MCE Board of Directors
March 19, 2026

FEBRUARY FILINGS
Docket No.: A.25-05-009
Exhibit No.:

Date: February 13, 2026
Witness: Ryan Matley

PREPARED DIRECT TESTIMONY OF RYAN MATLEY

ON BEHALF OF

THE CALIFORNIA COMMUNITY CHOICE ASSOCIATION
APPLICATION OF PACIFIC GAS AND ELECTRIC COMPANY (U 39 E) FOR

AUTHORITY, AMONG OTHER THINGS, TO INCREASE RATES AND CHARGES FOR ELECTRIC AND GAS SERVICE EFFECTIVE ON JANUARY 1, 2027

PUBLIC VERSION
i

TABLE OF CONTENTS
I. INTRODUCTION AND SUMMARY OF RECOMMENDATIONS ..................................... 1 II. THIRTEEN HYDRO FACILITIES, POWERHOUSES AND ASSOCIATED ASSETS SHOULD BE REMOVED FROM PG&E’S REQUESTED REVENUE REQUIREMENT BECAUSE THEY NO LONGER BENEFIT CUSTOMERS ............................................... 7 A. Four Hydro Facilities on PG&E’s Books Have Been Retired or Sold ............................... 9 B. Nine Hydro Powerhouses and Associated Assets on PG&E’s Books Are No Longer Used and Useful ......................................................................................................................... 10

1. Kilarc-Cow Creek .......................................................................................................... 11

2. Kerckhoff 1 .................................................................................................................... 13

3. Potter Valley .................................................................................................................. 14

4. San Joaquin #1A, San Joaquin #2, and San Joaquin #3 Powerhouses ........................... 15

5. Centerville ...................................................................................................................... 17

6. Inskip .............................................................................................................................. 18

7. Hamilton Branch ............................................................................................................ 19 C. Recommendation .............................................................................................................. 20 III. PG&E SHOULD REVINTAGE EIGHT HYDRO FACILITIES TO ENSURE PROPER COST ALLOCATION ACROSS BUNDLED AND UNBUNDLED CUSTOMERS ....... 21 A. The Commission’s Foundational PCIA Principles Should Inform Cost Responsibility for PG&E’s Proposed UOG Investments in This Case .......................................................... 23 B. The Commission’s Evolving Precedent on the Vintaging of Utility Reinvestments in UOG Should Inform Cost Responsibility for These New UOG Investments .................. 25 C. PG&E Proposes to Extend the Asset Life of Eight Hydro Facilities Due to FERC Relicensing in This Proceeding ........................................................................................ 27

1. FERC License Expiration and Relicensing .................................................................... 28

2. Asset Life Extensions via PG&E’s Depreciation Study ................................................ 29

3. Depreciation of Original Cost of Asset .......................................................................... 30 D. PG&E’s Hydro Facilities Have Experienced a Change in Function ................................ 33

1. Meeting Demands of New Net Load Curve ................................................................... 33

2. Changed Operational Performance of PG&E’s Hydro Fleet ......................................... 36

3. Reduction in Energy and Contribution to Peak Capacity During System Peak ............ 41 E. PG&E Is Proposing Significant Investment in Its Hydro Facilities ................................. 43 F. Revintaging Recommendations and Impact on the PCIA ................................................ 46 IV. PG&E SHOULD FIX TWO SIGNIFICANT DEPRECIATION ACCOUNTING ISSUES

THAT UNNECESSARILY RAISE RATES FOR CUSTOMERS AND RISK COST SHIFTS BETWEEN CUSTOMER GROUPS .................................................................... 50 A. PG&E’s Hydro Depreciation Expense Request Is Artificially Inflated ............................ 50 B. PG&E’s Method of Tracking Accumulated Depreciation Hinders Its Ability to Fairly Allocate Costs to Its Current Customer Base ................................................................... 53 V. RATEMAKING ISSUES ....................................................................................................... 57 A. PG&E’s Improper Calculation of Deferred Tax Assets for the Diablo Canyon Power Plant Increases the Electric Generation Revenue Requirement ........................................ 57 ii

1. Overview of Accumulated Deferred Income Taxes ....................................................... 58

2. DCPP Pre-Extended Operations Investments Should Have Been Fully Depreciated by 2026................................................................................................................................ 59

3. Had PG&E Taken Advantage of Accelerated Depreciation Options Available to It, There Would Likely be No DCPP DTA ........................................................................ 60

4. The Inclusion of the DCPP DTA in the GRC 2027 Violates the Consistency Rule ...... 61

5. PG&E Does Not Need to Include the DCPP DTA in the 2027 GRC Rate Base Because IRS’s Safe Harbor Provision Protects PG&E from Inadvertent Normalization Violations ....................................................................................................................... 62

6. PG&E’s Should Remove the DCPP DTA from Rates or Provide Balancing Account Treatment to Allow for a Future Refund. ...................................................................... 63 B. PG&E’s Failure to Implement Changes to the Uniform System of Accounts Required Under FERC Order No. 898 Overstates the Revenue Requirement by $18.8 Million ..... 64 VI. PRODUCTION O&M EXPENSE FORECAST .................................................................... 67 A. PG&E’s Forecasts Overestimate Its Expense for Hydro Operators ................................. 68 B. PG&E Must Include in Its Revenue Requirement Cost Savings Realized from Its Asset Management System Investments ..................................................................................... 71 VII. COST ALLOCATION ISSUES .......................................................................................... 73 A. PG&E’s Allocation of Common, General, and Intangible Plant Violates the Requirements of FERC Order No. 898 and Results in Improperly Allocated Costs to the Functional Revenue Requirements ..................................................................................................... 73 B. PG&E’s Outdated Fees for Performing Billing Services on Behalf of CCAs Are Resulting in CCA Customers Paying More to PG&E than the Cost to Serve Those Customers ......................................................................................................................... 77 VIII. HYDRO LICENSING BALANCING ACCOUNT EXPANSION ..................................... 79 A. PG&E’s Expansive Use of Its Hydro Licensing Balancing Account Is Unwarranted and Reduces the Incentive for PG&E to Control Costs ........................................................... 79

