PREPARED TESTIMONY OF ERIC BORDEN. Submitted on Behalf of THE UTILITY REFORM NETWORK. 785 Market Street, Suite 1400 San Francisco, CA 94103

Transcription

1 CPUC Docket: A , et al. Exhibit Number: TURN - Witness: Eric Borden Date: August 1, 01 PREPARED TESTIMONY OF ERIC BORDEN ADDRESSING THE PROPOSALS OF PACIFIC GAS AND ELECTRIC COMPANY AND SOUTHERN CALIFORNIA EDISON FOR NON-LIGHT-DUTY CHARGING STATION INFRASTRUCTURE PROGRAMS Submitted on Behalf of THE UTILITY REFORM NETWORK Market Street, Suite 0 San Francisco, CA Telephone: (1) - Facsimile: (1) -

2 Table of Contents I. Overview of PG&E and SCE Non-Light-Duty Infrastructure Proposals and TURN s Recommendations... 1 II. PG&E s and SCE s Program Design and Cost Estimates Are Broad and As Proposed Will Not Lead to Effective Investments that Maximize Benefits and Minimize Costs... a. PG&E s and SCE s Cost and Vehicle Estimation Methodology is Not Robust, Accurate, or Targeted Enough to Provide a Basis for Reasonable Costs that Can be Approved by the Commission.. b. PG&E s and SCE s Programs Will Not Cost-Effectively Reduce GHG Emissions or Mitigate Risks to Ratepayers... III. Comparison of PG&E s and SCE s Estimated Costs Demonstrate Flaws and Uncertainty... a. SCE Errs in its Forklift Forecast Methodology, Resulting in Approximately $0 Million in Extraneous Costs... b. SCE s Medium-Duty Cost Estimates Are Excessive, Demonstrating the Extreme Uncertainty in Actual Costs...1 c. PG&E and SCE Contingencies Add Considerably to Costs and Are Unsupported...1 d. PG&E s Approach to Charging Station Rebates is More Reasonable than SCE s...1 e. SCE s Phase 1 Program Does Not Need to be Larger than PG&E s...1 IV. TURN Recommendations to Accelerate Transportation Electrification in Non-Light- Duty Sectors, Mitigate Ratepayer Risk, and Collect Fundamental Data... 1 a. An Initial Budget of $1 Million Over Four Years Supports Significant Acceleration of Transportation Electrification While Incenting Deployment of Cost-Effective Projects and Mitigating Some Ratepayer Risk...1 b. Minimum Site Selection Criteria and Emphasis on Critical Sectors Will Improve Program Administration, Cost-Effectiveness, and Initial Program Success... c. TURN Recommends Rebates for Electric Forklifts Rather than Infrastructure Subsidies... d. Fundamental Data Collection During Phase 1 Will Guide Future Programs to Cost-Effectively Accelerate Transportation Electrification... V. Performance Accountability Measures as a Risk-Sharing Mechanism Are Necessary to Incent Prudent Utility Investment... VI. SCE Must Abide by All Affiliate Transaction Rules Regarding its Stake in an Electric Bus Manufacturer... VII. Appendix 1 Eric Borden Statement of Qualifications... VIII. Appendix Comparison of PG&E and SCE Total Forecasted Sites and Vehicles.. IX. Appendix Data Requests Referenced in TURN'S Testimony...

3 I. Overview of PG&E and SCE Non-Light-Duty Infrastructure Proposals and TURN s Recommendations Pursuant to the April 1, 01 Scoping Memo and Ruling of Assigned Commission and Administrative Law Judges, TURN submits this testimony on Pacific Gas and Electric s (PG&E s) and Southern California Edison s (SCE s) standard review medium and heavy duty (MD-HD) infrastructure applications. The proposals include ratepayer subsidized utility and customer-side charging station infrastructure, as well as differing levels of charging station rebates, for essentially all non-light-duty electric vehicle (EV) sectors. This includes MD-HD trucks and busses as well as off-road equipment like port cargo handling equipment, forklifts, airport ground support equipment, transport refrigeration units (TRU s) and truck stop electrification (TSE). PG&E requests $ million 1 and SCE requests $ million over a year period to deploy infrastructure for these market segments. TURN strongly supports state goals to accelerate Transportation Electrification (TE), including for the non-light-duty EV sector. Electrification of this sector can decrease emissions, particularly criteria pollutants and NOx which adversely impact disadvantaged areas of the state causing severe health issues for vulnerable communities. Further, non-lightduty infrastructure programs have an even greater potential than light-duty programs to result in incremental EV adoption and related increased load, which, in addition to environmental benefits, can result in downward pressure on electric rates for all ratepayers. However, the latter point is only true if programs are carefully designed to be effective and targeted to where subsidies can provide the most impact for each investment. Ideally, investments would be selected that maximize emissions reductions for each ratepayer dollar, with particular emphasis on how to accelerate emissions reductions in disadvantaged communities (DACs). Unfortunately, this is not how PG&E and SCE s proposals are structured. PG&E and SCE admit that their vehicle forecasts, costs, and program impacts are highly uncertain, but still choose to place the risks of their programs entirely on ratepayers by not including sufficient deployment criteria, off-ramps, accountability measures, or reasonable cost caps that are necessary to incent cost-effective deployments. There are 1 Pacific Gas and Electric Transportation Electrification SB 0 Prepared Testimony, January 0, 01 ( PG&E Testimony ), p. -. Testimony of Southern California Edison in Support of its Application For Approval of its 01 Transportation Electrification Proposals, January 0 01 ( SCE Testimony ), p.. 1

4 virtually no criteria proposed to determine which areas of the non-light-duty sector will be targeted or emphasized, nor any minimum site requirements that would help result in costeffective program implementation. The utilities even request that no after-the-fact reasonableness review be conducted. This should concern the Commission. PG&E and SCE are requesting approval to spend an unprecedented amount, over $1 billion in ratepayer funds, for which the utilities accept no accountability to meet and exceed state goals, spend ratepayer funds prudently and intelligently, or provide the positive impacts claimed. Consistent with P.U. Code Section 0.1(b), the utilities must be held accountable for the performance of these programs. Ratepayers will be burdened financially by these programs, at the very least utilities should be required to provide the asserted benefits. TURN finds fundamental flaws in the utilities cost proposals and underlying vehicle forecasts such that, in conjunction with the lack of site and program criteria, the programs cannot be approved by the Commission (Section II). Approval of utility proposals in their current form would shift control of state energy policy to the utilities for these sectors where the only sure winners are PG&E s and SCE s shareholders. However, by adopting TURN s modifications, the Commission can shape utility programs to accelerate TE while mitigating ratepayer risk through Phase 1 deployment that simultaneously collects fundamental data, currently lacking in utilities testimonies, to inform better-designed programs in the near future. Given the imperative to begin acceleration of TE in the non-light-duty sector now, but keeping in mind the substantial uncertainties and corresponding ratepayer risk, TURN recommends the following modifications be adopted by the Commission for PG&E s and SCE s non-light-duty EV programs: SCE Testimony, p., lines 1 to 0. PG&E Testimony, p. -. PG&E states the Commission should adopt a policy that PG&E should be allowed to recover all its costs as long as the expenditures were made pursuant to, and consistent with the specific SB 0 TE Program spending authority and guidance provided by this Commission. This is somewhat unclear, though taken at face value may impede parties ability to oppose the reasonableness of utility expenditures. P.U. Code Section 0.1(b) states: «The commission shall approve, or modify and approve, programs and investments in transportation electrification, including those that deploy charging infrastructure, via a reasonable cost recovery mechanism, if they are consistent with this section, do not unfairly compete with nonutility enterprises as required under Section 0., include performance accountability measures, and are in the interests of ratepayers as defined in Section 0.. See Section V for a discussion of TURN s Performance Accountability Measures proposals.

