SCE Portfolio Overview

SCE's proposed portfolio is based on a wide variety of programs for most sectors. Many of the programs are continuations and expansions of well-tested programs with established track records. Some programs will seek out innovative ideas for new opportunities, such as the InDEE and IDEEA, and Emerging Technology initiatives. In addition, SCE has developed three "Flagship" programs that attempt to find efficiencies in implementation by combining multiple previous programs under a few umbrellas. These are the Business Incentive Program, the Residential Energy Efficiency Rebates, and the Comprehensive HVAC program. Among them, these three large programs account for about one-third of the overall annual average budget.

Goals Attainment - SCE

Southern California Edison will be spending $674.8 million ($750.3 million including costs recovered from others and EM&V) over three years to save an incremental 3,364 GWh and 731 MW. No therms are included in the TRC. The three-year portfolio is forecast to have a TRC benefit/cost ratio of 2.76 and a PAC ratio of 3.58. This is a substantial programmatic effort at an average of $225 million/year, an increase in the annual budget of 246 percent from 2004-2005 ($91.5 million/yr), but is still forecast to be very cost-effective.

Comparison with CPUC Goals

For the three portfolio years, 2006-2008, the planned SCE energy savings can be expressed in two ways: with credit for pre-2006 activity, and without credit. If the pre-2006 is included, SCE is targeting 4,030 GWh (129% of goals), but without the pre-2006 activities, the goal is 3,364 GWh, or about 107 percent of the CPUC energy goals5. On the kW side, SCE forecasts 820 MW (120% of goals) with pre-2006 savings and 731 MW, or about 106 percent of the peak savings goals without the pre-2006 savings credit. These numbers reflect the total of the LIEE and EE savings for each year. It is our understanding that pre-2006 savings are not supposed to be used to compare to the goals, but that isn't the understanding of SCE. The MW savings are not the critical summer peak impacts but the cumulative GWh multiplied by the CEC factor of 0.21.

Comparison with Potential

As shown in Table 39, the expected savings from this program is forecast to exceed the three-year potential and the CPUC goals. At this time, there is no published report for industrial potentials, however, there is an industrial potentials study currently being finalized by KEMA. For the SCE industrial potential we used preliminary estimates from the soon-to-be-published 2005 industrial potentials study being completed by KEMA. The industrial potentials should be considered proxy estimates that will need to be adjusted once the KEMA study is released in 2005. The TecMarket Works Team acknowledges that these potential estimates will change over the course of KEMA's efforts to more fully develop the estimates.

KEMA's published potential reports provide 10-year estimates of sector potentials. In order to adjust the KEMA potentials to the 3-year 2006-2008 program cycle, we multiplied the KEMA potentials by 0.3. We use 3-year potentials in this assessment because the current program planning cycle is three years in length.

Budgets and Service Offerings Balance

SCE has a wide variety of program offerings with a reasonable split between residential and all other. There appears to be an effort to serve all customer segments, including manufactured home residents. The two largest three-year budgets are the Business Incentive Program at $106 million and the Residential Energy Efficiency Program with a budget of $67 million, although the multi-family sector will also be well-served with a budget of $53 million.

Energy Savings Issues

For all utilities, the TecMarket Works Team attempted to determine how reasonable the savings estimates were for the majority of measures in the overall portfolio. For those with a basis in the DEER database, we compared SCE's estimates to those in the DEER database. For the many measures that are not linked directly to DEER, we examined the workpapers that described how the calculations are done and upon what assumptions the estimates are based. For SCE, the vast majority of kWh and kW in the savings estimates were resulting from measures without a direct link to the DEER database (See Table 41).

DEER Measures Estimates

For SCE, only about 15 percent of the kWh savings and 27 percent of the demand savings of programs could be traced back to a DEER-based energy savings estimate.

All of the measures that were estimated using DEER were reviewed for accuracy and consistency with the DEER 2005 Database. The DEER data was downloaded from http://www.rtf.nwppc.org/deer2005/# on May 13, 2005.