ATTACHMENTS
Attachment RAM-1: Curriculum Vitae of Ryan Matley
Attachment RAM-2: Select Discovery Responses
Attachment RAM-3: Revenue Requirement by Hydro Facility Attachment RAM-4: Depreciation Expense by Hydro Facility Attachment RAM-5: Capital Expense Forecast by Hydro Facility Attachment RAM-6: O&M Expense Forecast by Hydro Facility Attachment RAM-7: Cumulative Gross Hydro Investment in Real Dollars Attachment RAM-8: Hydro Unit Capacity Contribution
Attachment RAM-9: Reallocation of CGI Plant for FERC Order 898 Attachment RAM-10: Common, General, Intangible Plant Allocations Attachment RAM-11: Hydo Licensing Balance Account Capital by Category Attachment RAM-12: Financial, Operational, and Regulatory Summary of Proposed Revintaged Hydro Facilities
1 1 I. INTRODUCTION AND SUMMARY OF RECOMMENDATIONS
The California Community Choice Association (“CalCCA”) presents this Direct 2 Testimony in the Application of Pacific Gas and Electric Company (U 39 E) for Authority, Among 3 Other Things, to Increase Rates and Charges for Electric and Gas Service Effective on January 1, 4 2027 (“Application”). This testimony was prepared on behalf of CalCCA by Ryan Matley, Senior 5 Manager at NewGen Strategies and Solutions, LLC (“NewGen”). Mr. Matley’s qualifications are 6 set forth in Attachment RAM-1. 7
CalCCA represents 24 community choice aggregator (“CCA”) electricity providers in 8 California, including 11 that are located in Pacific Gas and Electric Company’s (“PG&E”) service 9 territory. 1 CCA customers receive generation services from their local CCA and receive 10 transmission, distribution, billing, and other services from their investor-owned utility (“IOU”). 11 As such, CCA customers in PG&E’s service territory pay the same electric distribution, 12 transmission, and non-bypassable charges (“NBC”) as PG&E’s bundled customers, including the 13 Power Charge Indifference Adjustment charge (“PCIA”). The PCIA charge—which recovers the 14 cost of procurement on behalf of bundled customers and allocates to unbundled customers the 15 above-market costs of resources procured on their behalf—is derived from the Electric Generation 16 revenue requirement set in Phase I General Rate Cases (“GRC”) such as this one, as well as from 17 other generation cost recovery proceedings. This charge significantly impacts CCAs’ competitive 18 position as IOU-alternative generation providers. 19
1 California Community Choice Association represents the interests of 24 community choice electricity providers in California: Apple Valley Choice Energy, Central Coast Community Energy, Clean Energy Alliance, Clean Power Alliance, CleanPowerSF, Desert Community Energy, East Bay Community Energy, Energy For Palmdale’s Independent Choice, Lancaster Energy, Marin Clean Energy, Orange County Power Authority, Peninsula Clean Energy, Pico Rivera Innovative Municipal Energy, Pioneer Community Energy, Pomona Choice Energy, Rancho Mirage Energy Authority, Redwood Coast Energy Authority, San Diego Community Power, San Jacinto Power, San José Clean Energy, Santa Barbara Clean Energy, Silicon Valley Clean Energy, Sonoma Clean Power, and Valley Clean Energy. 2 In this 2027 GRC, PG&E has proposed a substantial increase to the Electric Generation 1 functional revenue requirement. The 2027 forecasted Electric Generation revenue requirement of 2 $1,507 million2 reflects a $283 million increase over the 2026 forecasted Electric Generation 3 revenue requirement of $1,224 million3—a 23 percent increase in a single year. The 2027 4 forecasted Electric Generation revenue requirement reflects an even larger $352 million increase, 5 or 30 percent, over the 2024 actual Electric Generation revenue requirement of $1,155 million.4 6 This increase is primarily the result of PG&E’s significant proposed investment in its hydro 7 facilities in this case. PG&E is planning capital investment that will, in just six years’ time, more 8 than double the rate base associated with its hydro facilities, many of which have been in operation 9 for up to and in some cases over 100 years. PG&E is also forecasting a 35 percent increase in 10 operational expenses for its hydro facilities as compared to 2024 actuals, despite the fact that it has 11 plans to retire or sell multiple facilities. 12
CalCCA recommends that the California Public Utilities Commission (“Commission”) 13 make the following adjustments to PG&E’s request in this case: 14 • Remove thirteen hydro facilities that erroneously remain on PG&E’s books and 15 that are no longer used and useful in providing generation services for the benefit 16 of customers from the Electric Generation revenue requirement, while recovering 17 any remaining balance of the original investment through a regulatory asset, with 18 no return on rate base; 19
• Move eight hydro facilities from the Legacy Utility-Owned Generation (“UOG”) 20 vintage to the 2027 vintage for purposes of PCIA cost recovery to reflect the fact 21 2 Exh. PG&E-10, App. A, p. AppA-18 (Table 17) (Nov. 15, 2025). 3 Id. at 1-1 (Table 1-1).
4 Exh. PG&E-10 Workpapers at WP 14-51 Line 1, Columns (B) and (E). 3 that the substantial proposed investments in these assets of over $1.6 billion would 1 result in a significant overhaul of these facilities, supporting life extensions as well 2 as changed functions for these assets; 3
• Reduce the depreciation expense in the 2027-2030 forecasted revenue requirement 4 for hydro facilities by $16 million annually to account for the actual book 5 accumulated depreciation as reported by PG&E; 6
• Order PG&E to begin recording all elements of its hydro facilities’ capital 7 accounting at the hydro facility level, not just at the Federal Energy Regulatory 8 Commission (“FERC”) account level; 9
• Remove the Deferred Tax Asset (“DTA”) associated with the Diablo Canyon 10 Power Plant (“DCPP”) from the Electric Generation revenue requirement since it 11 was incorrectly accounted for and is not required to be included to avoid a potential 12 normalization violation with the Internal Revenue Service (“IRS”); 13 • Reclassify assets according to FERC Order 898 so that reclassification is reflected 14 in the 2027–2030 GRC-approved revenue requirement, particularly since Order 15 898 required this to be complete by January 1, 2025; 16 • Reduce PG&E’s Production Operations and Maintenance (“O&M”) forecast by 17 $34.8 million annually in 2027-2030 because (1) PG&E has not justified the need 18 for its proposed substantial increase in hydro operators; and (2) PG&E has not 19 forecasted any offsetting savings from its increased asset and risk management 20 investments, despite recent experience demonstrating that substantial savings are 21 possible; 22
4 • Reject PG&E’s request to make a one-off change in allocation for the hydro license 1 electric intangible assets without a comprehensive study of allocation changes for 2 all Common, General, and Intangible (“CGI”) assets; 3
• Require PG&E to cap its revenues associated with CCA/Direct Access (“DA”) 4 billing services at the cost included in its forecast for such services and refund any 5 revenue received above that revenue cap to CCA/DA customers through the Energy 6 Resource Recovery Account (“ERRA”) proceeding; and 7
• Limit the use of the Hydro Licensing Balancing Account (“HLBA”) to costs that 8 are truly unpredictable, variable, and generally outside of the utility’s control, 9 therefore: (1) disallowing HLBA recovery of the Large Uncontrolled Water 10 Release (“LGUWR”), Security, and Spillway categories of costs; and (2) limiting 11 HLBA recovery of the License Conditions projects such that PG&E may only 12 recover for facilities that have not yet received their FERC license at the time of 13 filing a GRC. 14
The following table summarizes the estimated 2027 revenue requirement impact of CalCCA’s 15 16 recommendations.
5 Table RAM-1
2027 Electric Generation Revenue Requirement Impact of CalCCA’s Recommendations CalCCA Recommendation
CalCCA 2027 Electric
Generation Recommended
Revenue Requirement
Change
Walk
Forward of Electric
Generation Revenue
Requirement (in Millions)
PG&E’s Requested Revenue
Requirement
N/A $1,507.25
Remove Hydro Plants
Recorded in Error and no
Longer Used and Useful
($10.6 million) $1,496.6
Reduce Depreciation Expense
Associated with Hydro Units
($15.7 million) $1,480.9
Remove DCPP DTA from
Rates
($4.3 million) $1,476.6
Implement FERC Order 898
Reclassifications
($18.8 million) $1,457.8
Refrain from Additional One-Off CGI Allocation Changes ($36.0 million) $1,421.8 Cap Billing Service Revenue
Included in Expense Forecast
($2.5 million) $1,419.3
Reduce Forecasted
Production O&M Expenses
($34.8 million) $1,384.5
Total Recommended
Revenue Requirement
$1,384.5
($122.7 less than PG&E’s
$1,507.2 request)
1 In addition to these adjustments to the overall Electric Generation revenue requirement, 2 CalCCA’s recommendations also result in PCIA vintaging adjustments of the Balch, DeSabla-3 Centerville, Drum-Spaulding, Helms, Kerckhoff 2, McCloud-Pit, Phoenix, and Upper North Fork 4 Feather River hydro facilities. CalCCA’s revintaging recommendations do not change the overall 5 Electric Generation revenue requirement, but do impact the customer group—i.e., bundled versus 6 unbundled—from which portions of that revenue requirement are recovered via PG&E’s ERRA 7 proceedings. CalCCA recommends the revenue requirements for these eight hydro facilities be 8 reassigned from the Legacy UOG vintage to the 2027 vintage for cost recovery purposes. 9 5 Exh. PG&E-10, App. A, p. AppA-18 (Table 17) (Nov. 15, 2025). 6 The revintaging of these eight facilities will ensure that unbundled customers are not 1 required to subsidize PG&E’s proposed generation investments on behalf of its bundled customers. 2 Absent this revintaging, unbundled customers would be unfairly saddled with between $81 million 3 to $120 million of extra costs recovered via the Legacy UOG vintage over the course of the 2027-4 2030 time period governed by this GRC. That would result in departed load customers providing 5 a significant subsidy to bundled customers, contrary to state law and the Commission’s principle 6 7 of bundled customer indifference.

Table RAM-26
2026 Actual and 2027-2030 Forecast (With CalCCA’s Proposed Revintaging) – PCIA Revenue Requirement Split Between Bundled and Unbundled Customers ($ million)
2026 PCIA -
Actual
2027
Forecast –
With
Revintaging
2028
Forecast –
With
Revintaging
2029
Forecast –
With
Revintaging
2030
Forecast –
With
Revintaging
CCA/DA
Revenue
Requirement
$1,544.8 $1,463.8 $1,461.4 $1,441.3 $1,424.5
Bundled
Customer
Revenue
Requirement
($243.5) ($162.7) ($160.3) ($140.2) ($123.5)
Total PCIA
Revenue
Requirement
$1,301.2 $1,301.2 $1,301.2 $1,301.2 $1,301.2
Change in
CCA/DA
Revenue
Requirement
from 2026
PCIA Actual
N/A ($81.0) ($83.4) ($103.5) ($120.3)
8 6 These figures are calculated based on the current inputs and assumptions to the 2026 PCIA currently in effect. Some of these inputs will change through the course of the 2027–2030 implementation of the PCIA through the ERRA proceedings, but this analysis provides a snapshot at a point in time of the magnitude of the revintaging impact.
7 1 II. THIRTEEN HYDRO FACILITIES, POWERHOUSES AND ASSOCIATED ASSETS 2 SHOULD BE REMOVED FROM PG&E’S REQUESTED REVENUE
3 REQUIREMENT BECAUSE THEY NO LONGER BENEFIT CUSTOMERS
Through this GRC filing, PG&E continues to request cost recovery for four hydro facilities 4 and associated assets that it has already retired or sold. In addition, there are nine other hydro 5 facilities, powerhouses and associated assets which appear to no longer be operational and yet 6 would still earn a return through PG&E’s proposed revenue requirement, despite not being used 7 and useful for customers. 8
Foundational regulatory accounting principles dictate that a generation asset generally 9 should only remain on a utility’s books if it is used and useful for the purposes of generating energy 10 and providing capacity. That fundamental principle is also codified in the California Public 11 Utilities Code: 12
In establishing rates for any electrical . . . corporation, the commission may 13 eliminate consideration of the value of any portion of any electric . . . generation or 14 production facility which, after having been placed in service, remains out of 15 service for nine or more consecutive months, and may disallow any expenses 16 related to that facility. Upon eliminating consideration of any portion of a facility 17 or disallowing any expenses related thereto under this section, the commission shall 18 reduce the rates of the corporation accordingly . . . 7 19 Based on the current operational condition of these thirteen hydro assets, the Commission 20 should require PG&E to remove them from the revenue requirement in this case. The table below 21 22 presents a summary of the units that should be removed because they are no longer used and useful. 7 California Public Utilities Code § 455.5(a).
8 Table RAM-3
Hydro Powerhouses and Associated Assets Retired, Sold, or Otherwise Not Used and Useful
Hydro Facility Powerhouse(s) Capacity
2027 Revenue
Requirement8 Status
Coal Canyon Coal Canyon N/A $0.3 million Operationally retired in 2023; plant erroneously
left on books.

Lime Saddle Lime Saddle 2.0 MW $1.2 million Operationally retired in 2023; plant erroneously
left on books.

Kern Canyon Kern Canyon 11.5 MW $0.0 million Plant sold; not all plant assets retired.

Chili Bar Chili Bar 7.0 MW $0.0 million Plant sold; not all plant assets retired.

Kilarc-Cow
Creek
Kilarc and Cow
Creek

3.4 MW ($0.1) million Operationally retired in
2024; asset retirement
pending CAISO
determination.

Kerckhoff 1 Kerckhoff 1 25.4 MW $3.2 million Last operated in 2017, mothballed9 since 2018;
not capable of generating
electricity; retirement
planned in 2028.

Potter Valley Potter Valley 9.2 MW $5.2 million Last operated in 2021, mothballed since 2023;
not capable of generating
electricity; no plans to
return to service;
retirement planned in
2028.

Crane Valley San Joaquin
#1A, #2, and #3

7.4 MW $7.2 million #1A last operated in 2016;
#2 last operated in 2023;
#3 last operated in 2017;
no firm plans to return any
to service.
8 See Attachment RAM-3. This represents the estimated amount of revenue requirement associated with each hydro powerhouse and associated assets that should be removed from the total GRC revenue requirement.
9 See Attachment RAM-2 (PG&E’s response to DR CalCCA 013-004) (“The term mothball is an official CAISO approved status for a generating unit and describes a unit that is no longer available to the grid, while keeping the unit and interconnection facilities in place until a final decision is made regarding the future of the unit (return to service, sale, or decommission)”). 9 Table RAM-3
Hydro Powerhouses and Associated Assets Retired, Sold, or Otherwise Not Used and Useful
Hydro Facility Powerhouse(s) Capacity
2027 Revenue
Requirement8 Status
DeSabla-Centerville Centerville 6.4 MW $1.0 million Last operated in 2011, mothballed since 2021;
not capable of generating
electricity; active sale
process ongoing.