5 A budget of no more than $1 million over four years ( Phase 1 ) subject to minimum site criteria and program requirements (Section IV): o Each site host that participates in the program should commit to purchasing at least two electric vehicles, but sites which commit to more electric vehicle adoption (in both the near and medium-term) should be prioritized over those that commit to less. o A minimum of 0% of sites should be located in disadvantaged communities (DACs). o $1 million of the allocated budget should be reserved exclusively for the deployment of electric busses. o TURN s budget is sector-specific such that funds may not be shifted among sectors (in contrast to the general spending request proposed by the utilities), with any unspent funds returned to ratepayers. Conversion of fossil fuel forklifts to electric should be accomplished with a rebate program rather than subsidized infrastructure (Section IV(c)). Phase 1 data collection (Section IV(d)) should include the following, which should directly inform future utility programs: o A bottoms-up accounting of each non-light-duty sector to determine a more accurate estimate of the number of vehicles that can reasonably be forecast and displaced with EVs particular to each utility territory. o Relative cost-effectiveness (dollar per greenhouse gas emissions avoided) of each site and non-light-duty sector. o Specific data collection and proposal for how to mitigate the potential for stranded costs, particularly over the long-term. These are notably absent from utility proposals. o Better understanding and analysis of what investments best reduce particulate emissions in DACs. o Calculation of ratepayer payback period for each site and comparison of sectors. o The commercial, economic, and technical maturity of electrification for each respective sector. o Factors that drive per site cost variability and how to minimize costs in future deployments. o Site host and sector willingness to pay. With the exception of airport GSE, Port CHE, and TSE, where TURN estimates a combined budget of $. million and does not oppose a one-way balancing account for these three sectors.

6 Further, TURN s analysis of the drivers behind the differences in PG&E s and SCE s cost estimates (Section III) demonstrate the following: Due to an error in its forecasting methodology, SCE over-estimates costs for the forklift sector by at least $0 million. SCE s per site costs for the medium-duty sector are excessive. Contingencies for both utilities are unsupported and excessive. PG&E s approach to charging station rebates is more reasonable than SCE s and should be approved for Phase 1, with TURN s modifications. SCE has not demonstrated its program should be larger than PG&E s in Phase 1. In the following sections, TURN details PG&E s and SCE s approach to vehicle forecasting, cost estimates and program design, illustrating why the proposed programs cannot be approved as proposed (Section II). TURN also provides a comparison of PG&E s and SCE s cost estimates that illustrate significant shortcomings of the utility proposals as well as one aspect of PG&E s proposal that should be approved with TURN s modifications (Section III). Next, TURN s proposal to mitigate program risk with various program design elements and robust data collection metrics is discussed in Section IV, TURN s recommended Performance Accountability Measure (PAM) is addressed in Section V, and finally Section VI discusses the need for SCE to adhere to affiliate transaction rules due to its unregulated affiliates investment in an electric bus manufacturer. II. PG&E s and SCE s Program Design and Cost Estimates Are Broad and As Proposed Will Not Lead to Effective Investments that Maximize Benefits and Minimize Costs Generally, PG&E and SCE do not have sufficient granular data and program experience to develop reasonable cost estimates for their non-light-duty infrastructure programs. Further, even the most fundamental program design guidelines are lacking. The absence of these key aspects result in proposals that do not maximize benefits and minimize costs to ratepayers, as required by both SB 0 and the Assigned Commissioner Ruling. Assigned Commissioner s Ruling Regarding the Filing of the Transportation Electrification Applications Pursuant to Senate Bill 0, September 1, 0 ( ACR ), Appendix A, p. A1.

7 a. PG&E s and SCE s Cost and Vehicle Estimation Methodology is Not Robust, Accurate, or Targeted Enough to Provide a Basis for Reasonable Costs that Can be Approved by the Commission With a few exceptions PG&E s and SCE s cost estimates are based on statewide vehicle forecasts for each MD-HD and off-road sector which are then apportioned to the respective utility territory. Generally, PG&E and SCE rely on a reference or in between scenario (usually the middle of a low and high vehicle estimates) and assume the utility provides 0% subsidized infrastructure cost for the forecasted vehicles over a five-year period. PG&E uses CARB, CEC, and ICF studies for California-wide estimates; where more granular studies or data is not available to PG&E, the utility apportions % of the statewide forecast to its territory. SCE relies solely on the In Between Scenario from the California Transportation Electrification Assessment ( CalTEA ) study for its vehicle forecast, apportioning % of the total based on SCE s share of commercial and industrial activity with the exception of airport ground support equipment (GSE) sector (% of the CalTEA In Between forecast). These vehicle estimates then directly drive the utilities cost requests, whereby the utilities, using similar estimates (many from EPRI) on attach rates (chargers required per EV) and power levels, incorporating multiple sensitives, derive a cost forecast for all utility and customer-side charging infrastructure costs. PG&E and SCE also add considerable contingency factors to the capital estimates (addressed separately in Section III(c)), and SCE estimates costs to provide charging station rebates for 0% of the base cost for all stations and sectors, based again on the five-year vehicle forecast. It is important to note that while PG&E and SCE use vehicle forecasts and related charging infrastructure to directly derive the request for ratepayer funds, the utilities do not commit to any specifics on how the funds would actually be spent. The approach described Namely, PG&E provides a more granular territory-wide estimate for transit buses, airport GSE, port cargo handling equipment, truck stop electrification, and forklifts. SCE apportions % of airport GSE statewide vehicles rather than %. For PG&E, these costs represent (at maximum) 1% of the utility s total cost forecast. Airport GSE infrastructure represents less than 1% of SCE s total cost estimate. Contingency not included in these percentage estimates. See Table. California Air Resources Board, California Energy Commission, ICF International. PG&E Testimony, p. -. ICF International (01), California Transportation Electrification Assessment Phase 1: Final Report ( CalTEA study ), FINAL_Updated_001.pdf. SCE Testimony, D-.

8 above results in the following infrastructure cost estimates per sector (not including contingency or rebates): Table 1. PG&E and SCE Infrastructure Cost Forecasts by Vehicle Sector 1 $ Thousands, Nominal PG&E SCE Forklifts $1, $1, Medium-Duty Vehicles $,0 $1, Heavy-Duty Vehicles $1, $, Port Cargo Handling Equipment (CHE) $,00 $1, Transport Refrigeration Units (TRU) $, $, Truck Stop Electrification (TSE) $1, $,00 Airport GSE $, $, Total $1, $, There are five primary reasons PG&E s and SCE s cost forecasts cannot be approved by the Commission: 1) Forecasts based on statewide totals do not accurately reflect the actual potential for TE in the utility s territory. As stated by SCE forecast accuracy is challenging. 1 PG&E admits [g]iven the uncertainties in adoption rates for TE in the non-light-duty vehicle sector, actual market demand for PG&E s FleetReady infrastructure will be higher or lower than the reference case over the five-year period. 1 This is demonstrated by the extreme uncertainty in the forecasts; for example, in 01 the CalTEA study has a range for its low and high scenario of between, Medium Duty vehicles (the reference case used by PG&E is for that year, which is then apportioned to the utility s territory). 1 ) Apportionment according to the utility s share of California likely compounds the first inaccuracy the number of vehicles adopted will be driven by a variety of factors, such as the presence of unique distribution centers and warehouses located in each utility s territory for the medium-duty sector. A bottoms-up 1 For PG&E see TURN-01, question 1, attachment SB0 Program Master Program Cost Summary, individual sector tabs. For SCE data from TURN-01, question 1, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs. 1 SCE Testimony, p. D-. 1 PG&E Testimony, p. -1, lines -. 1 PG&E Testimony, p. -1, Table -.

