differences in avoided cost values. Certain suggestions also run contrary to our goal of adopting a methodology that is based on public information and easily updated.

We agree with the parties that we should use the most current market data, to the extent possible, and for this reason, we will update the E3 methodology to reflect current energy prices. The adopted approach is workable for the 2006 energy efficiency programs, but will remain subject to modification for future uses.

4.4. Market Price Referent (MRP) Assumptions

CCC suggested that the Commission should draw upon the record in the California Renewables Portfolio Standard (RPS) R.04-04-026, as the basis for a more detailed and sophisticated source of all-in CCGT costs for use in E3's long-run avoided cost methodology.²⁹ In R.04-04-026, we have considered the inputs and methodology to estimate the costs of baseload (CCGT) and peaking generation (CT) in order to establish a benchmark price for Renewables Portfolio Standard purchases called the Market Price Referent (MPR).

SDG&E/SoCalGas note that the E3 method and the Commission's RPS methodology are generally consistent, using natural gas forward prices for the first years transitioning to forecasts of fundamentals, but the specific methods are different.³⁰ SDG&E/SoCalGas state that both methods seem equally acceptable and suggest that the Commission allow the EE proceeding to use both forecasts for purposes of program evaluation.

Discussion

As SDG&E/SoCalGas notes, the E3 methodology gas forecast and the MPR approach are generally, but not entirely consistent. When E3's gas price forecast is input into the MPR model, the resulting annual average market prices are nearly identical to the results in the draft report.

We agree with the parties that the use of the same gas price forecast methodology would be ideal, given our intent to eventually develop consistent inputs and assumptions across all uses of avoided cost data. However, the MPR methodology currently does not offer monthly estimates of price or price differences by location. Since the energy savings associated with energy efficiency measures can vary greatly depending upon when and where they are used or installed, we view the availability of time and location specific forecasts as critical to the development of accurate avoided cost forecasts.

Consistent with the goals of this rulemaking, we will update the E3 methodology to reflect the CCGT cost inputs used in calculating the MPR, where applicable, but will not substitute the MPR methodology for the E3 gas forecast methodology at this time. Specifically, we will utilize the following CCGT MPR inputs in the E3 model, as set forth in Appendix C of the Revised MPR Staff Report issued on February 11, 2005, via ACR, as briefly described here: All "Capital Inputs" (although the E3 model only uses an average heat rate, not the new and clean value or a heat rate degradation factor); All "Finance Inputs" to the extent that these inputs are currently utilized in the E3 model; All "Power Delivery Inputs" to the extent that these inputs are currently utilized in the E3 model; and All the "Tax Rate Inputs" listed in Appendix C to the Revised MPR Staff Report.

With regard to updating the gas forecast, the E3 models will utilize a 60-day average of NYMEX gas futures data (similar to that done in the MPR gas forecast methodology.) The 60-day sample of NYMEX data should extend back from the effective date of this decision. The E3 models will also incorporate the same fundamentals forecast as was used in the MPR gas forecast model. Specifically, the MPR fundamentals forecast should replace the existing fundamentals forecast currently used as an input in the E3 methodology.

4.5. Transmission and Distribution Costs

SCE and SDG&E/SoCalGas raise issues concerning E3's proposed methodology for calculating T&D costs. In contrast to the existing values contained in the Policy Manual, E3's forecasts of T&D avoided costs are differentiated by utility service territory, customer class and season to recognize the time- and area-specific nature of the avoided costs. The report provides gas T&D avoided cost streams for core residential customers, core commercial/industrial customers and total core consumption. The avoided costs of each customer class are further allocated to the winter season (November through March), when the utilities normally experience peak demand. This approach is designed to allow the Commission to attribute greater value to DSM programs that (1) are implemented in areas with higher avoided costs; and (2) provide reductions when they are most needed -- at the time of the peak load for transmission and local peaks for distribution, as opposed to measures that affect off-peak consumption.