Battle Creek Inskip 8.0 MW $1.9 million Last operated in 2017; not capable of generating
electricity; diversion dam
removed.

Hamilton
Branch
Hamilton
Branch

4.8 MW $0.6 million Last operated in 2018;
active sale process
ongoing.

Total 85.1 MW $20.5 million
1 2 A. Four Hydro Facilities on PG&E’s Books Have Been Retired or Sold Four of the hydro facilities10 listed in Table RAM-3 erroneously remain on PG&E’s books 3 despite having been either sold or retired. Specifically, the following facilities still appear in 4 PG&E’s calculations of its Electric Generation revenue requirement: 5 • Coal Canyon; 6
• Lime Saddle; 7
• Kern Canyon; and 8
• Chili Bar. 9
PG&E admitted in discovery that the Coal Canyon11 and Lime Saddle12 facilities remain 10 in PG&E’s accounting records in error. It has committed to correcting that error in an upcoming 11 10 I will use the term “hydro facilities” to refer to the collection of dams, conveyances, powerhouses, turbines, and supporting equipment all contained under a single FERC license. It is the facilities within a FERC license to which PG&E applies the remaining life method in its depreciation study. 11 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-007). 12 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-006). 10 errata filing.13 In addition, there is a small amount of net plant attributable to the Kern Canyon and 1 Chili Bar hydro facilities that remains on PG&E’s books. 14 The Kern Canyon facility was 2 mothballed in 2017 and sold in 2020.15 The Chili Bar facility was sold in 2021, and PG&E filed 3 final accounting information from the sale with the Commission on September 20, 2023, which 4 showed an adjustment to remove gross plant and accumulated depreciation from rate base.16 5 PG&E has admitted in discovery that the net plant attributable to Kern Canyon and Chili Bar 6 remains on its books erroneously and that it plans to update that in a future Results of Operations 7 (“RO”) model.17 8
9 B. Nine Hydro Powerhouses and Associated Assets on PG&E’s Books Are No 10 Longer Used and Useful
In addition to the accounting errors identified at the Coal Canyon, Chili Bar, Lime Saddle, 11 and Kern Canyon facilities, there are nine other hydro powerhouses and associated assets that are 12 non-operational, 18 per PG&E’s own admissions, and that PG&E is either (1) planning to 13 decommission, (2) actively in the process of selling, or (3) not actively planning to return to 14 15 service. Table RAM-4 below summarizes the status of these hydro assets. 13 Attachment RAM-2 (PG&E’s response to DR CalCCA 012-005) indicating the correction was not made in the Nov. 11, 2025, errata filing, but would be completed later in the GRC cycle. 14 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-001) showing $6,898 of net plant attributable to Kern Canyon in FERC account 335 as of December 31, 2024. 15 Exh. PG&E-5 Workpapers at WP 3-145.
16 PG&E’s Advice Letter 7031-E, filed September 20, 2023. 17 Attachment RAM-2 (PG&E’s response to DR CalCCA 010-007); Attachment RAM-2 (PG&E’s response to DR CalCCA 010-012).
18 For purposes of this testimony, I define “non-operational” as facilities that have not generated electricity in the past nine or more consecutive months. This standard is consistent with the standard laid out in California Public Utilities Code § 455.5(a).
11 Table RAM-4
Hydro Assets No Longer Used and Useful
Hydro Powerhouses and
Associated Assets Status
Kilarc and Cow Creek No longer operational per PG&E’s own admissions and its energy and capacity
data;19 decommissioning to begin during
the 2027–2030 GRC period.

Kerckhoff 1 Last operated in 2017; decommissioning
planned during the 2027–2030 GRC
period.

Potter Valley Has not operated since 2021;
decommissioning planned in future.

San Joaquin #1A, San Joaquin #2, and
San Joaquin #3 powerhouses, part of the
Crane Valley FERC license
Have not operated since 2016 (#1A),
2023 (#2), and 2017 (#3); no firm plans
to return to service.

Centerville powerhouse, part of the
DeSabla-Centerville FERC license
Has not operated since 2011; active sale
process ongoing.

Inskip powerhouse, part of the Battle
Creek FERC license
Has not operated since 2017;
decommissioning planned in future.