9 analysis of the particular industry and commercial activity that utilizes conventional versions of the vehicles that can be electrified through the program is required for accuracy rather than the top-down methodology used by PG&E and SCE for most of the estimated costs. PG&E shows this is feasible for some of its forecasts. ) PG&E and SCE assume 0% of the necessary infrastructure to support the vehicles in each category should be subsidized by ratepayers. This is inappropriate because as stated in the ACR utility ratepayers will not be able to bear all the costs of accelerating TE in California. 1 In addition, some of these vehicles would be converted to electric without ratepayer subsidy, which inflates the utility request for ratepayer funds. ) The cost estimates developed by PG&E and SCE are extremely uncertain, with the utilities essentially averaging (usually weighted to the more expensive estimate) potential costs. For example, SCE s site cost estimates show variances up to almost 00% between the high and low estimates. 1 This further compounds the forecast uncertainties found in the rest of the program. ) The final reason the utility forecasted costs cannot be approved is PG&E and SCE do not commit to a specific funding level (or any specific amount) for each sector according to their own cost forecasts. The utilities thus ask approval of ratepayer funding by the Commission with just a superficial understanding of how the funds will be spent, so the potential effects of the program are essentially unknown. b. PG&E s and SCE s Programs Will Not Cost-Effectively Reduce GHG Emissions or Mitigate Risks to Ratepayers There is tremendous uncertainty in the non-light-duty market regarding commercial deployment of various technologies and how customers will respond to subsidies offered by the utilities, as acknowledged by both PG&E and SCE: PG&E: [T]he costs for the FleetReady make-ready infrastructure project are highly uncertain, due to both the nascent status of market demand for non-light-duty EVs, and the uncertainty regarding the precise mix of non-light-duty vehicles and EV infrastructure Though TURN has not vetted each analysis, PG&E provides a more granular territory-wide estimate for transit buses, airport GSE, port cargo handling equipment, truck stop electrification, and forklifts. SCE apportions % of airport GSE statewide vehicles rather than %. For PG&E, these costs represent (at maximum) 1% of the utility s total cost forecast. Airport GSE infrastructure represents less than 1% of SCE s total cost estimate. Contingency not included in these percentage estimates. 1 Assigned Commissioner s Ruling Regarding the Filing of Transportation Electrification Applications Pursuant to Senate Bill 0, September 1, 0 ( ACR ), p For the heavy-duty estimated site cost, combination of both utility-side and customer-side costs. TURN-01, question 1, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs.

10 sites that will need make-ready infrastructure over the limited five-year period of the FleetReady program. Thus, the forecast costs for the FleetReady program below and the estimated demand for the program should not be considered forecasts of the actual market demand and mix of types of non-light-duty EVs and EV infrastructure sites that will actually be served by the program. 1 SCE: For the standard review Medium- and Heavy-Duty Vehicle Charging Infrastructure Program, vehicle composition and resulting GHG and pollutant reductions will vary by customer composition and participation, which will be influenced by several variables out of SCE s control. For example, the GHG and NOx reductions from a program composed of forklifts be different than one composed of medium-duty trucks. 0 Rather than propose a reasonable level of funding with clear program criteria to guide initial deployment, PG&E and SCE place the risks associated with the development of a nascent industry on ratepayers. Essentially, PG&E s and SCE s proposals are to provide customers with subsidies on a first-come, first-serve basis, without identifying clear criteria for Commission approval a priori to prioritize where or how ratepayer funds should be invested. SCE s hypothetical in the response cited above is instructive would the State, Commission, and other stakeholders really support massive subsidies just for the electric forklift sector? TURN hopes not, and reiterates the need for cost-effective deployment of ratepayer dollars that can most meaningfully accelerate TE, reduce emissions, mitigate impacts on low and moderate income ratepayers, and allow all ratepayers to benefit from TE with cleaner air, especially in the most burdened communities, and lower bills. Furthermore, PG&E and SCE do not have reasonable site criteria to ensure cost-effective deployment of ratepayer funds. One basic criteria is the number of vehicles per site; especially because SCE estimates an average cost per site of over $00,000, 1 any site that applies for subsidy should not be simply approved. Ratepayers should not pay $00,000 (or more) for the adoption of one electric vehicle. Yet, this appears to be possible, if not likely, under the utilities program design: SCE: The minimum number of a) chargers per site is one and b) vehicles per site is one. SCE believes that a floor on the number of chargers installed may limit program participation. 1 PG&E Testimony, p. -0, lines -. 0 TURN-SCE-01, question. 1 Data from TURN-SCE-01, question 1, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs. TURN calculation. TURN-SCE-0, question 1.

11 Program participation should be limited, and ideally targeted, to ensure some level of relative cost-effectiveness and prudent use of ratepayer funds. TURN discusses a smarter program design and a vision for future deployment, along with fundamental data collection during Phase 1 in Section IV. III. Comparison of PG&E s and SCE s Estimated Costs Demonstrate Flaws and Uncertainty As discussed above, TURN does not believe the utilities cost estimates are reasonable or sufficiently supported to warrant approval by the Commission. Nevertheless, it is instructive to explore why the two utility program cost estimates ($ million for PG&E, $ million for SCE) vary so widely, particularly given that the methodology to forecast costs use, for the most part, very similar methodologies, and aside from rebates, support the same activities - utility and customer-side charging station infrastructure - for the same non-light-duty sectors. TURN s analysis demonstrates this is due primarily to four areas of the cost forecast: forklifts, mediumduty vehicles, contingency, and charging station rebates.

12 1 1 1 Table. Comparison of PG&E and SCE Five Year Cost Estimates (All Program Costs) $ Thousands, Nominal Percent of SCE PG&E SCE- PG&E Estimate Difference Forklifts $1, $1, $,1 1% Medium-Duty Vehicles $1, $,0 $, % Contingency $, $,0 $, 1% Charging Station Rebates $, $,000 $, % Heavy-Duty Vehicles $, $1, $,0 % Port Cargo Handling Equipment $1, $,00 $, % Transport Refrigeration Units $, $, -$, -% Truck Stop Electrification $,00 $1, $, 1% Airport GSE $, $, -$ 0% Labor/non-labor costs, Other O&M and capital $1, $, -$1,00 -% Total $, $, $,1 The four primary areas of cost difference (forklifts, medium-duty vehicles, contingency, and charging station rebates) are discussed in the following sub-sections. In addition to the reasons stated above, the discussion demonstrates the majority of utility cost estimates for infrastructure cannot be approved by the Commission in part because they are speculative, uncertain, and in the case of forklifts for SCE, contain an error. Further, aspects of PG&E s approach, such as targeted charging station rebates that are expensed, should be adopted by the Commission because they are more reasonable and provide, overall, for greater leveraging of non-ratepayer funds, an aspect of the ACR essentially ignored by SCE. For PG&E see TURN-01, question 1, attachment SB0 Program Master Program Cost Summary, individual sector tabs. For SCE data from TURN-01, question 1, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs. ACR, Section..

13 a. SCE Errs in its Forklift Forecast Methodology, Resulting in Approximately $0 Million in Extraneous Costs As explained by EPRI in its market analysis of potential electrification in PG&E s territory related to forklifts, there are five classes of forklifts on the market today: Forklifts are categorized under five different classes and may operate throughout the day and night depending on the specific operation, number of hours used, and number of shifts. Classes 1,, and represent all electric forklifts, while Classes and account for the internal combustion (IC) portion of the overall forklift market. In order to forecast the potential for electrification of forklifts in PG&E s territory, EPRI thus estimated the potential to convert Class and forklifts to Class 1: EPRI estimates the electrification opportunity based on expected Class and sales, using Class 1 market growth to inform realistic conversion rates. This is a reasonable methodology to forecast potential electrification of the forklift sector. The key is that organic growth of class 1- forklift does not represent electrification, because these vehicles are, by default, electric. Conversion of Class and forklifts, which are powered by diesel or propane, to Class 1, represents electrification, as demonstrated by EPRI s approach for PG&E. SCE has not factored in this distinction; the utility s basis for its forklift forecast, the In Between scenario from the CalTEA study, is inappropriate because it includes organic growth of electric forklifts, as well as conversions of class / to electric (class 1 or ). The utility appears to acknowledge this in response to a data request though SCE does not seem to recognize it as inappropriate: The forklift forecast used for estimating the program size and associated cost includes only incremental class 1 and class electric forklifts. This would include conversions from class and combustion forklifts to class 1 and electric as well as new growth in class 1 and forklifts (Class 1 and are electric substitutes for Class and combustion forklifts). Again, SCE s forecasting methodology is flawed because it includes forklifts that would be electric regardless of SCE s program because no non-electric alternative exists. Put another Pacific Gas and Electric Company, Electrification Case Study Report, March 01 ( EPRI Report ), p. -1. EPRI Report, p. -. TURN-SCE-00, question 1.