In the "measure list" tables of the utility workbooks, the "Measure ID" was typically a match with the "RUNID" field in the DEER data. In some cases, averages across the DEER "Measure ID" were used instead of the more specific "RUN ID". Comparisons of the utility estimate and DEER estimate were made based on the information available in the filed workbooks. For example, if the utility noted that they used an average for all DEER refrigerators, the evaluation team attempted to replicate their calculations for a valid comparison.

SCE's per unit estimates matched with those in the DEER database for the largest proportion of the kWh savings. For those that were over the DEER value, the main problem was that SCE was using a baseline that was below the current code regulations. DEER provides estimates above code for these measures, and it is not clear why that value was not used in these cases considering that it was used in other cases. Some measures also had differences in the measure units being used in DEER and SCE. Most of these measures were for building shell improvements like insulation, where "square feet" or "whole house" were used as the base unit and causes some comparison problems.

Overall SCE's savings estimate appears to be slightly more optimistic than would have been generated with DEER. Making certain to use above code estimates from DEER would help rectify the discrepancy.

The largest proportion of SCE's kW savings was not clearly related to the DEER database. DEER estimates appear to be smaller by a factor of 1000, which suggests there may have been a mistake in the conversion in units from DEER in addition to using a value different from DEER. DEER reports demand savings in watts, not kW. Again, SCE should have used the above code estimates provided in DEER and it is not clear why they did not do this for a select group of measures.

Non-DEER Measure Estimates

SCE provided many workpapers to support their savings estimates. Based on our review of the measures making up the majority of the non-DEER related savings, about 7 percent appeared reasonable based on our review, and 20 percent of the portfolio savings presented specific concerns (See Error! Reference source not found.). For more detail about the specific concerns for these measures, see Appendix A. Based on concerns about these measures and lack of documentation for other measures, SCE was asked to provide further documentation of the measure savings estimates. Several of the measures for which additional documentation was provided were those in customized industrial and commercial applications. It appears that the documentation provided will clarify the forecasting assumptions, but there may not be enough information to comment on specific measure savings estimates, but there was insufficient time for complete review. These measures made up nearly 29 percent of the portfolio's savings. Likewise, there were several lighting measures for which additional documentation was provided, and it appears they will clarify the assumptions used in the measure savings estimates, but the evaluation team did not have sufficient time to review each measure. These lighting measures made up about 20 percent of the portfolio savings.

Cost-Effectiveness - SCE

TRC Issues

SCE is forecasting that only three programs will not be cost-effective on a TRC basis. Several have unexpectedly high TRCs that may result in unfulfilled promises.

With a TRC of 0.43, the CA New Homes Program appears particularly expensive. We have suggested an approach in our Portfolio Overview and in the policy section of this SCE review that the CA New Homes program could be legitimately combined with Codes and Standards and other programs to create a strategic and cost-effective ensemble. Similarly, the Home Energy Efficiency Survey program is treated in some non-California jurisdictions as part of the marketing effort for very cost-effective Residential Energy Efficiency Programs, with a good combined cost-effectiveness. The savings that come from the rebate programs are seen as a result of the work of all of the elements that combine to bring consumers to make the efficiency transaction, most particularly the information and audit programs. Thus the cost of those programs is included in calculating the cost-effectiveness of the overall rebate program.

TRC Range-of-Estimate Issues

The TRC values range from 0.43 to 6.57. There are at least three TRCs that would appear to be potentially too high, so there is risk of under-achievement. The first is the Home Energy Efficiency Survey that approaches a TRC of 1.0, based on results from the 2002 evaluation of PY 2000 participants in similar programs. Although several national studies have been done that indicate savings can occur from these types of informational audits, the basis for these claimed savings in the HEES program is undermined by the fact that the evaluation did not include a nonparticipants group to control for actions that may have been taken during the start of the energy crisis, which was going on concurrently with the program (Ridge and Associates, 9/2000, Study SCE-116.01). In addition, among the assumptions that were made, the actions taken after the audit were due to the audit, even though 40% of respondents reported having participated in other efficiency programs. Thus, the evaluated TRC may end up disappointing all parties when the results come in from the projected program.

