9. Discussion of Remaining Contested Issues
TURN contests aspects of the settlement agreement business case, and maintains that if its suggested modifications are made the business case would not be cost effective. In addition, TURN opposes several other elements of SCE's AMI deployment and funding proposal, and recommends adoption of a penalty mechanism if the AMI deployment project is approved and does not result in at least 65% of the expected demand response benefits. In order to find these aspects of the settlement agreement reasonable, it is necessary to analyze TURN's specific objections to these terms, including analyzing the settlement agreement assumptions and the alternatives recommended by TURN. TURN's proposed penalty mechanism is discussed in Section 16, below.
9.1. Cost Effectiveness of the Business Case
In this section, we address the contested assumptions in SCE's business case, and analyze TURN's arguments for modifying those assumptions.
9.1.1. Use of the Cost-Effectiveness Framework Proposed by Parties to R.07-01-041
In the business case contained in SCE's initial cost-effectiveness testimony and in that contained in the settlement agreement, SCE uses a business case analysis framework similar to that proposed in the July 21, 2004 ruling in R.02-06-001, and used in the analysis of the AMI proposals of PG&E and SDG&E. TURN argues that instead, the proper framework for the AMI project cost effectiveness analysis is the "consensus framework" for cost effectiveness proposed by parties in R.07-01-041.24 TURN further argues that if the proposed Total Resource Cost (TRC) test is used to evaluate SCE's business case, incentives paid to customers would be included as a cost in the analysis, and these additional costs would make the overall case not cost effective.
The "consensus framework" is not the appropriate tool to use in assessing the cost effectiveness of SCE's AMI project for several reasons. First, this proposed framework has not been adopted by the Commission for use in evaluating DR programs, and in fact had not yet been proposed when SCE filed its initial application. At least one additional proposed cost-effectiveness methodology is still being considered in the cost-effectiveness phase of R.07-01-041, and no methodology for assessing the cost effectiveness of demand response programs has yet been adopted. In addition, if a cost effectiveness methodology for demand response programs had been adopted, it is not clear that such a methodology would be appropriate for estimating the cost effectiveness of an AMI system, which includes additional metering equipment and associated functionality. For these reasons, SCE cannot be expected to have applied this framework to its cost-effectiveness analysis. As discussed above, the Commission established in R.02-06-001 a specific framework for analyzing AMI proposals, and SCE properly applied the analytical framework also used in the SDG&E and PG&E AMI cases. TURN's objection is not persuasive, and the analytical framework used in the settlement agreement is reasonable and consistent with law and Commission policy.
TURN contests several of the assumptions used to calculate the costs and benefits of SCE's proposed PCT program included in the business case, and specifically contends that the settlement agreement underestimates the costs of PCTs and overestimates participation in the PCT program as well as the benefits that can be expected from PCTs. Specifically, TURN argues the following:
1. The business case underestimates the costs of the PCTs SCE will use in the program, resulting in the use of an erroneously low estimate of program costs in the business case.
2. By assuming a statewide mandate of the use of PCTs from the CEC by 2012 and a high enrollment percentage in PCT programs, the settlement agreement overestimates participation in the proposed PCT program and therefore the benefits that may accrue from the program.
3. SCE further overestimates the expected savings from participation in the proposed PCT program by failing to reduce the kilowatt-hours saved by the program to account for estimates of average air conditioning tonnage, inoperative air conditioning (A/C) units, and customer overrides.
TURN contends that the $50 cost to purchase a PCT that SCE includes in its business case underestimates the actual cost of the PCTs SCE will need to utilize the functionality of its proposed AMI system. The SCE/DRA settlement provides that $50 is a reasonable cost for PCTs based on the product design specifications and evidence presented of the costs of PCT component parts, and budgets an additional $75 per unit for installation. SCE provided a bill of materials for most of the PCT components,25 along with separate estimates of the additional parts needed to complete a PCT; SCE also notes that it intends to buy PCTs in bulk at wholesale rates. TURN argues that PCTs whether bought retail or wholesale, would cost more than $50 per unit, and more than $125 installed, and believes that SCE underestimates the actual cost of this program.26 Based on the evidence provided by SCE, the $50 per unit wholesale cost estimate for a PCT is reasonable. TURN has not provided compelling evidence that the wholesale cost of a PCT purchased in bulk will be higher than SCE's estimate.