PG&E and SCE question the validity of a universal T&D avoided cost adder, arguing that consideration of T&D costs on a case-by-case basis is a more reasonable approach. In support of their positions, SCE and PG&E both reference prior to Commission Decisions in R.99-10-025, concerning Distributed Generation, and claim that these decisions support a finding that the impact of DSM/EE programs on transmission and distribution can only be ascertained through case-by case analysis. E3 responds, and we agree, that while a case-by-case analysis should be applied to determine payments related to specific projects for long-term conservation measures it is appropriate to credit programs with T&D avoided costs for program evaluation purposes.

The E3 Report does not present costs for specific investments, but averages numerous investments within large geographic areas. These costs are meant to be used for evaluating long-lived DSM/CEE programs that are being credited with the avoided cost of representative (not specific) investments. In this application, for long-lived measures with fairly predictable kW reductions over many hundreds of hours, E3 believes that the issue of "reliably in place" is sufficiently addressed through the use of hourly costs that capture the timing of the demand reductions, combined with traditional adjustments such as persistence factors.

The E3 report discusses the deration of T&D avoided costs, and supports well-reasoned adjustments to the level of T&D avoided costs used for program evaluations, but maintains that a general assumption of zero value for T&D avoided costs is inappropriate.

4.6. Natural Gas Issues

Price Forecast

SDG&E/SoCalGas suggest that the natural gas price forecast presented in the E3 report is flawed. SDG&E/SoCalGas proposed a different process based on an average of the basis differential for two years of basis swaps and three years of historical data. A basis differential between Henry Hub and the California Border/PG&E Citygate based on a five-year average would be comparable to the E3 approach to calculating the spark spread in the electric price forecast. SDG&E/SoCalGas suggest that an alternative would be to use the Henry Hub and San Juan Basin basis differential, or other appropriate basin, plus full transportation costs to the California Border/PG&E Citygate. SDG&E/SoCalGas note that the E3 method, the RPS Method, and the SDG&E gas price forecast included in its long-term resource plan are all consistent; all have some reference to gas forward prices and all rely on long-term gas price forecasts based on fundamentals prepared by government agencies or private consulting firms.

Storage

SDG&E/SoCalGas and TURN recommend that future updates to the gas avoided costs could include gas storage costs and core firm pipeline capacity costs. We agree in principal, and will consider SDG&E/SoCalGas' and TURN's suggestions in our consideration of revisions to the E3 report methodology in Phase 3, with the caveat that gas storage costs should not be included if the purpose of the project is the management of seasonal gas price swings. We also note that if core firm gas pipeline costs are already captured in the gas T&D adder, care must be taken to avoid double counting of that cost item.

4.7. Discount Rate

NRDC urges the PUC "to adopt a discount rate in the range of 2%-3% real. This is consistent with the 3% real discount rate that has been used for many years by both the CEC in evaluating energy efficiency standards and the Northwest Power Planning Council." (NRDC, pre-workshop comments, p. 4.)

The E3 Report uses the discount rate of 8.15% adopted in D.01-11-066 in the Energy Efficiency Policy Manual. We decline to adopt a specific discount rate for use here. The discount rate used should be determined in R.01-08-028, the energy efficiency rulemaking.

4.8. Conclusion

We are pleased with the degree of consensus that appears to have been reached regarding the E3 methodology. The utilities, while generally supportive of adoption of the E3 methodology for evaluating energy efficiency programs, each recommend slight modifications to the approach. On an overall basis, however, the E3 avoided cost methodology offers significant improvements compared to the existing methods. In contrast to the current approach and the production simulation method advocated by SCE, the E3 avoided cost methodology produces forecasts which are disaggregated by area and time for both electricity and natural gas over a 20-year period, from 2004 through 2023. For electricity, avoided costs are calculated by hour for each year for the 16 climate zones, 24 electric utility planning divisions, and three service voltage levels. This produces separate avoided cost estimates for customers served at each voltage level (transmission as well as primary and secondary distribution levels). For natural gas, E3 calculated the avoided costs by month for each year, utility, and customer type. The E3 avoided cost methodology more accurately estimates the electricity and natural gas a utility would avoid having to supply to its customers as a result of certain energy efficiency measures.

Compared to the current method, these hourly avoided costs will enable the Commission to recognize the full value of programs, such as air conditioner efficiency, which contribute disproportionately to reducing peak demand, especially on days of peak demand.