Hamilton Branch Has not operated since 2018; active sale process ongoing.
1 Each of these assets is discussed in more detail in the subsections below. As these assets 2 are no longer used and useful, they should be removed from PG&E’s revenue requirement request 3 in this case. 4
5 1. Kilarc-Cow Creek
The Kilarc-Cow Creek hydro project is a 5 megawatt (“MW”) facility20 consisting of the 6 Kilarc powerhouse, originally constructed in 1903,21 and the Cow Creek powerhouse, originally 7 constructed in 1907.22 Both units were retired by PG&E on October 17, 2024.23 According to 8 19 PG&E provided the annual energy generation from each of its hydro powerhouses over the 2022–2024 time period (see Attachment RAM-2 (PG&E’s response to DR CalCCA 006-001, Attach. 1)), as well as the monthly peak capacity contribution of each powerhouse over the 2016–2025 time period (see Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1)). 20 Exh. PG&E-5 Workpapers at WP 3-299.
21 Id. at WP 3-131 line 18.
22 Id. at WP 3-131 line 19.
23 Id. at WP 3-131 lines 18-19.
12 PG&E, it has not removed these units from its plant accounts because it has not yet received the 1 California Independent System Operator’s (“CAISO”) approval to permanently remove these 2 generation units from service.24 PG&E has indicated that only after it receives that approval will 3 it retire those units from its financial accounting. Until that time, PG&E plans to allow Kilarc-Cow 4 Creek to remain in the 2027–2030 revenue requirement. 5 This proposed approach to regulatory accounting is unreasonable and inconsistent with the 6 overwhelming factual evidence confirming that this asset is functionally retired. Kilarc 7 powerhouse last generated electricity in May 2019 due to damage to the conveyance/canal and has 8 been officially mothballed with CAISO since June 30, 2021. 25 Cow Creek powerhouse last 9 generated electricity in March 2023 due to storm damage to the headgate.26 According to PG&E, 10 both units would require significant investment to return to service, and the associated work would 11 take months to complete and would not be cost-effective to perform.27 12 In addition to the operational evidence suggesting that Kilarc-Cow Creek is functionally 13 retired, PG&E has also taken substantial regulatory steps to effectuate the retirement. PG&E filed 14 its FERC license surrender application for Project No. 606 on May 12, 2009.28 The surrender order 15 was issued on October 17, 2024, the same date that PG&E indicated it officially retired the units. 16 Therefore, PG&E has been actively working toward retiring these units for the past 16 years. With 17 respect to CAISO approval, the timing of the request to CAISO to permanently retire the units is 18 fully within PG&E’s control. As of November 13, 2025, PG&E forecasted submitting that 19 retirement request in November 2025 with a requested effective date of December 31, 2025.29 20 24 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-005). 25 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-001). 26 Id.
27 Id.
28 Exh. PG&E-5 Workpapers at WP 3-146.
29 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-001). 13 In light of all this operational and regulatory evidence, it is clear that Kilarc-Cow Creek is 1 no longer used and useful. To argue against the obvious conclusion that Kilarc-Cow Creek should 2 be removed from the 2027 GRC revenue requirement, PG&E points to one fact: that CAISO has 3 not yet approved the asset’s permanent retirement. This one fact should not outweigh all the others 4 demonstrating the status of the asset, especially given that the timing of making this request to the 5 CAISO is fully within PG&E’s control. 6
As of the end of 2026, PG&E will have collected a return on the Kilarc and Cow Creek 7 invested value for seven years and three years, respectively, since each powerhouse last operated. 8 PG&E plans to extend that return for another four years through this GRC cycle.30 This proposal 9 is unreasonable because it would result in PG&E collecting a return on the Kilarc and Cow Creek 10 powerhouses for an eleven-year period and a seven-year period, respectively, during which time 11 the assets will not be used and useful. 12
13 2. Kerckhoff 1
The Kerckhoff hydro facilities consist of two powerhouses: Kerckhoff 1 and 2. Kerckhoff 14 1 is a 25.4 MW powerhouse with two turbine units (Unit 1 and Unit 3) that began operations in 15 1920.31 Kerckhoff 2 is a 155.0 MW powerhouse with one turbine unit that began operations in 16 1983.32 Both powerhouses are licensed by FERC under a single license, Project No. 96, which was 17 granted on November 8, 1976.33 PG&E applied for a Final License Application to extend the 18 FERC license for the Kerckhoff 2 unit in November 2020, and as part of that application, it filed 19 30 See Attachment RAM-2 (PG&E’s response to DR CalCCA 002-005). 31 Exh. PG&E-5 Workpapers at WP 3-133 line 82.
32 Id. at WP 3-133 line 83.
33 Id. at WP 3-146.
14 a Proposed Retirement Plan for the Kerckhoff 1 unit.34 PG&E currently anticipates receiving an 1 order on this FERC application in December 2026.35 2
Units 1 and 3 at the Kerckhoff 1 powerhouse last operated in 201736 and were mothballed 3 in September 2018.37 The Kerckhoff 1 unit last contributed capacity during the peak hour on 4 PG&E’s system on June 22, 2017.38 The Kerckhoff 1 powerhouse is physically incapable of 5 generating electricity due to several inoperable components that are uneconomic to restore, and 6 PG&E has no plans to return it to service.39 7
Despite the fact that the Kerckhoff 1 powerhouse has not operated since 2017, as of the 8 9 end of 2026, PG&E will have collected a return on its invested value for nine years since the asset last operated. PG&E plans to extend that return for another four years through this GRC cycle. 10 This proposal is unreasonable because it would result in PG&E collecting a return on the Kerckhoff 11 1 asset for a 13-year period, during which time the asset will not be used and useful. 12 13 3. Potter Valley
Potter Valley is a three-unit powerhouse with a total capacity of 9.2 MW that began 14 operations in 1908.40 PG&E determined that the project was no longer economically viable to 15 operate in 2019 and began the formal license surrender process with FERC at that time.41 The 16 FERC license for the facility, Project No. 77, expired on April 4, 2022,42 and PG&E has been 17 operating under annual FERC license extensions since then.43 PG&E estimates that FERC may 18 34 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-004). 35 Exh. PG&E-5 Workpapers at WP 3-146.
36 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-004). 37 Exh. PG&E-5 Workpapers at WP 3-145.
38 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1). 39 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-004). 40 Exh. PG&E-5 Workpapers at WP 3-132 line 4.
41 Id. at WP 3-374.
42 Id. at WP 3-146.
43 Id. at WP 3-374.
15 issue a license surrender order in 2028.44 However, PG&E has no plans to remove the Potter Valley 1 revenue requirement from the Electric Generation revenue requirement calculated for the 2027-2 2030 GRC period.45 3
Potter Valley powerhouse is no longer operational. It has been out of service since July 4 2021, when its GSU46 transformer failed to pass inspection testing, 47 and it last contributed 5 capacity toward PG&E’s monthly peak demand on April 29, 2021. 48 The Potter Valley 6 powerhouse does not have the physical capacity to generate electricity and PG&E has no plans to 7 return it to service.49 8
As of the end of 2026, PG&E will have collected a return on Potter Valley’s invested value 9 for five years since it last operated. PG&E plans to extend that return for another four years 10 through this GRC cycle. This proposal is unreasonable because it would result in PG&E collecting 11 a return on the Potter Valley asset for a nine-year period during which time the asset will not be 12 used and useful (and for approximately two years after the asset’s expected license surrender). 13 14 4. San Joaquin #1A, San Joaquin #2, and San Joaquin #3 Powerhouses San Joaquin #1A is a 0.4 MW single turbine unit powerhouse that was placed into service 15 in 1919.50 San Joaquin #2 is a 3.2 MW single turbine unit powerhouse that was placed into service 16 in 1917.51 San Joaquin #3 is a 4.2 MW single turbine unit powerhouse that was placed into service 17 in 1923.52 The three powerhouses are operating under one FERC license, Project No. 1354, which 18 44 Attachment RAM-2 (PG&E’s response to DR CalCCA 010-008). 45 Id.
46 This is likely an acronym for Generation Step-Up transformer, but it was not spelled out in PG&E’s response to DR CalCCA 013-005.
47 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-005). 48 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1). 49 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-005). 50 Exh. PG&E-5 Workpapers at WP 3-133 line 80.
51 Id. at WP 3-133 line 79.
52 Id. at WP 3-133 line 78.
16 also includes the Crane Valley and A.G Wishon powerhouses. This license was received on 1 September 16, 2003, and is set to expire on August 31, 2043.53 2 The San Joaquin #1A, San Joaquin #2, and San Joaquin #3 powerhouses are no longer 3 operational. The San Joaquin #1A powerhouse last operated in December 2016, the San Joaquin 4 #2 powerhouse last operated in February 2023, and the San Joaquin #3 powerhouse last operated 5 in February 2017.54 The San Joaquin #2 powerhouse last contributed to PG&E’s monthly peak 6 demand on December 15, 2022, while the San Joaquin #3 powerhouse last contributed to PG&E’s 7 monthly peak demand sometime prior to January 2016.55 The San Joaquin #1A powerhouse shares 8 a meter with the A.G. Wishon unit, and as such its peak contribution cannot be separately 9 determined.56 10
PG&E could restore the San Joaquin #1A powerhouse to service if economics or 11 operational needs change,57 but it has not stated any plans to do so. PG&E is planning to return the 12 San Joaquin #2 powerhouse to service, but that return to service is pending budget allocation to 13 repair the culvert failure that resulted in the forced outage of the unit.58 PG&E has not clearly 14 shown in its list of forecasted capital projects that it has allocated that budget in the 2027-2030 15 forecast. Finally, the San Joaquin #3 powerhouse ceased operation due to dam integrity issues and 16 was placed in mothball status with the CAISO on July 30, 2021. 59 In order to return the 17 powerhouse to service, PG&E would need to complete a dam assessment and evaluate whether 18 the dam repairs plus needed replacements of several powerhouse components would be economic 19 53 Id. at WP 3-146.
54 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-006). 55 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1). 56 Attachment RAM-2 (PG&E’s response to DR CalCCA Oral001-001). 57 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-006). 58 Id.
59 Id.
17 to complete. It has not provided a timeline for that evaluation nor firm plans to complete that 1 evaluation.60 2
As of the end of 2026, PG&E will have collected a return on the San Joaquin #1A 3 powerhouse for ten years, on the San Joaquin #2 powerhouse for three years, and on the San 4 Joaquin #3 powerhouse for nine years since these powerhouses last operated. PG&E plans to 5 extend that return for another four years through this GRC cycle.61 This proposal is unreasonable 6 because it would result in PG&E collecting a return on the San Joaquin #1A, #2, and #3 7 powerhouses for periods of fourteen, six, and thirteen years, respectively, during which periods 8 the assets will not be used and useful. 9
10 5. Centerville
Centerville is a 6.4 MW two-unit powerhouse that was placed into service in 1900.62 The 11 powerhouse is operating under a FERC license, Project No. 803, which also includes the DeSabla 12 and Toadtown powerhouses. This license expired in 2009, and PG&E has been operating on an 13 annual license extension since that time.63 PG&E has been in the relicensing process with FERC 14 since prior to the license’s expiration in 2009 and currently expects to receive relicensing approval 15 in June 2028.64 16
The Centerville powerhouse is not operational. It has been out of service since February 17 2011 due to penstock integrity issues65 and was placed in mothball status with the CAISO in June 18 2021.66 The Centerville powerhouse is not physically capable of generating electricity, and PG&E 19 60 Id.
61 Exh. PG&E-10 Workpapers at WP 9-36.
62 Exh. PG&E-5 Workpapers at WP 3-131 line 26.
63 Id. at WP 3-215.
64 Id. at WP 3-146.
65 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-007). 66 Attachment RAM-2 (PG&E’s response to DR CalCCA Oral001-001). 18 has no stated plans to return it to service.67 Instead, PG&E is actively working to sell the DeSabla-1 Centerville project.68 In addition, PG&E intends to surrender the FERC license for the DeSabla-2 Centerville project (presumably if it is unsuccessful in selling the project).69 3 As of the end of 2026, PG&E will have collected a return on the Centerville powerhouse 4 for 15 years since it last operated. PG&E plans to extend that return for another four years through 5 this GRC cycle. This proposal is unreasonable because it would result in PG&E collecting a return 6 on the Centerville powerhouse for a 19-year period during which time the asset will not be used 7 and useful. 8