The Appliance Recycling Program (projected TRC of 6.57) is a mature and well-evaluated program that may anticipate actual savings higher than will be achieved. Not only are the number of secondary appliances expected to be picked up increasing over the three years, but the program has been in the market a long time, potentially restricting the amount of net growth that is possible. The expected savings are what would be anticipated from refrigerators that were at least 27 years old according to manufacturers' ratings. While it is true that the seals and fluids of younger refrigerators could result in lower efficiency when tested in the evaluations, they will need to be tested in the next evaluation with actual content-thermal mass to represent actual real world consumption. Just as the actual equipment condition is metered, not just using the test standard, so should the usage with thermal mass be used, since these appliances were supposedly in use in order to be eligible for the program. In addition, the NTG is assumed to be 0.35 across all three years of the program, yet the history of the program has shown a clear downward trajectory of the NTG, which might indicate that the out years could be lower still. Finally, the proposed program design does not require that the refrigerators be manufactured before 1992 in order to be picked up, when this has been the rule for the last two years. This may result in the mix of measures included having a higher starting efficiency than program forecasts.

The other program where the forecast TRC seems to be out of an expected range is the Small Business Direct Install program. It has a TRC forecast of 5.42 for a program delivery approach that has been much less cost-effective in previously evaluated programs. These types of programs are generally very expensive, and the evaluated similar programs in CA cost 100 percent more per first year kWh than is forecast in this program -- the previous evaluations of the third-party small business direct install programs in CA showed that the programs cost $0.22/kWh and $0.25/kWh. This proposal has an expectation of $0.10/kWh. While improved program efficiencies could be expected, doubling the program effectiveness seems to be a risky projection.

Energy Star Refrigerators

There should be serious consideration given to dropping incentives for Energy Star refrigerators, because they already have 42% market share and are expected to continue increasing market share over the three years of the program. The measure saves only 58 kWh a year for a $50 utility rebate; an Energy Star clothes washer above an MEF of 1.80 would save about 70 kWh plus significant water savings, but that measure is not being rebated by the REEP program. It would seem that the $11.5 million planned for refrigerator rebates would better be used for peak saving measures like residential HVAC.

Issues Addressed - SCE

Administrative Costs

At 8.5 percent, the SCE administrative costs are moderate, and probably low for most definitions of administrative costs. If the CPUC-ED staff clarifies the contents and definitions of such costs, a clearer picture will probably emerge when we compare utilities. The administrative costs vary across programs with some of the larger ones, such as the Residential Energy Efficiency Incentives Program having low costs - presumably due to some economies of scale. It is also possible that some programs that are turn-key, such as the Appliance Recycling program, have low internal utility administrative costs, but higher overall societal administrative costs.

Net to Gross

We noted in the overall statewide portfolio that while many of the NTGs found in the Policy Manual are fairly outdated at this time, the utilities did use the values in the Policy Manual as required. The majority of measures appear to have reasonable NTGs. While they are all based on the Policy Manual, there are some measures that probably have incorrectly high expectations for NTG (e.g., premium efficiency motors in industrial rebates, CFLs in residential homes and commercial lighting measures in the Business Incentive Program - all at 0.96). In addition, since the program is being planned for 3 years, the NTG is likely to be different by the third year as the market accepts more and more efficiency measures. Some measures for which trends are developing are appliance recycling, CFLs, Energy Star refrigerators, dishwashers, and air conditioners.

Flagship Programs versus Other Programs

While the BIP and REEP programs are the largest programs, and represent the Flagship Programs for the non-residential and residential programs respectively, there are still many other diverse program offerings that provide services. This diversity lowers the risk associated with concentrating program expectations in a single delivery mechanism.