TURN argues that two assumptions made in the settlement agreement business case reflect unrealistically high estimates of customer participation in PCT programs enabled by AMI, and recommends that these assumptions be modified to reduce estimates of customer participation. First, TURN contends that it is speculative to assume that the state will adopt a mandate for PCTs in new construction by 2012,27 and that without a mandate, SCE's participation estimate is unrealistically high.28 In addition, TURN argues that SCE's assumption that 25% of customers with PCTs installed under a Title 24 mandate will enroll in the PCT program is overly optimistic.29
SCE, on the other hand, states that its participation estimates do not depend on a CEC PCT mandate in 2012. SCE addresses TURN's concerns by including the cost of PCTs for its PCT program in its business case, so achieving the benefits of the PCT program does not depend on the possible state mandate.30 SCE and DRA agree that the enrollment rate of customers offered a free PCT is reasonable.31 While there is some uncertainty in this assumption, as in many assumptions, the settlement agreement business case estimates of PCT penetration and participation are reasonable.
TURN further claims that SCE overestimates the expected savings from participation in the proposed PCT program by failing to reduce the kilowatt-hours saved by the program to account for estimates of average A/C tonnage, inoperative A/C units, and customer overrides. TURN advocates for reducing the expected savings of 1 kW of demand response for four hours to reflect the following adjustments:
· Derate by 1% to adjust for average air conditioning tonnage,
· Reduce forecast DR by 15% to account for already-inoperative air conditioning units and
· Reduce forecast by 10% to account for customer overrides.
SCE argues that if it had used TURN's adjustment for A/C tonnage, the change would actually increase its expected PCT load reduction by 32%, not lower it by 1%,32 and shows that the effects of customer overrides and inoperative units are already accounted for in its model.33 Based on the evidence provided by SCE, the settlement assumptions appear to be reasonable, are within the range of value we would expect from a litigated outcome, and should be adopted for the purposes of the business case.
9.2. Peak-Time Rebate Benefits Assumptions
TURN takes the position that the settlement agreement business case overstates the likely benefits of a Peak-Time Rebate (PTR) program, and the assumptions underlying the analysis of PTR should be adjusted to reflect lower expected benefits. Specifically, TURN does not believe that the 50% awareness rate of the PTR program that is used in the calculation of program benefits is reasonable, and suggests that it should be reduced to a 25% participation rate, to reflect that not all customers aware of PTR events are likely to reduce load in response to a program event. In support of its position, TURN calls into question the applicability of participation rates from the State Pilot Program (SPP) to a broader mandatory PTR program, such as is used in the business case. TURN posits that the SPP population differs from the general population of the state because they have already shown a willingness to voluntarily participate in demand response programs, and are therefore more likely to participate than typical customers.
SCE counters by noting that analysis of the earlier SPP study looked for possible participation bias and found none, so concern over possible participation bias is not a sufficient reason for reducing the awareness and participation rates of customers enrolled in a PTR program, despite the differences between the SPP and general customer populations. Furthermore, the 50% awareness rate is comparable to the rate accepted by the Commission in evaluating the settlement of SDG&E's AMI business case. While approval of a settlement does not set a precedent nor imply a Commission endorsement of specific settlement elements, the same reasoning used in the comparable SDG&E case to accept this assumption is relevant here. In accepting this number, D.07-04-043, p. 51 states, "use of a conservative figure is prudent in order to allow for potentially improved results using different rate designs and/or improved communications and technologies in the future.''
In addition, TURN believes that it is inappropriate to use the customer elasticity of demand results from the SPP in evaluating the PTR, because the SPP study did not test a PTR rate. Instead, SPP customers were subject to a Critical Peak Pricing (CPP) rate, which TURN expects would be more effective in encouraging customers to lower their demand during an event then an equivalent PTR rate (i.e., a PTR with an incentive for load reduction equivalent to the penalty for increased use under a CPP rate). TURN argues that while the incentive for saving a kilowatthour (kWh) under the PTR rate is the same as under an equivalent CPP rate, that the cost of consuming an extra kWh is different under the two programs, and significantly higher under a CPP rate than under a PTR rate. TURN posits that "customers respond more to the threat of high prices than to the temptation of rebates,"34 and that customers under a CPP rate will be charged a high rate for every kWh used during a CPP event, while PTR participants will only receive incentives on the kWh saved during the same period, leading to a much greater savings from an CPP rate than a PTR, and large difference in bills between participants of the different programs. TURN argues that the PTR elasticities are more likely to be comparable to elasticities measured in a different pilot conducted in Ontario by Hydro Ottawa. Based on the results of this different pilot, TURN recommends adjusting the elasticities calculated from SPP data by 30% to account for these program differences.