14 way, a portion of the considerable costs represented in SCE s forecast would result in zero emissions savings (potentially an emissions increase) and zero ratepayer benefits. EPRI s methodology for PG&E, on the other hand, accounts for this important distinction by forecasting conversions of class and forklifts, which would otherwise burn fossil fuel, to class 1 (electric). Due to the forecasting error of including organic growth of electric forklifts in addition to conversions, SCE over-forecasts the number of sites and vehicles that can be electrified through its program: Figure 1. Forecasted Forklift Sites and Vehicles Over Respective Utility Five Year Program The correct methodology is to examine conversions of class and forklifts to class 1 or. Using SCE s framework, this can be done by subtracting the CalTEA study s low forecast (called In Line with Current Adoption ) from the In Between scenario (used by SCE) and then apportioning the difference to SCE s territory (% of the statewide forecast). This methodology targets forklift conversion because the In Line forecast represents just organic growth of the class 1 and market (no coversions of class / forklifts), while the In Between scenario contains accelerated conversion of class / forklifts. The difference between the scenarios is therefore conversions. This calculation actually results in a forecast similar to PG&E s -,00 vehicles at 1 sites over the five year period (versus PG&E s forecast of, As noted above, these forecasts are for cost purposes. Data request (DR) TURN-PGE-01, question 1, SB 0 Program Master Program Cost Recovery, DR TURN-PGE-0, question, PGE-SB 0 TE Vehicle Adoption Forecasts. TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tabs Aggregate Costs and Vehicle Population. This approach and the fact that the low scenario is just organic growth was confirmed via an exchange with an employee from ICF, Jeffrey Rosenfeld. 1

15 at 1 sites). 0 Using SCE s per site cost estimate for forklifts this results in a forecast of $ million (again, assuming SCE s methodology), rather than the stated request of $1 million. 1 Despite a lower vehicle estimate (with TURN s adjustment) SCE s cost are still almost three times those of PG&E s for the forklift vehicle category (PG&E s are about $1 million for the forklift sector). This is because SCE s average per site cost estimate is just under three times as expensive as PG&E, $0,000 versus $1,000 respectively. This demonstrates not only that SCE s costs are excessive, but that the tremendous uncertainty in cost estimates between the utilities demonstrates that real-world data collection is required before a larger program can be allowed by the Commission. b. SCE s Medium-Duty Cost Estimates Are Excessive, Demonstrating the Extreme Uncertainty in Actual Costs As shown in Appendix, while SCE does forecast more medium-duty vehicles than PG&E, the cost estimates assume a similar number of sites (in fact SCE assumes less), which is the main driver of infrastructure costs. SCE s per site costs are simply much higher than PG&E s, around $00,000 per site versus $,000 for PG&E, not including contingency. TURN was not able to identify exactly why this large discrepancy exists (e.g. what planning assumption is driving excessive costs for SCE). But at a minimum, this illustrates again the extreme uncertainty for what the actual cost of infrastructure deployment for these sectors will be. c. PG&E and SCE Contingencies Add Considerably to Costs and Are Unsupported SCE and PG&E add considerable contingencies to their capital cost estimates. SCE s contingencies are around % of capital costs; PG&E s are %. SCE s total dollar contingency 0 The organic growth of forklifts for the CalTEA study, represented by the In Line with Current Adoption forecast can be found on page of the study. TURN utilizes SCE s methodology which calculates the incremental number of forklifts using a compound annual growth rate (CAGR) calculated from the CalTEA study s forecast, and then apportioned to the utility s territory (%). The number of sites is calculated using SCE s assumed number of vehicles per site (about 1.). Data from TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tab Aggregate Costs. 1 Data from TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tab Aggregate Costs. Calculated by dividing total costs by total number of sites for the relevant cost categories. DR TURN-SCE- 01, question 1, MD-HD Cost Estimate Infrastructure, tab Aggregate Costs. TURN-PGE-01, question 1, PGE SB 0 Program Forklift WPs. TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tab aggregate costs. PG&E costs divide the $ million total MD Truck forecast by forecasted number of sites,. TURN-PGE-01, MD Truck WPs. 1

16 amount is much higher simply because its cost forecasts are higher. However, the large impact of this category demonstrates the tremendous ratepayer impact of contingency. TURN finds that PG&E s and SCE s contingency factors are excessive and unsupported. For SCE s program, contingency costs amount to $0 million and are designed to provide certainty and protection to SCE in case the utility cost estimate is substantially wrong: SCE: Since each customer site is unique with many factors influencing costs, SCE includes a percent contingency in its cost estimate. PG&E s contingency cost of around $ million represents a similar percentage of capital costs as SCE. Like Edison, the utility states that site cost variability drives its contingency forecast: PG&E: PG&E recognizes that there are many issues that may arise during implementation of any large, complex capital program, such as FleetReady, that will impact the costs. Given that the characteristics of the customer sites participating in the Program are not currently known, the actual geographic, site-specific, and technical conditions will likely vary from the assumptions used to inform the cost forecasts. For example, if PG&E s assumption that very few sites will be located in super-urban areas the most expensive site host location is incorrect, the program costs per installation will increase substantially. Additional variations in geographic, technical and site-specific factors such as those discussed above may lead to additional cost. Therefore, to address the variations in cost, PG&E has included a percent contingency overall on the EV service connection and supply infrastructure work. There are several reasons the utility forecasts for contingency are not reasonable. First, much of the variability cited by the utilities have already been factored in to the cost estimates. For example, SCE s estimates include essentially a weighted average site complexity (usually resulting in a low and high cost estimate) to determine each sector s per site cost. This results in site estimate costs ranging from a ratio of 1 (where low and high estimate were the same) to (where was a 00% difference between the low and high estimate) both the high and low site estimate determine the weighted average. Similarly PG&E utilizes a weighted average for each sector of various site complexities, including very expensive sites, to determine its make-ready costs. For example, the medium-duty service connection site estimate includes site costs that range from $,000 to $,000, whereby the utility heavily weights its estimate toward the more SCE Testimony, p.. While SCE s workpapers (TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tab aggregate costs. ) indicate capital contingency of around %, testimony states %. PG&E Testimony, pp. - to -0, lines 1 to. Data from TURN-SCE-01, question 1, MD-HD Cost Estimate Infrastructure, tab aggregate costs. These ratios combine utility-side and customer-side costs for simplicity. 1