9 6. Inskip
Inskip is an 8.0 MW one-unit powerhouse that was placed into service in 1979.70 The 10 powerhouse is operating under a FERC license, Project No. 1121, which also includes the Volta 11 #1, Volta #2, South, and Coleman powerhouses. This license expires July 31, 2026, and PG&E 12 informed FERC of its decision to not relicense the project in October 2020.71 PG&E plans to file 13 its License Surrender Application and Decommissioning Plan for FERC Project No. 1121 in May 14 2028.72 15
PG&E has separately filed a license amendment application for Inskip Removal.73 That 16 application, filed on October 22, 2022, requested permission to remove the Inskip Diversion Dam 17 67 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-007). 68 Id.
69 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-008). 70 Exh. PG&E-5 Workpapers at WP 3-131 line 16.
71 Id. at WP 3-146.
72 Attachment RAM-2 (PG&E’s response to DR CalCCA 010-019). 73 Exh. PG&E-5 Workpapers at WP 3-181.
19 due to significant safety concerns, and was approved by FERC on March 3, 2025.74 The diversion 1 dam has subsequently been removed by PG&E.75 2
The Inskip powerhouse is not operational. It has been out of service since October 2017 3 due to significant sand and erosion damage sustained as a result of a wildfire, 76 4 and it last contributed to PG&E’s monthly peak demand on September 2, 2017.77 PG&E determined that, as 5 a result of this damage, it was not economic to return Inskip powerhouse to service. 78 The 6 powerhouse has been in mothball status with CAISO since June 30, 2021, and it is not physically 7 capable of generating electricity if called upon by PG&E.79 8 9 As of the end of 2026, PG&E will have collected a return on the Inskip powerhouse for nine years since it last operated. PG&E plans to extend that return for another four years through 10 this GRC cycle. This proposal is unreasonable because it would result in PG&E collecting a return 11 on the Inskip powerhouse for a 13-year period during which time the asset will not be used and 12 useful. 13
14 7. Hamilton Branch
Hamilton Branch is a 4.8 MW two-unit powerhouse that was placed into service in 1921.80 15 The powerhouse is not FERC jurisdictional and therefore does not have a FERC license. 16 The Hamilton Branch facility is not operational. It has been out of service since June 2018 17 18 as a result of dam integrity issues; has been in mothball status with CAISO since July 30, 2021; 74 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-010, Attach. 1). 75 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-010). 76 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-002). 77 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1). 78 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-002). 79 Id.
80 Exh. PG&E-5 Workpapers at WP 3-131 line 29.
20 and is currently not capable of generating electricity if called upon by PG&E.81 The Hamilton 1 Branch facility last contributed to PG&E’s monthly peak hour on May 29, 2018.82 2 The facility is currently the subject of an active sale process, and the buyer plans to restore 3 the facility to service after acquisition.83 As a result, Hamilton Branch’s planned depreciation end 4 of life is 2029. However, PG&E has no plans to remove the Hamilton Branch revenue requirement 5 from the Electric Generation revenue requirement calculated for the 2027-2030 GRC period.84 6 7 As of the end of 2026, PG&E will have collected a return on Hamilton Branch for eight years since the facility last operated. PG&E plans to extend that return for another four years 8 through this GRC cycle. This proposal is unreasonable because it would result in PG&E collecting 9 a return on the Hamilton Branch asset for a 12-year period during which time the facility will not 10 be used and useful. 11
12 C. Recommendation
PG&E should remove all remaining plant assets associated with the Chili Bar, Kern 13 Canyon, Coal Canyon, and Lime Saddle hydro facilities from its books and records. PG&E should 14 also remove the revenue requirement associated with the Kilarc-Cow Creek, Kerckhoff 1, Potter 15 Valley, San Joaquin #1A, San Joaquin #2, San Joaquin #3, Centerville, Inskip, and Hamilton 16 Branch powerhouses from its Electric Generation revenue requirement in this case. Any remaining 17 Net Book Value (“NBV”) for these thirteen assets as of the end of 2026 should be placed in a 18 regulatory asset with no return to be recovered through this GRC over a four-year period. CalCCA 19 estimates the Electric Generation revenue requirement impact of this recommendation for all 20 21 thirteen assets in the table below:
81 Attachment RAM-2 (PG&E’s response to DR CalCCA 013-009). 82 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-006, Attach. 1). 83 Attachment RAM-2 (PG&E’s response to DR CalCCA 009-009). 84 Attachment RAM-2 (PG&E’s response to DR CalCCA 010-011). 21 Table RAM-5
Revenue Requirement Impact of Accounting Changes to Accurately Reflect Hydro Asset Status85
2027 2028 2029 2030
Revenue
Requirement
Adjustment
($20.5 million) ($20.4 million) ($20.5 million) ($22.0 million) Amortization of
2026 EOY
NBV86
$9.9 million $9.9 million $9.9 million $9.9 million
Net Change to
Electric
Generation
Revenue
Requirement
($10.6 million) ($10.5 million) ($10.6 million) ($12.1 million) 1 2 III. PG&E SHOULD REVINTAGE EIGHT HYDRO FACILITIES TO ENSURE 3 PROPER COST ALLOCATION ACROSS BUNDLED AND UNBUNDLED
4 CUSTOMERS
In this case, PG&E is proposing to invest over $1.6 billion87 in eight of its hydro facilities 5 to keep these assets in service beyond their originally anticipated lifespans and to continue the 6 ongoing project of adapting these assets to serve new functions to accommodate PG&E’s bundled 7 customers’ evolving electricity needs. 8
The Commission has indicated in past decisions that certain new, significant investments 9 in UOG assets should trigger reconsideration of the default assignment of the asset’s revenue 10 requirement to the asset’s original PCIA vintage.88 This reconsideration of an asset’s PCIA vintage 11 assignment is to be considered on a case-by-case basis, based on a fact-specific analysis of whether 12 the investment causes one of the following triggering events to occur:89 13 85 See Attachment RAM-3 for more detailed calculations. 86 2026 EOY NBV estimated at $45.2 million.
87 See Attachment RAM-5, which also shows total hydro capital investment from 2027–2030 of over $2.1 billion when including investment proposed and approved in the Helms Uprate proceeding, and 2025–2030 forecasted hydro investment of over $2.6 billion when including the Helms Uprate proceeding capital. 88 D.18-10-019, p. 135.
89 D.23-11-069, p. 511; D.18-10-019, p. 135.
22 • An extension in the asset’s useful life (i.e., the original investment in the asset is 1 fully depreciated, and the investment is used to extend the asset’s lifespan); 2 • A capacity expansion of the asset; 3
• A change in the function of the asset; and/or 4
• A large investment in the asset that leads to a “significant overhaul” of the facility. 5 Despite PG&E’s assertion in testimony that none of its proposed investments in its PCIA-eligible 6 generation assets will result in an asset life extension, a capacity addition, or a change in function,90 7 PG&E’s proposed investments in eight of these facilities implicate several of the Commission’s 8 revintaging triggers. 9
PG&E bears responsibility for providing capacity and generation service to its bundled 10 customers, while CCAs bear responsibility for providing capacity and generation service to their 11 customers who have elected to leave PG&E’s generation service. Therefore, by definition, when 12 PG&E decides to invest in an existing asset today to extend the life of the asset so that the facility 13 may continue to contribute to PG&E’s resource adequacy obligations and to serve its customers’ 14 generation needs, it is making this decision on behalf of its bundled customers. Because PG&E is 15 making these new investments to serve its current bundled customer set, these eight assets should 16 be revintaged from the Legacy UOG vintage91 to the 2027 vintage. This will ensure that bundled 17 customers remain responsible for new generation investments made on their behalf, and that costs 18 are not unfairly shifted to departed load customers. 19 90 Exh. PG&E-5 at 7-21 (Table 7-5).
91 With the exception of select customers that are exempt from PG&E’s PCIA surcharge. 23 1 A. The Commission’s Foundational PCIA Principles Should Inform Cost 2 Responsibility for PG&E’s Proposed UOG Investments in This Case The Commission has set forth clear directives regarding departing load cost responsibility. 3 In Decision (“D.”) 08-09-012, the Commission found that “[t]he law permits the recovery of 4 stranded costs from those customers who are responsible for stranded costs related to resource and 5 contractual commitments made by the IOU up until the time of the customer’s departure[,]” but 6 that “departing customers should bear no cost responsibility for such commitments the IOU makes 7 after their departure.”92 This directive helps ensure that each customer will “pay its fair share of 8 the costs the IOU incurred on [its] behalf[,]” which “is an integral part of the principles of bundled 9 customer indifference and prevention of cost-shifting.”93 10 Based on these underlying principles, the Commission relies on the PCIA mechanism to 11 recover the cost of procurement on behalf of bundled customers and allocate to unbundled 12 customers the above-market costs of resources procured on their behalf. The above-market costs 13 associated with each of PG&E’s generation resources are recovered from different subsets of 14 customers via the PCIA charge depending on when PG&E made the commitment to invest in that 15 resource and whether a customer was receiving bundled service from PG&E at that time. The 16 process of assigning resources and customers to particular cost recovery tranches is called PCIA 17 “vintaging.” 18
A distinct portfolio of generation resources is identified for each vintage year based on the 19 year in which the generation resource commitment was made. The Commission’s current approach 20 to determining the year in which the generation resource commitment was made is set forth in 21 PG&E Electric Preliminary Statement HS—for UOG resources, it is the year that the resource 22 92 D.08-09-012, p. 59 (emphasis added).
93 Id., Finding of Fact 2.
24 commitment was made, which is historically the date of Commission approval of UOG 1 construction. 94 Unbundled customers are assigned to a vintage year corresponding to their 2 departure date and are responsible for their share of above-market costs associated with generation 3 resource commitments made through the end of their assigned vintage year.95 Current bundled 4 customers are effectively assigned to the most recent PCIA vintage and are responsible for their 5 share of all PCIA-eligible resources. 6
PCIA rates are derived from the utility’s Indifference Adjustment, which is updated 7 annually in each IOU’s ERRA forecast proceeding. The Indifference Adjustment is the difference 8 in the target year between the cost of the IOU’s supply portfolio and the market value of the IOU’s 9 supply portfolio, as shown in the graphic below. 10
11 Figure RAM-1: Indifference Adjustment Calculation
12 Total Portfolio Cost includes variable power supply costs, which are determined in the 13 IOU’s annual ERRA forecast proceedings,96 plus the UOG capital investment recovery and fixed 14 maintenance costs determined in GRCs and other generation cost recovery proceedings.97 15 94 PG&E Preliminary Statement HS, Sheet 1, available at https://www.pge.com/tariffs/assets/pdf/tariffbook/ELEC_PRELIM_HS.pdf. 95 D.08-09-012, Finding of Fact 38 and Conclusion of Law 14. 96 Variable power supply costs include purchased power such as that from power purchase agreements (“PPAs”), fuel costs for UOG and PPAs with tolling agreements, and CAISO grid charges and revenues, net of any sales.
97 D.11-12-018, pp. 8–9.
25 An Indifference Adjustment is calculated for each vintage, and customers are responsible 1 for the cumulative costs included in all vintages prior to and including their assigned vintage year.98 2 The total Indifference Adjustment is collected through PCIA rates, ensuring that PG&E receives 3 full recovery of the generation-related revenue requirement approved by the Commission in the 4 GRC and any other proceedings.99 As significant new investments in PG&E’s UOG may be 5 approved in this proceeding, it is critical that any such investments be properly and equitably 6 assigned to a vintage year that accurately reflects customer cost responsibility. 7 8 B. The Commission’s Evolving Precedent on the Vintaging of Utility 9 Reinvestments in UOG Should Inform Cost Responsibility for These New 10 UOG Investments