SCE counters that TURN's position has no empirical basis, and provides analyses that the demand elasticities computed for CPP under the SPP are "statistically indistinguishable" from the elasticities of PTR pilots conducted elsewhere.35 The fact that two numbers are statistically indistinguishable does not necessarily mean that the underlying calculations used to arrive at those numbers will always lead to the same outcome. However, TURN does not provide a solid basis for its suggested reduction; as TURN acknowledges, when the margins of error of the two studies are considered, the apparent 30% difference between the elasticities calculated from the different pilots is not statistically significant.36 The best available comparison between the elasticities of the CPP and PTR rates do provide some reasonable confidence that at the levels studied, the two rates lead to similar outcomes.
In addition, TURN bases its argument that load reductions from PTR will go down over time on the assumption that customers can drop only a small portion of their load, and therefore will receive small savings from a PTR rate. TURN believes that a combination of a lack of a penalty for increased use combined with a low absolute level of savings available under a PTR rate would over time discourage customers from reducing usage under a PTR rate. TURN argues that this will lead to customer fatigue and a reduction in long-run elasticities of demand that are lower than the short-run elasticities. SCE counters that customers responding to a PTR rate would likely be able to reduce their load by 50% or more, leading to a much greater savings than TURN predicts, and also cites economic theory, under which long-run elasticities are generally higher than short-run elasticities.37
Current evidence does not provide a definite picture of customer behavior under a PTR rate, since such rates are not currently in widespread use. However, based on existing evidence it is reasonable to conclude that the elasticity of customer electric demand under a PTR rate may be comparable to under a CPP rate. Similarly, though it is not possible to be certain how customers will react to a PTR rate on a long-term basis, it is reasonable to apply economic theory to this question and assume that long-run elasticities will not be lower than short-run elasticities. Over the long run, for example, customers may have access to more enabling technology allowing them to respond more easily to PTR rates and increase their resulting demand response. For these reasons, the elasticities used in the settlement agreement business case, which are based on elasticities calculated from CPP rates and are assumed to remain stable over time, are reasonable for the purposes of estimating future energy savings from PTR rates and their associated benefits.
9.3. Transmission and Distribution Costs
SCE forecasts that its AMI project will provide approximately $221 million in benefits for deferred or avoided transmission and distribution (T&D) costs. TURN argues that in order to avoid or defer T&D investments through demand response, a utility must show that the specific demand programs meet "right place" and "right certainty" criteria identified in the proposed cost effectiveness Consensus Framework introduced by parties in R.07-01-041. TURN further requests that if the Commission includes T&D benefits in the business case analysis, that it adjust the T&D amounts to remove impacts from its Air Conditioner Cycling Program (ACCP). TURN argues that SCE has already used its ACCP to mitigate some local and regional emergencies in the absence of an AMI system, making it inappropriate to attribute associated T&D benefits to the AMI system. Finally, TURN states that even if some T&D costs are included in the business case, that only the costs of some limited transformer costs should be included, because other T&D costs are not allowed under the Consensus Framework proposed in R.07-01-041.38
SCE counters that its estimate of a 20% benefit for T&D is reasonable and its simplified assumptions for calculating the T&D benefit are consistent with the Commission's adopted business case framework. SCE also notes that TURN does not demonstrate that use of the consensus framework would provide results materially different from those used by SCE.39
As discussed in Section 9.1.1 above, the consensus framework for cost effectiveness analysis proposed in R.07-01-041 is not the appropriate tool to use in assessing the cost effectiveness of SCE's AMI project. SCE properly applied the analytical framework established for AMI proposals in R.02-06-001, and that has been used in the SDG&E and PG&E AMI cases. Section 9.1.2 accepts the reasonableness of including the settlement agreement's PCT benefits in the business case. Therefore, it is not appropriate to exclude T&D benefits associated with the PCT cost. TURN's suggestion that benefits related to the ACCP program are not appropriately included because the ACCP program has been dispatched in local areas in the absence of an AMI system does not account for the likelihood that the AMI system will enable SCE to dispatch the ACCP program more frequency and in more targeted areas, which may significantly increase the benefits of that program and increase its T&D benefits. The level of estimated T&D benefits used in the settlement business case is reasonable in the context of the whole record.40 TURN's objection is not persuasive, and the analytical framework and estimated values used in the settlement agreement are reasonable and consistent with law and Commission policy.