17 expensive/complex sample site. If utilities had assumed only low costs, and then added contingencies, the result would likely be more reasonable; however this is not the case. Second, for the utility side costs, which comprises around half of the infrastructure costs, the utility has decades, if not over a century, of experience upgrading distribution infrastructure due to increased or new loads. The utility cannot claim ignorance as to what potential costs will be. Admittedly, different numbers of vehicles and charging levels will drive variability in these costs, but again much of this has been factored in to the utility cost estimates. Third, there is no clear basis for why contingency should be added to costs; TURN could argue with the same logic that infrastructure costs may be % lower than utility forecasts. While TURN does not go as far as to recommend reducing infrastructure estimates by %, the utilities have not sufficiently supported the need for large additional cost contingencies that will burden ratepayers and incentivize wasteful projects and expenditures. d. PG&E s Approach to Charging Station Rebates is More Reasonable than SCE s SCE proposes rebates for 0% of the base cost of EV charging stations for all sites and vehicle sectors, totaling $ million. On the other hand, PG&E proposes rebates targeted to DACs, electric public transit, and school bus charging, estimating a cost of $ million. SCE s approach to rebates is excessive, unnecessary, and unsupported. Given the significant level of subsidy already proposed by the utilities, available incentives for vehicle purchases, and the operating savings of electric vehicles over gasoline-fueled vehicles for fuel and maintenance costs, it is important that non-ratepayer funding be leveraged as prescribed in the ACR. 0 While it is not clear whether the rebates will, in fact, be needed, PG&E s approach to target rebates to DACs and electric busses is a more reasonable approach than SCE because it is most likely to influence GHG emission reductions where they are needed most and leverage at least some funding from the majority of site hosts. Further, TURN supports deployment of electric busses and increased emphasis on disadvantaged communities to provide benefits to the communities most severely impacted by MD-HD sector emissions. While TURN initially does not oppose PG&E s approach to rebates (and they are included in TURN s alternative cost TURN-PGE-01, question 1, Attachment PGE SB0 Program MD Truck WPs.0.1, tab Service Connection Summary. PG&E Testimony, p. -. Several are listed on PG&E Testimony line, Table -1, p ACR, Section.. 1

18 forecast but modified for a four-year program), more data should be collected during Phase 1 to understand whether these rebates are actually needed in the sectors foreseen by PG&E and how effective they are in driving increased EV adoption. SCE (again in contrast with PG&E s approach) also proposes to capitalize its rebates as a regulatory asset rather than treat them as an expense. Given that the program is almost all capital costs (% as proposed) 1 and presents little risk to SCE, the utility does not need any additional incentives to deploy its program. Further, there is no sound basis for this accounting treatment and it serves only to increase costs to ratepayers with no corresponding benefit. SCE shareholders stand to make significant profits from the capital costs associated with this program, aside from how rebates are treated. In fact, if rebates are not capitalized in SCE s program, % of the costs will still be capitalized. Figure. SCE Non-Light-Duty Infrastructure Program ($ Thousands) Total Cost Proposal if Rebates are Not Capitalized $00,000,1 $00,000 $00,000 $00,000 $0,000 1, $0 Total Capital Expenditure Without Rebate O&M If Includes Rebate TURN s proposal for limited rebates targeted to certain sectors, and expensed as proposed by PG&E, is discussed in Section IV(a). 1 TURN-SCE-01, question 1, attachment TE Filing Master Summary. $,1 / $, = %. Data from SCE Testimony, Table III-, p..

19 e. SCE s Phase 1 Program Does Not Need to be Larger than PG&E s SCE contends, in its response to an ORA data request, that its program should be larger than the other IOU s: In addition to serving the Port of Long Beach (POLB), SCE also serves the surrounding distribution centers and warehouses fed by Port of Los Angeles (POLA) and POLB. According to the World Port Rankings.[t]he two ports moved more goods than any other port in the United States. POLA and POLB facilitate nearly seven times more containers than the next largest port in California (Port of Oakland). Comparatively, PG&E does not serve the Port of Oakland, and even though SDG&E serves the Port of San Diego, the volume of goods is a fraction of what moves through POLA/POLB. Therefore, relative to the other IOUs, SCE would be expected to have a larger population of medium-duty, heavy-duty and non-road (including forklifts) vehicles in its service territory. There are several reasons SCE s explanation does not warrant a larger program at this time. First, the vehicle and cost forecast methodology used by SCE (see above) is not granular enough to reflect the differences claimed by SCE in the non-light-duty vehicle sectors. As described in Section II(a), SCE actually applies a slightly smaller percentage of the relevant statewide forecast to its territory compared with PG&E. SCE has not done the required work, including granular sectoral analysis that TURN recommends be accomplished during Phase 1 for all sectors, to show that the number of vehicles that can be electrified in its territory over the next several years is greater than PG&E. Second, as discussed in Section IV, TURN recommends the Commission order the utilities to conduct a Phase 1 program with a specific budget for each sector. Pursuant to this recommendation SCE need not have a larger program than PG&E because a Phase 1 program should only electrify a portion of the non-light-duty vehicles in the utilities territory. This allows for acceleration of TE while utilities and relevant agencies collect and analyze fundamental data (Section IV(d)) currently lacking that will help achieve a successful program. IV. TURN Recommendations to Accelerate Transportation Electrification in Non- Light-Duty Sectors, Mitigate Ratepayer Risk, and Collect Fundamental Data As described above, utility proposals are uncertain and unsupported, placing tremendous risks on ratepayers. The programs can more meaningfully accelerate TE if they incorporate program design features to guide deployment, maximize benefits and minimize costs where possible, and fill in the considerable data gaps present in the utilities testimony. There are three DR ORA-SCE-001, question.b.iii. 1

20 primary ways this can be accomplished: Adopt a budget of $1 million over years for Phase 1, subject to TURN s recommended sectoral budget caps, site criteria, and program design features. This allows the utility to deploy infrastructure to accelerate TE while collecting the fundamental data required to support additional ratepayer funds. The development of TURN s budget is detailed in Section IV(a). Site selection criteria and a focus on particular segments is critical to initial stage deployment and a minimum level of cost-effective deployment (Section IV(b)). o Each site host that participates in the program should commit to purchasing at least two electric vehicles, and sites which commit to more electric vehicle adoption (in both the near and medium-term) should be prioritized over those that commit to less. o A minimum of 0% of sites must be in disadvantaged communities. o $1 million of each utility s budget should support deployment of electric busses. Conversion of class and forklifts (internal combustion) to class 1 or (electric) should be funded through a rebate program rather than infrastructure subsidies (Section IV(c)). Fundamental data collection is critical to the long-term success of cost-effectively reducing emissions from the non-light-duty sector. This should include data collection on GHG emission reductions per site, sector, as well as analysis of vehicle technological maturity and business case (cost savings) for participants. Critically, a bottoms-up analysis of each vehicle sector to develop a more precise inventory and vehicle forecast for each utility must be completed. Section IV(d) provides an overview of the data collection that should occur in Phase 1. a. An Initial Budget of $1 Million Over Four Years Supports Significant Acceleration of Transportation Electrification While Incenting Deployment of Cost-Effective Projects and Mitigating Some Ratepayer Risk While TURN generally prefers short pilot periods with low budgets to both mitigate ratepayer risk and collect learnings quickly, this may be difficult to accomplish for MD-HD sectors. As stated by SCE, some EVs require longer lead times to manufacture and purchase. For example, transit agencies phase in their vehicle purchases over several years, because of manufacturing and financial constraints. In addition, TURN recognizes data collection may SCE testimony, p., lines -1. 1

21 need to take place over a longer period due to the nascent status of the market, significant data gaps, and historic lack of utility involvement in the relevant sectors. The majority of utility cost forecasts are unsupported, overly-broad, and in many cases, unreasonably high. However, TURN finds the following aspects of PG&E s proposal as containing sufficient support to form a basis for a Phase 1 budget for PG&E and SCE, described further below: PG&E provides more granular territory-specific forecasts for transit buses, airport GSE, port cargo handling equipment, truck stop electrification, and forklifts. With the exception of transit busses and forklifts, TURN recommends % of PG&E s estimated budget for both utilities be approved for Phase 1 so that ratepayers are not responsible for 0% of TE in these sectors. This equates to $. million for these sectors. PG&E proposes targeted charging station rebates to incent adoption in DACs, electric public transit, and school busses. TURN supports this proposal and the goals of the rebates. TURN s budget provides for four years of rebates rather than five. PG&E and SCE s approach of asking the Commission to approve large and uncertain cost forecasts with the promise to return unspent funds does not protect ratepayers from funding wasteful projects and incents utilities to spend up to their (excessive) forecasts. Instead, TURN recommends a more reasonable cost cap that can incent smarter and more cost-effective project selection while dramatically accelerating TE and collecting fundamental data to inform a future phase. This provides a level of certainty to program funding while protecting ratepayers from incurring excessive or unnecessary costs. TURN uses aspects of PG&E s cost estimates as a starting point for our proposed budget, and provides more reasonable cost caps for other sectors. The discussion above in Section III(e) demonstrates SCE has no basis for a larger budget than PG&E without the required data collection. TURN s four-year budget is as follows: PG&E Testimony, p. -1, Table -. This represents at most 1% of PG&E s total budget request. See Table. TURN recommends greater emphasis on transit busses with a $1 million set-aside and a rebate program for forklifts rather than subsidized infrastructure. TURN adopts the five year budget to ensure sufficient funding despite our recommendation for a four year Phase 1 program. 1