Currently, PG&E has a general practice of assigning each PCIA-eligible UOG resource to 11 a specific PCIA vintage based on the year in which the generation resource commitment was 12 originally made (i.e., Commission approval of UOG construction) and assigning all future costs at 13 that facility, even new investments to serve current bundled customers, to the same initial 14 vintage.100 This approach requires unbundled customers to continue to pay costs associated with 15 significant changes to, and new investments in, UOG resources made after those unbundled 16 customers have departed—i.e., investment decisions that were not made on departed customers’ 17 behalf. This vintaging practice contravenes the indifference principle and the statutory prohibition 18 on cost shifting because it requires unbundled customers to continue to bear costs that PG&E 19 98 Id., p. 9.
99 Prior to D.18-10-019, the PCIA rate was set only on a forecast basis with no after-the-fact true-up for unbundled customers. D.18-10-019 approved a true-up for the PCIA using actual recorded net costs for PCIA-eligible resources and billed revenues from both bundled and unbundled customers. This true-up now occurs via the PABA, a rolling true-up between the forecasted Indifference Adjustment and the actual costs and revenues that PG&E realizes during the year related to its PCIA eligible resource portfolio. 100 PG&E Electric Preliminary Statement Part HS, Sheet 1, available at https://www.pge.com/tariffs/assets/pdf/tariffbook/ELEC_PRELIM_HS.pdf. 26 incurred solely to serve its bundled customers. Unbundled customers are thus not at all 1 “indifferent.” They are instead subsidizing bundled customers’ generation costs. 2 Under this default approach, PG&E can, for example, extend the life of a generation 3 resource beyond the original expected life—either through life-extending capital investment or via 4 an administrative process (e.g., an updated depreciation study)—but continue to assign that 5 resource to its original PCIA vintage for cost recovery purposes. As there is no limitation on the 6 number of years that a PCIA vintage may be applied to an unbundled customer or any end date 7 associated with initial vintage assignments, departed customers could be required to continue to 8 compensate PG&E for generation costs long after the original anticipated life of the resource 9 procured on their behalf. 10
The Commission’s stance on the treatment of ongoing generation-related costs has been 11 evolving in response to concerns raised by CCAs in various proceedings. In D.18-10-019, the 12 Commission recognized CCAs’ concerns regarding ongoing costs of UOG and the need to develop 13 a framework to ensure that such costs are equitably allocated. Specifically, the Commission found 14 that “[i]t is possible that new investments in an old power plant may represent such a significant 15 overhaul of the facility as to justify a ‘re vintaging’ of the facility. Likewise, it is possible that plant 16 investments for certain upgrades may justify a different vintage treatment for those investments 17 than for the underlying facility.”101 The Commission concluded that “any such analysis must be 18 fact-specific to the plants and spending in question and is better suited to a GRC evaluating such 19 spending.”102 20
In response to this directive, the CCAs took this concern to utility GRCs. In the final 21 decision issued in PG&E’s most recent GRC, D.23-11-069, the Commission again recognized the 22 101 D.18-10-019, p. 135.
102 Id.
27 need to ensure that new investments in UOG are properly categorized for PCIA vintaging and cost 1 recovery. To ensure that these issues can be analyzed in future GRCs, the Commission directed 2 PG&E: 3
…to include in its future GRC filings its position and any supporting evidence 4 concerning (1) the details of any PG&E proposal for new asset life extensions, 5 incremental capacity additions, or changed functions for any of its UOG assets and 6 why it is undertaking these changes, (2) on whose behalf it is making these new 7 investments, and (3) the appropriate vintaging treatment for each asset in light of 8 this testimony along with any future GRC proposals.103 9 The Commission has therefore implicitly acknowledged that these are the critical details 10 for any revintaging analysis triggered by new investments in UOG. These types of major proposed 11 changes to UOG—asset life extensions, incremental capacity additions, or changed functions— 12 should be treated as new resource “commitments” that trigger an evaluation of the asset’s vintage 13 assignment and whether full or partial asset revintaging is appropriate. In contrast, routine, 14 expected, or preventative maintenance types of operating and capital costs at a UOG resource 15 should by default remain tied to the initial vintage, as such costs were expected at the time that the 16 utility made the original generation resource commitment. This direction from the Commission in 17 D.23-11-069 also confirms that the key question for determining cost responsibility is the question 18 of cost causation—i.e., on whose behalf is PG&E making this new investment? 19 20 C. PG&E Proposes to Extend the Asset Life of Eight Hydro Facilities Due to 21 FERC Relicensing in This Proceeding
The Commission has indicated that an asset life extension is one of the triggering events 22 that necessitate a review of the original vintage assignment of the asset.104 While to date the 23 Commission has not defined in detail what factual circumstances constitute a “life extension” for 24 the purposes of revintaging, CalCCA submits that the following factors are highly relevant to and 25 103
D.23-11-069, p. 511.
104 See id.
28 indicative of UOG asset life extensions: (1) FERC license expiration and relicensing; (2) asset life 1 extensions as reflected in PG&E’s depreciation studies; and (3) full depreciation of the original 2 cost of the asset. 3
4 1. FERC License Expiration and Relicensing
A decision by PG&E to pursue a FERC relicensing following the expiration of an operating 5 license in order to keep the facility in service represents a new commitment by PG&E to extend 6 the service life of the facility. Relicensing often extends the license term by between 40 and 50 7 years. Generally, there is a significant financial investment associated with this decision to meet 8 new license conditions and to prepare these facilities for additional decades of service; for 9 example, PG&E has invested between $150 million and $350 million during the ongoing 10 relicensing process for three large hydro facilities.105 Importantly, these decisions to relicense are 11 not compulsory—PG&E has full discretion over whether it decides to retire, sell, or relicense its 12 facilities at the license expiration date. 13
The following table shows the FERC license expiration dates as well as the expected dates 14 of relicensing orders for the eight facilities for which PG&E is currently pursuing relicensing.106 15 105 The Upper North Fork Feather River, McCloud-Pit, and Drum-Spaulding hydro facilities had their FERC licenses expire in 2004, 2011, and 2013, respectively, and have been in the relicensing process since then (Exh. PG&E-5 Workpapers at WP 3-146). Those plants have received significant capital investment since their license expiration, with Upper North Fork Feather River receiving $350 million in plant additions from 2005–2023, McCloud-Pit receiving $148 million in plant additions from 2012–2023, and Drum-Spaulding receiving $231 million in plant additions from 2014–2023 (Attachment RAM-7). 106 Exh. PG&E-5 Workpapers at WP 3-146.
29 Table RAM-6
FERC License Expiration for Facilities Undergoing Relicensing FERC License Project Name
License
Expiration Date
Application
Filing
Deadline/Date
Filed
Estimated
Order Issuance
1061 Phoenix 8/31/2022 8/31/2020 Jun. 2025
2105 Upper NF
Feather River
10/31/2004 10/31/2002 Nov. 2025
96 Kerckhoff #1
and #2
11/30/2022 11/30/2020 Dec. 2026
2310 (Upper
Drum)/
14531 (Lower
Drum)107
Drum-Spaulding 4/30/2013 4/30/2011 May 2027
803 DeSabla-Centerville 10/11/2009 10/11/2007 Jun. 2028 2106 McCloud-Pit 7/31/2011 7/31/2009 Dec. 2026
175 Balch #1 and #2 4/30/2026 4/18/2024 Apr. 2029
2735 Helms 4/30/2026 4/19/2024 Apr. 2029
1 2 2. Asset Life Extensions via PG&E’s Depreciation Study Another indicator of an asset life extension is a regulatory accounting decision by PG&E 3 to update its depreciation study to reflect a new expected end of life date for the asset. Generally, 4 a decision by PG&E to pursue relicensing for a hydro facility is the main driver in its decision to 5 extend the facility’s accounting end-of-life. 6
The following table shows the probable retirement dates for each hydro facility going back 7 to the 2014 GRC.108 In the current GRC, the probable retirement dates for 15 hydro facilities have 8 been extended. 9
107 PG&E has proposed the split the license for Drum-Spaulding into multiple licenses. 108 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-002). 30 Table RAM-7
Hydro Facility Probable Retirement Date, 2014–2027 GRCs Retirement Date
2014
GRC
2017
GRC
2020
GRC
2023
GRC
2027
GRC
Upper NF Feather River 2047 2055 2055 2061 2075
Kilarc-Cow Creek 2014 2020 2020 2021 2024
DeSabla-Centerville 2047 2029 2029 2022 2038
McCloud-Pit 2048 2055 2055 2061 2075
Drum-Spaulding 2051 2057 2057 2063 2077
Potter Valley 2022 2030 2030 2026 2028
Phoenix 2022 2030 2030 2022 2064
Kerckhoff 1 2022 2057 2057 2027 2028
Kerckhoff 2 2022 2057 2057 2065 2076
Battle Creek 2026 2026 2026 2026 2037
Balch 2026 2026 2026 2069 2079
Helms 2026 2026 2026 2069 2079
Mokelumne 2031 2031 2031 2031 2031
Hat Creek 1 and 2 2032 2032 2032 2032 2032
Rock Creek 2031 2031 2031 2031 2034
Haas-Kings River 2041 2041 2041 2041 2041
Pit 1 2042 2042 2042 2042 2042
Pit 3, 4, 5 2043 2043 2043 2043 2043
Crane Valley 2043 2043 2043 2035 2043
Spring Gap-Stanislaus 2047 2047 2047 2047 2047
Poe 2047 2053 2053 2068 2068
Bucks Creek 2018 2055 2056 2062 2072
1 The eight highlighted facilities in the table are currently undergoing relicensing and 2 generally show the largest changes in retirement dates across the past four GRCs. But other 3 facilities have also seen changes proposed to their retirement dates in the current GRC outside of 4 the relicensing process, as PG&E adjusts to other operating or regulatory changes. 5 6 3. Depreciation of Original Cost of Asset
Finally, if an asset has achieved full depreciation, this is another indicator that PG&E has 7 recovered its original investment in the asset—i.e., its original generation commitment—and that 8 31 any further significant investment in the asset represents a new investment distinct from PG&E’s 1 original commitment. 2
The following table shows the original cost, book reserve, and NBV by hydro facility at 3 the end of 2024.109 4
Table RAM-8
Hydro Facility NBV, EOY 2024
($ millions)
Hydro Facility
Original
Cost
Book
Reserve NBV
Drum-Spaulding $596 $150 $446
Upper NF Feather River $526 $154 $373
Pit 3, 4, 5 $458 $198 $260
Helms $1,060 $870 $190
Miscellaneous Hydro $236 $57 $179
McCloud-Pit $244 $95 $149
Mokelumne $329 $202 $127
Rock Creek $298 $180 $118
Kerckhoff 2 $196 $87 $109
Poe $149 $44 $105
Spring Gap-Stanislaus $141 $50 $91
Crane Valley $203 $113 $90
Bucks Creek $113 $24 $89
Haas-Kings River $142 $63 $80
Pit 1 $108 $39 $69
Balch $78 $26 $52
Battle Creek $119 $102 $17
Potter Valley $66 $56 $10
Hat Creek 1 and 2 $23 $14 $9
Kerckhoff 1 $22 $16 $6
Hamilton Branch $8 $7 $1
Lime Saddle $16 $15 $0
Chili Bar $0 ($0) $0
Kern Canyon $0 $0 $0
Coal Canyon $6 $6 ($0)
Kilarc-Cow Creek $8 $9 ($1)
Phoenix $28 $30 ($2)
DeSabla-Centerville $91 $97 ($6)
Total $5,266 $2,704 $2,563
109 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-001, Attach. 1). PG&E indicated on January 16, 2026 it planned to provide a revised version of Attachment 1 to PG&E’s response to DR CalCCA 002- 001, but as of the filing of this testimony PG&E has not provided that revision. 32 The NBV shown in the above table is indicative of how much each facility’s total 1 investment is fully depreciated on a relative basis. It is more challenging to draw conclusions for 2 each facility on an absolute basis due to PG&E’s process for reallocating book reserves in each 3 GRC, which will be discussed in more depth in Section IV. Two of the eight facilities currently 4 undergoing relicensing and life extensions (highlighted in the above table), Phoenix and DeSabla-5 Centerville, have a NBV less than zero as of the end of 2024 which indicates the original 6 investment is fully depreciated. The remaining six (Drum-Spaulding, Upper NF Feather River, 7 Helms, McCloud-Pit, Kerckhoff 2, and Balch) have NBVs as of the end of 2024 ranging from $52 8 million to $446 million. That these facilities do not appear to be fully depreciated despite operating 9 past their initial FERC license period (and therefore initial depreciation lifetime) is a function of 10 PG&E’s accumulated depreciation allocation process110 and does not represent the cumulative 11 depreciation expense paid by existing customers. It is also a result of capital investment incurred 12 after the initial FERC license has expired,111 which is investment related to extending the life of 13 the facility. 14