Energy Accounting Issues

As noted in the Portfolio Overview, sector-specific programs are referring consumers to the Flagship rebate programs, often from more than one sector. At the same time, they are providing audit and custom incentives at the facilities. The accounting for actual achievements and the ability to match up participants in different programs for evaluations will be a chronic problem without some innovative approaches to tracking built-in up front. Double counting is also a potential issue that cannot even be investigated without an appropriate tracking system.

Transparency of Data Issues

Energy Savings

As noted above, we did not have sufficient information to track the bases for the savings for the measures in this portfolio.

Risk Issues

Energy Savings

At the utility level, the risk of not accomplishing the savings that are forecast is always there, but it is a relatively lower risk with a diversified portfolio. At the program level, the expansion of the Residential Rebate program with doubled savings, but tripled costs is one program that is large and could get out of control, and the IDEEA and InDEE programs are forecasting a substantial amount of savings (112GWh at a cost of $39 million) without knowing what new and innovative technologies will be proposed. The optimistic certainty of savings is further emphasized by SCE's forecasting TRCs for the yet-unknown programs of 4.23 and 3.84. Both of these TRCs are higher than just about any other program proposed by the other three California utilities. The other potential risks are noted in the Program Level Summary Table (Table 48).

Oversight Risk

SCE proposes flexibility in shifting funds that allows it to shift both procurement and PGC funds with a great deal of independence. SCE proposes fund shifting guidelines according to the following table.

This proposal provides for flexibility for the program administrator, but only provides after the fact notification to the CPUC for most cases. One clear distinction in the SCE proposal from other utilities is that it allows reductions in the evaluation budget of up to 25% without CPUC approval. Evaluation has often been an area that has suffered from program administrator priorities. This should remain as protected as the Emerging Technologies, Codes and Standards, and marketing and outreach programs.

· For Emerging Technologies, Codes and Standards, Statewide Marketing and Outreach, but not EM&V, pre-Commission approval is required before funds are shifted out of these programs.

· For fund shifts among categories exceeding the 25% limitation, pre-Commission approval is required before funds are shifted.

It is clear that some flexibility is needed to allow utilities to respond swiftly to changes in the markets, but allowing too much flexibility becomes a "faith based" oversight. If the CPUC believes that the 20% is the right amount for bidding, that administrative costs need to be controlled, that programs need to focus on all sectors and provide both kW and kWh benefits, and that sufficient funding is available for evaluations, then some more oversight than "notification" may be required. Otherwise, the current proposals and all of the guidelines behind them become immediately fungible. In fact the Energy Division review of the portfolio balance will have little meaning if it can change dramatically without prior authorization.

We have suggested in other parts of this review that a strategic reserve be accumulated from many programs upfront, which can be used to provide immediate flexibility without having to de-fund programs, take away money from third party implementers, and meet emerging needs. The TecMarket Team believes that flexibility to change incentives when the market moves is very sensible, and shifting uncommitted funds among programs within a sector makes sense, but situations that allow shifting of incentive dollars to increased administration may require prior approval. Not everything is an emergency.

Delivery Risk

Almost all programs will be operated at a much higher level of activity than in the past in order to meet the new goals. The program descriptions often reflect confidence that the proven programs will be improved, but there is little discussion of how the utility will staff up and train all the new actors in the fairly short time needed to ramp to the higher levels, often with new program designs and relationships in the market - such as the Comprehensive HVAC programs. In addition, the large expansion of the retro-commissioning program not only involves the risk of actually getting the same level of savings once the program gets beyond the lower hanging fruit, but entails the added risk that the utility will not be able to get the market penetration among building owners to reach the implied square footage needed to make the targets.

New Implementers

Several programs will have new implementers without a record of working in the program designs in which they are involved. For example, installation contractors with turn-key operations in the Small Business Direct Install Program will work through local governments, Community Based Organizations and Faith Based Organizations. This could be a risk and savings issue if CBOs and FBOs are expected to be re-trained to provide services outside of their areas of experience.