22 Table. TURN Recommended Phase 1 Budget for PG&E and SCE ($ Million) TURN Four Year Budget Source/ Sector / Budget Element Estimate Basis Make-Ready Infrastructure - Electric Busses $1.0 1 Charging Station Rebates (1) $1. Make-Ready Infrastructure- Airport GSE, Port CHE, TSE () $. Make-Ready Infrastructure- Transport Refrigeration Units $.0 Make-Ready Infrastructure - Other Heavy-Duty $.0 Forklift Rebate Program $. Make-Ready Infrastructure - Medium-Duty $.0 Non-Infrastructure Costs (e.g. O&M, Management, marketing, etc.) $.1 Total $1. Notes: (1) Treated as an expense. () TURN assumes an even allocation (around $1. million each) for each of these sectors. TURN recommends a one-way balancing account in this instance for the three sectors. Sources: 1. Based on PG&E's estimate for heavy-duty and TURN's recommendation for emphasis on the electric bus sector.. Based on PG&E's four-year estimate for rebates to encourage electric bus and disadvantaged community deployment.. Seventy-five percent of PG&E s cost estimate to ensure sufficient funds for this sector while likely not subsidizing 0% of TE.. TURN estimate of a reasonable cost cap that can also accelerate transportation electrification in this sector for Phase 1, pending data collection.. PG&E's four-year forklift vehicle estimate multiplied by a $,000 rebate. Slight reduction (0% of total population) to ensure ratepayers do not subsidize 0% of TE in this sector.. PG&E estimates 1% of costs will be non-infrastructure (mainly Program Management Organization costs, which appear excessive); SCE estimates around %. TURN provides % of total costs as "non-infrastructure" though hopefully utilities can minimize these costs further. 0

23 1 1 1 In order for the Commission to have transparency into what sectors are targeted, a one-way balancing account for each line item in TURN s proposed budget should be used to record costs and impose a reasonable cost cap for the sector over the course of the program. HD sectors while appropriately balancing affordability, equity, and ratepayer risk issues. Using each utility s own cost and vehicle estimates (other than for forklifts which incorporates TURN s rebate approach), TURN s budget can support the following number of EVs by sector in Phase 1: Table. Electric Vehicles Supported by TURN s Budget 0 SECTOR PG&E SCE Medium Duty,0 Heavy Duty TURN emphasizes its budget can result in significant acceleration of TE for the MD- Off- Road Heavy duty (transit) 1,0 1 Heavy-duty (non-transit) Airport ground support equip. (GSE) Port cargo handling equip. (CHE) Transport refriger-ation units (TRU) 1, Truck-stop electrific-ation spaces 1 (TSE) Forklifts (class 1)* 1, 1, Total,0, The largest discrepancies between the two utilities are for the medium-duty sector due to SCE s much larger per site cost estimate (see Section III(b)); Transport Refrigeration Units, where SCE estimates a much greater number of vehicles per site 1 and heavy duty transit, where PG&E estimates a lower per site cost ($1,000 versus $0,000) and greater number of vehicles per TURN recommends a one-way balancing account for the four-year program period. 0 The forklift forecast is the same for both utilities due to TURN s recommendation to institute a rebate program for this sector, based on PG&E s estimate which is similar to the corrected forecast provided by TURN for SCE in Section III(a). For PG&E s airport GSE, Port CHE, and TSE sectors TURN multiplies PG&E s four-year vehicle forecast by %. Other sector forecasts use PG&E s estimated cost and vehicles per site to derive the number of vehicles supported in each sector by TURN s budget. Cost per site estimates from separate sector workpaper attachments TURN-01, question 1. PG&E vehicle forecast from TURN-0, question, attachment, tab FleetReady Program Forecast. Applying TURN s budget to SCE s assumptions for cost and vehicles per site, TURN similarly calculates the number of vehicles supported by TURN s budget. This data is derived from SCE DR TURN-01, question 01, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs. 1 1 (SCE) versus 1 (PG&E). 1

24 site ( versus ). b. Minimum Site Selection Criteria and Emphasis on Critical Sectors Will Improve Program Administration, Cost-Effectiveness, and Initial Program Success PG&E s and SCE s proposals do not have sufficient site selection criteria to promote cost-effective deployment or maximization of emission reductions per dollar spent. Further, though the utilities mention the importance of these programs to air and health quality in DACs, they do not provide sufficient assurance the program will be targeted to markedly focus investments on reducing emissions and criteria air pollutants in DACs. TURN proposes the following minimum site criteria, which should be revised after Phase 1 data collection to ensure cost-effective investments and a move away from the one size fits all approach advocated by the utilities: Each site host that participates in the program should commit to purchasing at least two electric vehicles, but sites which commit to more electric vehicle adoption (in both the near and medium-term) should be prioritized over those that commit to less. This approach allows the large fixed costs for each site to influence greater emissions reductions, vehicles, and ultimately transportation electrification, than the purely market approach advocated by the utilities. As stated by PG&E California s disadvantaged communities (DAC) are often the most affected by the harmful environmental impacts associated with the transportation sector. The Commission should therefore require that a minimum of 0% of sites be located in disadvantaged communities. TURN provides for additional rebates for these communities in its recommended budget though it is not clear these site hosts require additional subsidy. Nevertheless, additional rebates for DAC sites ensures the 0% target is achievable. Further, % of the applications received by SCE for its light-duty ChargeReady program through May 01 are from sites located in DACs, representing % of the charge ports requested. TURN s goal of 0% is reasonable and attainable, not to mention well-aligned with the goals of the utility programs. Site cost data for PG&E from individual sector cost workpapers DR TURN-01, question 1 (TURN divides total non-contingency infrastructure costs by number of sites). PG&E vehicles per site from TURN-01, question 1, SB 0 Program Master Program Cost Summary, tab Fleet Ready Deployment Plan. Data for SCE site cost and per site vehicle estimates from TURN-01, question 01, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs. PG&E Testimony, p. 1-1, lines -. As TURN has pointed out in the past, the fact that a site is located in a disadvantaged community does not mean it is financially disadvantaged. TURN expects that large corporations will be a large recipient of the subsidies at hand; many likely may have distribution centers, warehouses, etc. in disadvantaged communities. ChargeReady Advisory Board Meeting Presentation, May 1, 01, p..