In sum, for eight of PG&E’s hydro facilities, PG&E is (1) pursuing FERC relicensing, (2) 15 changing the probable retirement date for the asset in its depreciation study, and (3) has 16 cumulatively collected the original cost of the asset through depreciation expense, despite its 17 allocation procedures attributing that accumulated depreciation to other facilities. 18 110 As detailed in Section IV.B, PG&E’s allocation method has the result of decreasing the accumulated depreciation attributed to a specific facility when it extends its depreciable end of life of that facility, all else equal. As a result, once a facilities life is extended in PG&E’s depreciation study, its NBV appears to be higher than it otherwise would be.
111 For example, PG&E has placed in service $231 million in plant at the Drum-Spaulding Facility since its 2013 FERC license expiration, $349 million at the Upper NF Feather River facility since its 2004 FERC license expiration, and $148 million at the McCloud-Pit facility since its 2011 FERC license expiration (Attachment RAM-7). This investment increases the 2024 NBV of those facilities but represents investment specific to the life extension, not to the original investment. 33 1 D. PG&E’s Hydro Facilities Have Experienced a Change in Function The Commission has also indicated that a change in function in a UOG asset should trigger 2 review of the default vintage assignment for the asset.112 3 PG&E’s entire hydro fleet is serving a different function today than that for which it was originally designed. According to PG&E, this 4 fleet was originally constructed as a source of capacity and energy, and now it is a source of 5 flexibility needed to integrate high levels of renewable generation. This change has occurred over 6 approximately the last 10 to 15 years.113 This change in function reflects the evolving needs of 7 PG&E’s current customer base. PG&E is responding to those evolving needs by undertaking new 8 investments in its fleet to serve current customer demands. PG&E is investing broadly across its 9 hydro fleet, and while that investment could be a triggering event to propose revintaging of all of 10 PG&E’s hydro facilities, in this testimony CalCCA focuses on the need to revintage the eight 11 hydro facilities that reflect the most egregious examples of improper vintaging. 12 13 1. Meeting Demands of New Net Load Curve
PG&E’s hydro fleet is already demonstrating that it has responded to changing demands 14 resulting from increasing penetrations of renewable generation, and PG&E has stated that it plans 15 to invest further as those demands continue to change. In seeming contradiction to its position that 16 its generation units have not and are not planned to undergo a change in function, PG&E discussed 17 the “Changed Demands on Generating Equipment”114 as support for its large, forecasted increase 18 in capital expenditures and operating costs, particularly for its hydro units. In testimony, PG&E 19 stated: 20
With large amounts of intermittent renewables added to the grid and climate change 21 creating extreme periods of heat during the summer months and extreme rainfall 22 112 See D.23-11-069, p. 511.
113 Attachment RAM-2 (PG&E’s response to CalCCA DR 005-025); Attachment RAM-2 (PG&E’s response to CalCCA DR 005-025, Attach. 1).
114 Exh. PG&E-5 at 3-2 to 3-3.
34 from atmospheric rivers during the winter months, PG&E has experienced new 1 challenges and greater operational demands to support grid reliability and flexible 2 operation of its generation fleet, requiring assets to be operated differently than the 3 modes for which they were originally designed. These changing demands on 4 existing aging generation infrastructure requires significant investment so that the 5 CAISO can rely on PG&E’s fleet for both flexible and reliable power when called 6 upon. For example, PG&E’s hydro and natural gas resources have been called on 7 more frequently in recent years, requiring plants to adjust schedules to ramp up and 8 down to meet hourly fluctuations in electricity demand and offset short-notice 9 outages or curtailments at base-load power plants. PG&E expects continued 10 increases in expense and capital costs in its generation portfolio to address cycling-11 induced life-cycle impact to various components of its facilities.115 12 The increase of intermittent renewables has changed the operational profile, characteristics, 13 and role of hydro units in PG&E’s generation fleet. According to PG&E, the CAISO “duck curve” 14 has been getting more pronounced since 2015.116 The duck curve shows net load, which is total 15 forecasted load less variable generation, and is characterized by a trough in the middle of the day 16 as solar output is highest, followed by a steep increase in the evening hours as solar production 17 falls off and load increases. It is the role of dispatchable generation, namely hydro, natural gas, 18 and energy storage, to meet the net load in each hour of the day. The following curve from the 19 Energy Information Administration (“EIA”), provided by PG&E, shows the change in net load 20 21 across CAISO from 2015 to 2023.
115 Id. at 3-2 line 21 to 3-3 line 2.
116 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-025). 35 1 Figure RAM-2: CAISO Net Load, 2015–2023
2 PG&E’s hydro fleet’s dispatch has changed in response to the changing net load curve. 3 Data provided by PG&E117 shows the average hydro generation profile for 2024 and 2025, which 4 5 shows the hydro fleet output backing down during midday and ramping up in the evening. 117 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-025). 36 1 Figure RAM-3: Hydro Average Generation Profile – 1/1/24 to 11/5/25 2 PG&E’s hydro fleet is well suited for this role of helping to meet the net load curve and 3 therefore integrating high levels of renewable generation. According to PG&E, “PG&E’s hydro 4 generation facilities typically start up quickly, have fast ramp rates, and can easily, quickly, and 5 economically vary output in response to changing customer loads and system conditions. In 6 addition, PG&E’s hydro generating units can operate at no load or low load with much higher 7 efficiency than the alternative natural gas-fueled peaking plants.”118 8 9 2. Changed Operational Performance of PG&E’s Hydro Fleet One result of the changed demands on PG&E’s hydro fleet is that it is being dispatched 10 differently and in a way that is less efficient and creates more wear and tear on the units—a change 11 that is forecasted to lead to more operational and maintenance expense in the future. The changing 12 operational characteristics of PG&E’s fleet are illustrated in the following chart compiled by 13 118 Id.
37 PG&E. This shows that PG&E increasingly needs to deviate from the typical operating setpoint, 1 which maximizes efficiency and minimizes wear and tear,119 as shown by the trendline of daily 2 average movements away from the unit set point.120 3
4 Figure RAM-4: Daily Average Movements Off Typical Operating Set Point 5 The changing role of PG&E’s hydro fleet is further evidenced by data on the annual output 6 of hydro units from CAISO.121 An analysis of this data for large hydro generation on an annual 7 basis shows that hydro generation statewide has become more variable. 8 119 Id.
120 Attachment RAM-2 (PG&E’s response to DR CalCCA 008-012). 121 CAISO Library, Production and Curtailments data, accessed at https://www.caiso.com/library/production-curtailments-data on December 9, 2025. This data is for all hydro units in California, but since PG&E’s hydro units make up 28% of all hydro capacity in the state, it seems a safe assumption that the characteristics of statewide hydro generation would hold true for PG&E’s hydro units. CAISO provides historical generation output in five-minute increments for each generation technology.
38 The following chart shows hydroelectric output every five minutes in calendar year 2016. 1 2 The mean generation output was 2,574 MW with a minimum generation output of 941 MW, a maximum generation output of 5,909 MW, and a standard deviation around the mean of 941 MW. 3 4 Figure RAM-5: CA Large Hydro 2016 Output in 5-Minute Increments 5 In contrast, the following chart shows hydroelectric output every five minutes in calendar 6 year 2024, which shows a marked change in output variability from 2016. The mean generation 7 output in 2024 was 2,344 MW with a minimum generation output of 0 MW, a maximum generation 8 output of 5,984 MW, and a standard deviation around the mean of 1,207 MW. The standard 9 deviation—a statistical measure of variability—of hydro generation output increased by 28 percent 10 from 2016 to 2024. In addition, large hydro generation output was 0 in 131 instances in the 2024 11 39 data set, meaning that for 10 hours and 55 minutes in 2024, the entire hydro fleet in California was 1 not producing any electricity. 2
3 Figure RAM-6: CA Large Hydro 2024 Output in 5-Minute Increments 4 The mean large hydro generation was similar between 2016, when the mean was 2,574 5 MW, and 2024, when the mean was 2,344 MW. However, 2016 was categorized as a Dry water 6 year in the San Joaquin basin and a Below Normal water year in the Sacramento Valley by the 7 California Department of Water Resources, while 2024 was characterized as an Above Normal 8 water year.122 Therefore, California’s large hydro generation fleet was called upon on average at 9 122 Department of Water Resources, Chronological Reconstructed Sacramento and San Joaquin Valley Water Year Hydrologic Classification Indices, accessed at https://data.ca.gov/dataset/cdec-water-year-type-dataset/resource/105614f4-c71d-4191-b1f9-ea510afd8b62 on December 9, 2025. 40 the same level in 2024 and 2016 despite the state having a greater supply of water in 2024 which 1 would allow for more electricity production in that year as compared to 2016. The fact that large 2 hydro output was more variable in 2024 than 2016 but had the same average capacity underscores 3 the fact that in 2024 the hydro resources were being called upon to integrate renewable energy 4 while in 2016 they were relied on as a source of energy and capacity during system peaks. 5 The increasing variability in large hydro output can also be seen by looking at the standard 6 deviation around mean for large hydro production in each year from 2016 through 2025. The 7 trendline on the chart below shows increasing standard deviation over time. That trend might be 8 even more pronounced were 2021 and 2022 not Critically Dry water years. Lack of water supply 9 limited average generation in those years to 1,131 and 1,502 MWs, respectively, which reduced 10 the standard deviation as compared to years with higher mean generation since the minimum 11 generation cannot fall below zero. 12
13 Figure RAM-7: Standard Deviation of CA Large Hydro Generation, 2016–2025 14 41
1 3. Reduction in Energy and Contribution to Peak Capacity During 2 System Peak
A consequence of the changed function of PG&E’s hydro fleet is that it is therefore less 3 available to provide its original function: the provision of energy and capacity during PG&E’s 4 system peaks. The annual generation for both large hydro and small hydro across California has 5 trended downward over time as can be seen in the following chart.123 On a trended basis, large 6 hydro generation has fallen by 27 percent from 2001 to 2024 while small hydro production has 7 8 fallen by 40 percent over the same time period.
9 Figure RAM-8: CA Annual Hydro Generation, 2001–2024
10 11 This change in function can also be seen from the change in PG&E’s hydro fleet’s contribution to peak capacity during PG&E’s system peak. One of the original purposes of the 12 hydro fleet was to provide capacity at system peak, which is particularly true of the Helms plant, 13 123 California Energy Commission, Electric Generation and Capacity data for all power plants sited in CA that are greater than 1 MW, accessed at:
https://www.energy.ca.gov/data-reports/energy-almanac/california-electricity-data/electric-generation-capacity-and-energy on January 6, 2026. 42 which contains 32 percent of the peak capacity of PG&E’s entire hydro fleet.124 That plant was 1 built explicitly for capacity; 125 because it is a pumped hydro unit (which switches from an 2 electricity consumer when in pumping mode to an electricity producer when in generating mode), 3 it is a net consumer of energy. An analysis126 of PG&E’s hydro unit production during the system 4 peak127 shows that over the past 10 years, hydro’s contribution to peak has been decreasing, as 5 shown in the following chart. 6
An analysis of the coincidence factor128 for each of PG&E’s hydro units129 and for the 7 system as a whole from 2016 to 2025 shows that the coincidence factor has been decreasing. 8 9 Figure RAM-9: PG&E Hydro Fleet Coincidence Factor, 2016–2025 10 124 Exh. PG&E-5 Workpapers at WP 3-133 where it shows Helms has 1,212.0 MW of PG&E’s 3,839.6 MW of total hydro capacity.
125 Pacific Gas & Elec. Co., Project Nos. 2735 & 1988, 56 F.P.C. 2802 (1976). 126 The analysis looked at hydro contribution to peak load during the months of June through September.