New Program Characteristics

Some programs will have some risks associated with completely new ways to approach the market. This may be a problem for the Comprehensive HVAC programs, for example. In order to achieve the large savings from the programs, solid coordination and new sets of relationships with market channel participants will be needed in a hurry. Further, by providing HVAC incentives upstream, the utilities won't have their usual role in overseeing the installations, and (1) may allow many homes to change their HVAC system without fulfilling the code-requirements to seal ducts at the same time, or (2) may actually end up paying for code-required duct-sealing under another program. This will take some careful oversight.

New means of doing business with the utility will require change on the part of market actors also. New approaches to the market can be effective in the long run, but there are short-term risks for achieving the goals.

Comprehensiveness and Lost Opportunities - SCE

SCE has a very comprehensive and diverse program portfolio. After careful review, we only noted a few areas of potentially lost opportunities. These included the potential for Energy Star Clothes Washers to fall through the cracks if SCE expects SCG to take care of the measure (and they budget for very few rebates), as well as the lack of an efficient manufactured home new construction program. As noted in the statewide summary across all utilities, there is no evidence that the utilities are taking advantage of the large efficiency opportunity to replace high intensity discharge (HID) lighting with high performance T-8s and T-5s in grocery, warehouse, large retail, and other places where a wattage reduction can be almost half of the installed wattage and the related additional benefits of dimming and the ability to work with occupancy sensors open up a lot of other savings opportunities.

Bidding and Third-Party Issues - SCE

There is little information provided, although estimates of the expected savings and aggressive benefit/cost ratios are provided for some programs to fill out the goals and the budgets. Since some of the bid programs are asking for ideas and innovations that haven't already been adopted by the CA utilities over 15 years of programs, the plan that expects very highly cost-effective programs in the next three years seems extremely optimistic.

Partnership Program - SCE

There is approximately $45 million budgeted by SCE over three years for various Partnerships. Although there are descriptions of what each non-utility partner brings to the partnership, the bases for the expected savings and the very strong TRC (3.15) appear to be speculative in many cases, with some optimism based on prior successes with the individual partners, e.g., the California university system.

Policy Issues - SCE

Residential New Construction

SCE is planning a fairly robust new home construction program to follow the Statewide Energy Star New Homes Program. From observing the Public Advisory Group (PAG) process, it appears that there is a strong interest in having Residential New Construction programs at the utilities. Given that the Residential New Construction programs are not easily cost-effective, especially after new code adoption cycles, the Commission should consider providing policy guidance as to the continuation or focus of this effort and the level of funding within the portfolio that is appropriate.

Further, we suggest that these programs could be integrated with other programs, such as the Emerging Technologies Program, Codes and Standards Program, Sustainability programs and the Advanced Building Programs of other utilities, in order to establish a strategic initiative that is specifically designed (a) to provide cost-effective long-term savings through adding innovations (b) through a large building practice dissemination program, and (c) eventually through code changes. In that way the ensemble of programs is strategically designed and would meet the criteria of being operated to produce predictable, long-term, cost-effective savings.

This approach does requires that codes and standards be recognized by the CPUC as being positively influenced by utilities and crediting the utility programs with some part of the resulting large and cost-effective portfolio savings.

Non-Residential New Construction

The natural corollary of this would be to combine the Savings by Design, Emerging Technologies, Sustainable Communities, and Codes and Standards into a non-residential new construction market. (The latter three programs may serve both residential and non-residential portfolios, but it is easier to separate costs in accounting than to divvy up savings among programs in the same market as now occurs). As with the new homes program, there is a need for a "carrier" program in non-residential new construction to disseminate innovations into the market, so that it can be shown to be cost-effective and eventually become improved code. Because all of these programs address the same market actors and are targeted to the same goal of improved building energy efficiency, they should be designed, implemented, evaluated, and rewarded as a unified program. The policy alternative to develop a unified non-residential new construction market as a sub-portfolio may be an overlooked opportunity.