25 $1 million of the allocated budget should be reserved exclusively for the deployment of electric busses. While a relatively nascent sector, the technology has advanced rapidly in recent years and commercial products are available to transit agencies. Electric busses have the ability to disproportionately reduce emissions in DACs, and given the amount of gasoline they displace from relatively inefficient diesel busses and the amount of e-miles driven, will likely be among the most cost-effective and meaningful investments utilities can make. Further, electric busses represent one of the few opportunities low and moderate-income individuals have to enjoy the emission-free and quiet travel offered by EVs, particularly in contrast to the lightduty sector which currently skews to wealthy single family home owners. Pending data collection from Phase 1, TURN expects investment in electric busses to be among the most valuable, cost-effective, and impactful made by utilities through these programs, and the sector is well-aligned with TURN s emphasis on decreasing emissions in DACs. c. TURN Recommends Rebates for Electric Forklifts Rather than Infrastructure Subsidies It is important that the Commission and utilities look beyond capital infrastructure subsidies, where appropriate, to provide for more cost-effective solutions. Rather than subsidized capital infrastructure, utilities should support greater electrification of the forklift sector through rebates of $,000 per forklift. There are two primary reasons for this: 1) Electric forklifts are more technologically mature than other non-light-duty EVs. ) A rebate program is likely to be equally or more effective than subsidized infrastructure because the primary barriers to greater electric forklift adoption are not only financial. On the latter point, EPRI states in its report for PG&E that education regarding the capabilities of electric forklifts are a central barrier to greater adoption. Education may overcome many barriers for customers and allow for faster adoption of electric units. A full understanding of customer needs and any applicable limitations of electric units is necessary to be sure that electric forklifts are a viable and beneficial option. Although the upfront costs of electric forklifts are in general higher than IC units, immediate savings are available through reduced maintenance and fuel costs. As noted in the ACR at page 0: utilities should not over-invest in utility-owned TE infrastructure if instead they could support the private sector or individuals in making these investments, while still receiving adequate compensation for their contributions to TE. Given that most of the costs of the program, even under TURN s alternative Phase 1 budget, are capital, the utility has sufficient incentive. A forklift rebate program is one way to make the program more cost-effective. EPRI, p. -1.

26 EPRI goes on to recommend an emphasis on education for PG&E s program: PG&E can help to increase adoption of electric forklifts by increasing end-use awareness, providing technical support to customers and dealers, and offering incentives. Leveraging existing communication channels between technology manufacturers, dealers, and end users is an important aspect during program implementation. Following are some key ideas to increase the market base for electric forklifts: Train and educate account managers about the details of the program along with program goals. Train and educate employees about the program to help facilitate communication of the message to customers and end users. Leverage existing relationships between account managers and key C&I accounts to help with scheduling of meetings and presentations with customers and targeted end users. EPRI lists a range of free online tools available to PG&E for these efforts. The report also summarizes several utility programs in other states, all of which offer forms of financial rebatesnone offer subsidized infrastructure. Rebates for these programs range from $0 - $,000, and $,000 is used in EPRI s High Incentive scenario. TURN thus recommends EPRI s suggestion of $,000 per forklift for its High Incentive scenario to aggressively accelerate TE in this sector. TURN recommends the utility also attempt to ensure participants in the program are converting class / forklifts to class 1 if not, the participant will receive a subsidy for something they would have done anyway, resulting in zero incremental GHG benefits and wasted ratepayer funds. 0 Since a rebate program is more cost-effective than subsidized infrastructure TURN s recommendation results in greater electrification at less cost. d. Fundamental Data Collection During Phase 1 Will Guide Future Programs to Cost- Effectively Accelerate Transportation Electrification Given the nascent state of the market and historic lack of ratepayer subsidy in the non-lightduty sector, Phase 1 data collection is critical to the ongoing and future success of TE programs. The data collection metrics proposed by TURN, in conjunction with metrics proposed by the utilities and other party recommendations, should directly inform future phases of utility TE EPRI Report, p. -0. EPRI Report, p While this issue exists for other sectors (e.g. a participant could apply for EV subsidies that it would have purchased anyway), whether or not an investment is truly incremental seems more of a concern for the forklift sector, where the technology is more prevalent than medium-duty or heavy-duty vehicles.

27 programs so that they maximize the GHG and criteria pollutant reductions of each dollar spent with particular emphasis on reducing NOx and particulate emissions in DACs. Ratepayer funds can be more effectively deployed in the future to both reduce costs and emissions. At a minimum, TURN recommends the following areas of data collection during Phase 1: Conduct a bottoms-up analysis of each non-light-duty sector to provide a more granular estimate of the number of vehicles that can reasonably be forecast and displaced with EVs particular to each utility territory. As stated above, somewhat generic and wide-ranging statewide estimates are used to derive most of the utility vehicle forecasts. Ideally, forecasting and conventional vehicle inventory methodologies among utilities would be similar or the same among the utilities for comparison purposes. With Phase 1 data collection, PG&E and SCE should estimate the number of conventional vehicles in their territory per sector and provide a reasonable economic and technological analysis of the number of vehicles that can be electrified over the next five-year period. A reality check on this should be uptake in the Phase 1 program. PG&E demonstrates that such an analysis is possible with its more bottoms-up approach to the transit bus, airport GSE, port cargo handling equipment, TSE, and forklift sectors TURN simply recommends similar approaches for all sectors to derive more meaningful electrification forecasts. Calculate relative cost-effectiveness of each site that receives subsidy. This should be done by calculating avoided GHG, NOx, and particulate matter emissions based on the conventional vehicle replaced (e.g. diesel bus), miles driven per year, relative fuel efficiency (gas vs. electric) and emissions profile of the electric mix at the times vehicles charge. Along with a reasonable projection of these elements, this should allow for a per vehicle and site estimate of avoided GHG and particulates, which can be compared to the cost of the site (TURN expects this will be tracked by the utilities). TURN also recommends Energy Division begin a process to standardize inputs and reporting formats for emissions reductions across utilities, as well as develop a maximum cost-effectiveness threshold per TURN s Performance Accountability Metric recommendation. More detailed understanding of what investments best reduce particulate matter in DACs. Granular emissions tracking is complicated, particularly for mobile sources. TURN recommends utilities work with CARB and other relevant agencies to better understand the impact of various investments/sectors on local emissions. It is possible that some results will be counter-intuitive, for example local emissions reduction in areas outside of where investments are made. This is important for how the Commission evaluates future investments. Better understanding and plans for how to mitigate the potential for future stranded costs. Many of the charging and vehicle technologies that will be invested in through the utility programs are in early stages of commercialization. Charging station standards in the medium and heavy duty sectors are generally fragmented and vary depending on the particular vehicle manufacturer. If a manufacturer or OEM (Original Equipment Manufacturer) goes bankrupt, the vehicle and related utility infrastructure will become

28 stranded. Utility equipment is installed and depreciated over very long periods (up to and over 0 years). Utility proposals do not address this risk at all; this must be rectified when future ratepayer funds are considered. Calculate ratepayer payback period of each site. Given that load is directly correlated with GHG emissions as well as relative cost-effectiveness, additional revenues (from utility bills) from EVs for each site should be calculated, forecast, and used to derive a payback period (in years) for each site and sector. Phase proposals should include ratepayer payback periods and plans on how they can be shortened (including but not limited to greater site host contributions, use of rebate programs, etc.). Understand the commercial, economic, and technical maturity of each sector and reflect this in Phase proposals. The utility proposals provide for a blanket approach to each non-light-duty EV sector, regardless of technical and economic maturity. However, there will likely be opportunities to leverage greater site host funds, propose rebate programs that are more cost-effective (as TURN has recommended here for the forklift sector), and/or identify non-financial barriers impeding adoption. Utilities should reflect a more nuanced understanding of the technical and economic features of each sector in Phase proposals to provide for more targeted funding efforts. At a minimum this should include: o The technical and commercial maturity of each sector. For example, while there are few if any commercially available heavy-duty EVs for long-distance trucking, the medium-duty sector is progressing and electric forklifts are relatively mature. It is potentially wasteful and overly complex to employ the same approach for each sector regardless of its maturity. On one end of the spectrum, forecasts of vehicles and costs where there are not expected to be electric options are wasteful and unnecessary. On the other end, large and equal subsidies for sectors that have reached or are expected to reach cost parity with the conventional vehicle equivalent may be equally unnecessary. o Financial benefits of EVs for site hosts. Given that electricity is (generally) cheaper than gasoline and EVs require less maintenance, there is a financial gain to the site host for adopting an EV. Greater site host funding of infrastructure should be leveraged where payback periods are low (or are forecasted to be low) and the incremental price between electric and conventional vehicles substantially reduced. Some sectors or particular sites may only require subsidies for utilityside infrastructure rather than utility and customer-side. o Gaps in awareness/education. While a utility is not always the best vehicle to bridge this gap, there may be particular competencies within utilities to accomplish greater awareness of alternatives, including EVs. TURN hopes any spending on education would supplant, not complement, the usual utility proposals of capital infrastructure spending. Understand factors that drive per site cost variability. Phase cost proposals should not contain the same level of variability in cost projections with experience garnered in Phase