9.4. Societal Benefits
The settlement agreement identifies benefits beyond the financial benefits included in the business case, and quantifies these benefits at approximately $295 million. The societal elements that provide net benefits in the settlement agreement are the costs and benefits related to unaccounted-for energy and energy theft, which are estimated to result in a net benefit of $39 million, and benefits associated with increased meter accuracy, which the settlement agreement estimates at approximately $256 million. The settlement agreement does not include these additional benefits in the business case, but SCE and DRA agree that these benefits are real and can now be quantified. TURN argues that both of these proposed societal benefits are not sufficiently supported in the record. TURN argues that AMI technology does not by itself reduce energy theft and that the inclusion of this benefit contradicts SCE's own previous statements about AMI and energy theft.41 TURN also notes the relative absence of information in the record supporting the $256 million estimate for increased meter accuracy.
Both of these societal benefits are included in the settlement agreement, and while it would be helpful to have more information in the record on these specific issues, parties had an opportunity to question witnesses about the derivation of these benefits during hearings on the settlement agreement. SCE addresses energy theft in its rebuttal testimony, estimating savings from energy theft reduction at between $96 million and $150 million, in contrast to DRA's estimate of $26 million.42 The settlement agreement estimates the value of this societal benefit at $39 million, which is in the range defined by the testimony, and is reasonable in light of the record. Though there is little information in the record on meter accuracy benefits, TURN does acknowledge that in principle solid state meters may have accuracy benefits over older electromechanical meters,43 and the calculation of the estimated benefit for meter accuracy appears consistent with the Commission's acceptance of similar benefits in the SDG&E AMI case.44 In addition, even if, as TURN suggests, the meter accuracy benefits turn out to be minimal, estimated societal benefits of $39 million from energy theft benefits would be enough to support the cost effectiveness of the settlement.
The settlement agreement as a whole appears to be reasonable in the context of the overall record, and we are inclined to consider these societal benefits (which are not considered a part of the main business case) to be reasonable estimates of possible future benefits of AMI that have until recently been difficult to quantify. This is consistent with D.07-04-043 in SDG&E's AMI deployment application, which recognized some level of benefits of SDG&E's AMI system for both meter accuracy and energy theft.45 We will consider these societal benefits in our analysis of the settlement as a whole.
24 Joint Commments of California Large Energy Consumers Association, Comverge, Inc., Division of Ratepayer Advocates, Energyconnect, Inc., Enernoc, Inc., Ice Energy, Inc., Pacific Gas and Electric Company (U 39-M), San Diego Gas & Electric Company (U 902-E), Southern California Edison Company (U 338-E) and The Utility Reform Network Recommending a Demand Response Cost Effectiveness Evaluation Framework, filed November 19, 2007, in R.07-01-041 Phase 1.
25 DeMartini, Exhibit 14.
26 Schilberg, Exhibit 203, p. 23.
27 Opening Brief of The Utility Reform Network Concerning Southern California Edison Company's Application for Approval of Advanced Metering Infrastructure Deployment Activities and Cost Recovery Mechanism (TURN Opening Brief), p. 30.
28 TURN Opening Brief, pp. 31-32.
29 TURN Opening Brief, p. 32.
30 Reply Brief of Southern California Edison Company, p. 29.
31 SCE Exhibit 3 pp. 7-8, pp. 16-17.
32 SCE Opening Brief, p. 46.
33 SCE Reply Brief, p. 28.
34 TURN Opening Brief, p. 15.
35 SCE Opening Brief, p. 38.
36 RT pp. 51-52.
37 SCE Reply Brief, p. 21.
38 TURN Opening Brief, p. 33.
39 SCE Reply Brief, pp. 30-32.
40 SCE Exhibit 4, p. 27.
41 TURN Opening Brief, pp. 42-43.
42 SCE Exhibit 3, pp. 42-43.
43 TURN Reply Brief, p. 42.
44 D.07-04-043, pp. 38-39.
45 D.07-04-043, pp. 38-39.