The avoided costs resulting from E3's methodology are also transparent and easily updated. In the existing hybrid market, an open and dynamic method is if critical importance. For example, while the avoided costs methodology incorporates a substantial reserve margin beyond what is currently maintained by the California Independent System Operator, the costing methodology can be easily modified to reflect any changes. For example, if a new standard requires additional capacity purchases beyond what is already included in the estimate, an adder could be included based on these additional costs. Alternatively, if these standards are implemented in the bilateral energy market, they can be reflected as a multiple of the long-run cost proxy, which is the cost of a combined-cycle plant.

Finally, the recommended methodology is relatively simple, transparent and relies on no proprietary data or software. PG&E agrees that E3's methodology is straightforward, transparent, and easily updated, and that the resulting avoided costs are acceptable for immediate evaluation of EE programs and ranking them for spending within given budgets. SDG&E/SoCalGas also stated that the E3 avoided cost values, albeit with certain modifications, are appropriate for use in determining the benefits of potential EE programs.³¹

We concur with the parties that we should continue to refine the E3 methodology and forecast as part of our effort to develop consistency in the methodology and input assumptions for Commission applications of avoided costs. However, this effort requires more time and effort than is available prior to the PY 2006 -008 program cycle. Our primary objective in adopting the E3 methodology on an interim basis is to promote cost-effective energy efficiency programs to assist the utilities in meeting the energy savings goals identified by the Commission in D.04-09-060. In order to achieve this goal we must update the avoided cost calculations used to evaluate competing EE programs prior to the time for program selection and design.

We believe that the parties' proposed modifications to the E3 methodology and forecast for other resource options should be carefully considered in the third phase of this rulemaking prior to adoption of the method on a permanent basis, and concur with the parties recommending immediate adoption of the approach for use in the generation avoided cost energy forecasts to be used in evaluation of energy efficiency programs for program year 2006. As PG&E and TURN point out, the existing energy efficiency avoided costs are so dated that the current E3 avoided costs, even without further refinement, represent an improvement needed now to avoid inefficient energy efficiency program planning for 2005 and 2006.

We have set ambitious goals for energy efficiency programs in the Energy Action Plan. Based on these goals, we directed the utilities to optimize electric energy efficiency investments in their resource plan portfolios. We increased energy efficiency funding to over $800 million for the 2004-2005 funding cycle, or an average of approximately $400 million per year and augmented natural gas energy efficiency funding for PG&E, SDG&E and SoCalGas on an expedited basis, in order to expand current programs for the 2004/2005 winter season.³² In addition, in D.04-09-060, we recently established increased natural gas and electric savings goals for the utilities by service territory through the year 2013, subject to updates for 2009 and beyond, and directed that the next program cycle would cover program years 2006 through 2008.³³ The next program cycle begins on January 1, 2006, and the program selection process, for which the updated avoided costs adopted in this decision are necessary, will begin in the next few months.

²¹ PG&E, post-workshop reply comments, p. 5.

²² SCE, PHC Statement, p. 2.

²³ SCE, Comments, p. A2.

²⁴ GPI, post-workshop reply comments, p. 6.

²⁵ The three EE measures are for secondary voltage customer in PG&E's Climate Zone 12 (the Central Valley area, including portions of the Diablo, Mission, North Bay, Sacramento, Stockton, Sierra and Yosemite Planning Divisions).

²⁶ PG&E agrees that CO2 emission values be used for 2008 and beyond in screening and selecting resources, but that emission reduction values for non-regulated pollutants not be included in any pricing for generation resources. (PG&E, comments, p. 10.)

²⁷ California Energy Commission, "Climate Change and its Impacts on California," July 2, 2004, www.energy.ca.gov/global_climate_change/index.html.

²⁸ SDG&E/SoCalGas, post-workshop comments, p. 13.

²⁹ CCC, pre-workshop comments, p. 10.

³⁰ SDG&E/SoCalGas, post-workshop comments, p. 9.

³¹ SDG&E/SoCalGas, post-workshop comments, p. 2.

³² See D.04-12-019.

³³ See D.04-09-060, mimeo., pp. 37; Ordering Paragraph 1.