Evaluation Issues

The CPUC will be responsible for designing and implementing impact evaluations for the 2006-2008 portfolios. Without having direct responsibility for the design of the program and the data tracking, the staff may find that the necessary and useful data may not be accessible due to the design of the Flagship Programs. The REEIP rebate program and the upstream design of the Comprehensive HVAC initiatives are two examples where consumers may be getting services and rebates from more than one program. Knowing who the participant is and being able to follow measures to meters will be important for evaluating the program without double-counting. Examples include HVAC program participation and Business Incentive Program participation. The consumer may not even be aware of the upstream rebate behind measures. Net to gross estimation will be difficult as it is with any upstream program.

In addition, the same consumer gets support from two programs with measures that may interact, as in lighting and HVAC. The HVAC program is designed to work with the CA New Homes Program, but the impact of the measure can be double-counted. The point of purchase rebates are readily available to all consumers and can show up everywhere in the buildings of other programs. This isn't new, but from arm's length, the CPUC staff may find it difficult to sort out the issues that are within the utilities' tracking systems.

The recommendation is that the CPUC staff and the utilities start planning now for how the participants in each program will be tracked and associated with a premise or meter. Develop a process now to ensure that the eventual evaluators will have access to meaningful data needed to do the job, and that responsible parties within each utility are identified.

Finally, the CPUC staff needs to begin now to prioritize the evaluation research that will need to be done, given the large amount of savings that are promised from non-DEER measures and new program designs. Just as the utilities need some flexibility to take advantage of savings opportunities that arise, the CPUC needs to have agreements with the utilities whereby the evaluation results found by the CPUC and its contractors can be used to modify, expand or contract programs that are within what is still a speculative portfolio.

As part of the evaluation effort, we also recommend the following:

1. Development of a data dictionary that all users can access for information on definitions of measures, baselines, energy savings, costs, and references for non-DEER measures, including changes when they are made to the baseline or the measure effectiveness.

2. Periodic updating of the potential studies in all sectors (residential, commercial, industrial, and agricultural).

3. Market-based evaluations to see how specific markets are changing, only some of which may be due to the recorded rebates in the IOU programs. The potential studies should be updated as a large part of market based evaluations.

Conclusion

Our general conclusion is that SCE has a strong and diverse appearing portfolio with only a moderate risk of failing to achieve the projected savings. We also have included some suggestions for improvement in the policies and in the programs. These include:

· Grouping the new residential and non-residential construction programs into market-based packages of programs

· Questioning whether the $11.5 million being spent on refrigerator rebates might better be spent on a more aggressive residential HVAC program

· Adding measures to replace HID lighting with T-8s and T-5s

· Recommending a new manufactured home construction program

· Removing measures that are required by code as listed in the Statewide Portfolio Overview

· Updating gross savings for the Appliance Recycling program when in situ test results become available

· Recalculate TRC if necessary to comform to the CPUC Policy Manual, Version 3, as discussed in the Statewide Overview

· Adjust expected TRCs and savings after receiving the bids on the IDEEA and InDEE programs

· Update savings and forecasts against goals to reflect the correct values for CA New Homes, all instances where watts were taken to be kW in estimating savings, the removal of measures that only get to code, and any other errata that have become apparent during the review process.

Our general endorsement of what is being proposed is a "faith based assessment" that cannot be validated until we are better able to trace and understand the derivations of the non-DEER savings estimates, and until the programs are actually implemented.

Program-Level Assessment - SCE

This section of the report presents the program-specific assessment information and issue discussions that were identified during the portfolio review effort. The issues reviewed are presented in the left-most column of Table 48, and each subsequent column represents a specific program, allowing the reader to see if the review team determined there to be an issue associated with a specific program, and to understand the review team's perspectives associated with each issue.

5 This is further complicated by the fact that the filed Summary Table (Attachment II - Table 2.1) says 3,549 GWh, including Low Income EE.

Previous PageTop Of PageNext PageGo To First Page