29 Utilities should examine the primary factors that drive cost variance among sites and determine how they can be minimized with reasonable site criteria. Explore site host and sector willingness to pay. Subsidy levels set in the Phase 1 program will likely over-subsidize some participants, meaning a particular site would be willing to pay more to participate in the program. Leveraging participant and private funds can substantially alleviate ratepayer burden. The types of sites that participate will be instructive; for instance, UPS, FedEx, Amazon, and other corporations make billions of dollars in revenue yet are likely to be primary recipients of ratepayer funds. Further, better understanding through surveys of site host payback period tolerance 1 and budgeting constraints/timelines may inform how to spend ratepayer funds judiciously. V. Performance Accountability Measures as a Risk-Sharing Mechanism Are Necessary to Incent Prudent Utility Investment Senate Bill 0 states utility programs should include Performance Accountability Measures. These measures can, in part, help ensure programs seek to minimize overall costs and maximize overall benefits as also called for in SB 0. These may be viewed as risksharing mechanisms between utilities and ratepayers to help better align incentives and hold utilities accountable. These accountability measures are lacking in the utilities proposals and the programs can be improved by ensuring they minimize costs and maximize benefits with specific accountability metrics. TURN proposes a minimum dollar per greenhouse gas (GHG, for example carbon dioxide equivalent) emissions avoided metric serve as a utility accountability measure. This, along with the implementation details, should be determined by Energy Division, in conjunction with input from other state agencies (CEC, CARB) and stakeholders, using one to two years of actual Phase 1 data. Once the metric is established, if utility programs do not meet or exceed the minimum cost-effectiveness level, the utility must refund all or a portion of the earned return on equity (ROE) it earns to ratepayers. This should be assessed on a rolling basis, for example every three years. If the threshold is met, no refund is necessary. TURN believes this mechanism may provide sufficient incentive to invest reasonably during project implementation, and allows for shareholders to share some risk with ratepayers, 1 For example, managers should be able to articulate the corporation s expectation of payback period for investments, as well as whether there is any intrinsic value placed on reducing emissions (such as a sustainability plan) such that payback periods may be longer, but still supported, for green investments. There may also be lessons from the marketing arena where consumer willingness to pay studies are common. A draft paper summarizes some common techniques A Review of Methods For Measuring Willingness to Pay,

30 1 1 1 only if utilities are not able maximize emissions reductions with ratepayer investments. TURN suggests a portfolio approach to calculating the total dollar per GHG avoided statistic, which means some projects may exceed the minimum threshold but the overall portfolio must meet or exceed (be lower than) the threshold determined by Energy Division with stakeholder input. VI. SCE Must Abide by All Affiliate Transaction Rules Regarding its Stake in an Electric Bus Manufacturer A subsidiary of SCE s parent company, Edison International, owns a stake in electric bus manufacturer Proterra. SCE s proposal for ratepayer investment in MD-HD electrification which includes electric buses, may directly increase the profitability of the utility s unregulated affiliate. Accordingly, it is critical that SCE adhere to all affiliate transaction rules given the enhanced role ratepayer funds are expected to play in this sector pursuant to both TURN s recommendations and the utility proposals. SCE, See CPUC, Affiliate Rules and Holding Company Issues, available at

31 VII. Appendix 1 Eric Borden Statement of Qualifications I am presently an Energy Policy Analyst with The Utility Reform Network (TURN). I have provided testimony on behalf of TURN in prior utility applications for light-duty charging infrastructure. Prior to my position at TURN, I consulted for utilities, an inter-governmental energy agency, and an energy services company. I have also conducted research and published reports on energy sector topics. My Curriculum Vitae is detailed below. EDUCATION Master of Public Affairs, University of Texas at Austin, LBJ School of Public Affairs, 0-01 Specialization: Natural Resources and the Environment Thesis: Electric Vehicles and Public Charging Infrastructure in the United States B.S.B.A., Washington University in St. Louis, Olin School of Business, Majors: Finance, Entrepreneurship Minor: Psychology PROFESSIONAL EXPERIENCE Energy Policy Analyst February 01 Present The Utility Reform Network (TURN) Prepare testimony, conduct analyses, and represent TURN in various proceedings at the California Public Utilities Commission (CPUC) related to electric vehicle charging infrastructure, utility procurement, rate design, demand response, and rate cases. Senior Energy Analyst June 01 January 01 Thought Energy LLC, Chicago, IL Thought Energy specializes in designing, installing, and operating on-site natural gas combined heat and power (CHP) systems. Created financial models to forecast profits of potential site installations Researched state and regional public policy frameworks governing CHP Conducted analyses over electricity and natural gas price trends Developed presentations and marketing materials for investor meetings Consultant February 01 October 01 International Renewable Energy Agency (IRENA), Bonn, Germany Hired to write a report on worldwide electricity sector battery storage, including primary applications for renewable energy integration, market developments, trends, and case studies Conduct research, review literature, interview key industry players, develop case study material

32 Travel to Bonn, company sites, and research facilities Written report will be sent to policymakers in IRENA member countries German Chancellor Fellow July 01 November 01 Alexander von Humboldt Foundation, hosted by DIW Berlin, Berlin, Germany Research Project Title: Energy Storage Technology and the Large-Scale Integration of Renewable Energy Investigated the role of energy storage in Germany for renewable integration through literature review, interviews with German energy experts, and analysis comparing public policy support in Germany and the U.S. for storage technologies Invited to hold a presentation at the International Renewable Energy Storage Conference and Exhibition (IRES 01) Discussions with German businesses and governmental ministries; special visit to European Union and NATO headquarters in Brussels Attended energy conferences and workshops in Berlin Senior Consultant June 00-July 00 The Kenrich Group LLC, Chicago, IL Consulted for multiple energy utilities in legal disputes with the Department of Energy (DOE) Performed detailed research and quantitative/qualitative analysis to analyze financial impact related to construction of coal-fired power plants, liquid natural gas facilities, and other types of construction Contributed to final reports and presentations submitted in arbitration, settlement, or court of law presenting KRG s expert opinion Associate, Intellectual Property July 00 May 00 Charles River Associates, Chicago, IL Developed complex financial models including discounted cash flow, lost profit, and regression analyses to support expert reports within the context of intellectual property and financial litigation in multiple industries Created valuation models and supporting materials to value business entities Contributed to final reports and presentations submitted in arbitration, settlement, or court of law presenting CRA s expert opinion PUBLICATIONS Clean Energy Technology and Public Policy, LBJ Journal of Public Affairs, editor and contributor, 0. Electric Vehicles and Public Charging Infrastructure: Impediments and Opportunities for Success in the United States, The University of Texas at Austin, 01. Policy efforts for the development of storage technologies in the U.S. and Germany, DIW Discussion Paper, 01. Expert Views on the Role of Energy Storage for the German Energiewende, DIW Berlin and BMU Stores project, online here, 01. 0

33 Germany s Energiewende, chapter 1 in Global Sustainable Communities Design Handbook, ed. Dr. Woodrow Clark, Elsevier Press, 01. Battery Storage for Renewables: Market Status and Technology Outlook, International Renewable Energy Agency (IRENA), co-author with Ruud Kempener, 01. 1

34 VIII. Appendix Comparison of PG&E and SCE Total Forecasted Sites and Vehicles Table. Total Forecasted Number of Sites by Vehicle Segment (PG&E and SCE) Five Year Program SCE s forecasts are from 01-0, PG&E s from This does not appear to be a large driver of the difference in number of vehicles and sites. PG&E data from TURN-01, question 1, SB 0 Program Master Program Cost Summary, tab Fleet Ready Deployment Plan. Data for SCE from TURN-01, question 01, attachment MD-HD Cost Estimate Infrastructure TURN, tab aggregate costs.