That is because on average the load in those months is significantly higher than load in other months. 127 Data provided in Attachment 1 to PG&E’s response to DR CalCCA 009-006. 128 Coincidence factor is the ratio of hydro unit output during the peak hour to its nameplate capacity. A decreasing coincidence factor indicates that the hydro fleet is being called upon to deliver less capacity than it is capable of delivering at peak, most likely because other resources, particularly solar, are displacing the need for hydro capacity.
129 Hydro units that were sold or retired during the time period at issue were excluded from the analysis. 43 Of the 62 hydro units included in the analysis, 40 showed a downward trend in coincidence factor 1 and 20 showed an upward trend. The trend of hydro units decreasing their contribution to capacity 2 can be seen at Helms as well. Two of its three units have on average contributed less of their 3 available capacity at peak over the past 10 years, which is particularly notable since the facility’s 4 5 sole original purpose was to provide peak capacity.

Figure RAM-10: Helms Coincidence Factor, 2016–2025 6
7 In summary, PG&E’s hydro fleet has evolved over the past 10 to 15 years to serve different 8 functions that are distinct from those it originally served. PG&E has accomplished this shift—and 9 continues to effectuate this shift—through significant capital investments in its hydro fleet. These 10 changed functions for PG&E’s hydro fleet underscore the need to reconsider the default vintage 11 assignments for PG&E’s hydro assets. 12
13 E. PG&E Is Proposing Significant Investment in Its Hydro Facilities Finally, the Commission has indicated that, more generally, new investments in UOG may 14 in certain circumstances be substantial enough to represent a “significant overhaul” of the asset, 15 44 such that revintaging would be appropriate.130 In order to support its proposed life extensions and 1 the changing function of its hydro fleet, PG&E has requested $3.7 billion in forecasted capital 2 investment in its hydro assets from 2025 through 2030.131 That is an extraordinary amount of 3 investment by any measure. 4
PG&E’s year-end 2024 functional hydro generation gross plant was $4.9 billion. 132 5 PG&E’s year-end 2024 functional hydro generation reserve was $2.6 billion,133 which when 6 subtracted from the gross plant value results in hydro generation net plant of $2.3 billion. 7 Therefore, PG&E’s forecasted investment over the coming six years is 161 percent of the current 8 net plant value of existing hydro generation. In other words, in six years’ time, PG&E is proposing 9 to more than double the rate base of its hydro fleet—a fleet that it has built and operated for over 10 a century. This level of investment arguably amounts to a “significant overhaul”134 of PG&E’s 11 entire hydro fleet. 12
The following table shows PG&E’s proposed capital expenditures by facility. 13 130 See D.18-10-019, p. 135.
131 See Attachment RAM-5. This figure includes 2025–2026 investment which was governed by the 2023 GRC as well as Helms investment agreed to in the Helms Uprate proceeding but is shown here to provide the full scale of forecasted hydro capital investment.
132 Exh. PG&E-10 Workpapers at WP 7-48, Table 7-15.
133 Id. at WP 8-72, Table 8-28.
134 See D.18-10-019, p. 135.
45 Table RAM-9
Forecasted Capital Expenditure by Hydro Facility, 2025–2030 Hydro Facility
2025–2030
Forecasted Capital
Expenditure
Capex as a Percentage
of Original Cost
(2024 Real $)135
Helms136 $937.0 34%
McCloud-Pit $648.6 54%
Drum-Spaulding $463.2 38%
Upper NF Feather River $286.7 17%
Mokelumne $252.5 23%
Pit 3, 4, 5 $166.4 11%
Rock Creek $159.3 16%
Spring Gap-Stanislaus $155.3 46%
Balch $100.1 35%
Bucks Creek $95.7 35%
Kerckhoff 2 $86.2 17%
Haas-Kings River $84.2 17%
Poe $63.8 12%
Crane Valley $60.0 16%
DeSabla-Centerville $59.8 24%
Pit 1 $26.2 10%
Kerckhoff 1 $23.8 24%
Battle Creek $23.3 8%
135 The forecasted capital expenditure is shown as a percentage of the original cost of each hydro facility converted to real 2024 dollars. That is because some hydro facilities have experienced investment across a time period greater than 100 years, and the original cost is recorded based on expenditures occurred at the time of investment. Converting those costs to real 2024 dollars removes the effect of inflation from the comparison of capital investment today compared to investment across a long time period. Attachment RAM-7 shows the conversion of original cost to 2024 dollars. 136 There is $371.6 million of Helms capital investment forecasted in this GRC that is incremental to the $544.1 million forecasted from 2025–2030 for the Helms Uprate and Lifecycle Replacement Cost in the Helms Uprate proceeding. (Attachment RAM-2 (PG&E’s response to DR CalCCA 007-004); Attachment RAM-2 (PG&E’s response to DR CalCCA 003-017)) The total Helms investment forecasted between the GRC and the Uprate proceeding is $937.0 million.
46 Table RAM-9
Forecasted Capital Expenditure by Hydro Facility, 2025–2030 Hydro Facility
2025–2030
Forecasted Capital
Expenditure
Capex as a Percentage
of Original Cost
(2024 Real $)135
Phoenix137 $23.9 41%
Hat Creek 1 and 2 $11.3 16%
Potter Valley $9.5 5%
Hamilton Branch $3.3 14%
Total GRC & Helms
Uprate
$3,740.9 26%
1 The six-year forecast period presented in the table above results in an average capital 2 expenditure of $623.5 million per year. That compares to the $818.2 million of gross additions to 3 hydro production plant over the 2020 to 2024 time period,138 which is an average of $163.6 million 4 per year. PG&E is therefore proposing capital investment in its hydro fleet at nearly four times 5 historical levels. 6
7 F. Revintaging Recommendations and Impact on the PCIA Based on the evidence detailed above, PG&E’s proposals in this case related to eight of its 8 hydro UOG facilities should trigger PCIA revintaging for those assets. For these eight hydro 9 facilities, PG&E is (1) pursuing FERC relicensing and associated extension to each facility’s 10 useful life; (2) continuing to effectuate a change in plant function during this GRC period; and (3) 11 137 At the 2 MW Phoenix powerhouse, PG&E is proposing $23.9 million in capital investment over the 2025–2030 time period. That level of investment translates to nearly $12,000 per kilowatt (“kW”) of capacity proposed to prepare the Phoenix station to run for an additional FERC license term, which raises significant concerns about the cost-effectiveness of their plans. This is a significant expense for a plant that provides minimal capacity and energy benefit to the system. PG&E is in discussions to potentially sell the Phoenix facility (PG&E’s response to DR CalCCA 009-009). The Commission should look closely at the prudency of any capital investment made in that plant from 2025 onward, when PG&E was aware of a potential sale, in light of the purchase price achieved through the sale. PG&E’s customers should not pay for investments that a potential buyer is unwilling to pay. 138 Exh. PG&E-10 Workpapers at WP 8-356, Line No. 137 on Table 8-50. 47 proposing a significant capital investment of between 17 and 54 percent of each asset’s original 1 cost in this case. Any one of these three developments could trigger full facility revintaging; these 2 eight facilities are undergoing all three of them. 3
Specifically, the following hydro facilities should be reassigned from the Legacy UOG 4 vintage to the 2027 vintage: Phoenix, DeSabla-Centerville, Helms, Balch, Drum Spaulding, 5 Kerckhoff 2, McCloud-Pit, and Upper NF Feather River.139 The following table summarizes each 6 facility’s characteristics relative to the Commission’s revintaging criteria. 7 139 Refer to Attachment RAM-12 for a summary of each plant’s financial, operational, and regulatory history.
48 1
Table RAM-10
Hydro Facilities To Be Revintaged
Hydro
Facility
Relicensing Life
Extension?
Changing
Function?
2025–2030
Investment
Phoenix Yes, 2022 -> 2064 Yes $23.9 million;
41% of original cost
DeSabla-Centerville Yes, 2022 -> 2038 Yes $59.8 million; 24% of original cost
Helms Yes, 2069 -> 2079 Yes $937.0 million;
34% of original cost
Balch Yes, 2069 -> 2079 Yes $100.1 million;
35% of original cost
Drum-Spaulding Yes, 2063 -> 2077 Yes $463.2 million;
38% of original cost
Kerckhoff 2 Yes, 2065 -> 2076 Yes $86.2 million;
17% of original cost
McCloud-Pit Yes, 2061 -> 2075 Yes $648.6 million;
54% of original cost
Upper North
Fork Feather
River
Yes, 2061 -> 2075 Yes $286.7 million;
17% of original cost
2 CalCCA proposes that the new vintage for these facilities should be 2027, the first year of 3 requested cost recovery through this GRC. This approach is consistent with the Commission’s 4 current approach of using the year in which the resource commitment is made to determine the 5 vintage assignment. Approval of PG&E’s proposed investments in its hydro facilities would 6 trigger a new generation commitment starting in 2027, the first year that rates would be in effect 7 from this GRC. 8
The revintaging of these eight facilities will ensure that unbundled customers are not 9 required to subsidize PG&E's proposed generation investments on behalf of its bundled customers. 10 Absent this revintaging, unbundled customers would be saddled with between $81 million to $120 11 49 million140 of extra costs recovered via the UOG legacy revenue requirement over the course of the 1 2027-2030 time period governed by this GRC. That would result in departed load customers 2 providing a significant subsidy to bundled customers, contrary to state law and the Commission’s 3 4 principle of bundled customer indifference.

The following table presents the estimated 2027 revenue requirement for each of the 5 proposed revintaged facilities which are the inputs to the PCIA impact calculation. 141 6 Table RAM-11
Revenue Requirement of Proposed Revintaged Hydro Facilities Facility
2027 Rev. Req.
($ million)
Decommissioning
Cost ($
million)142
2027 Rev. Req.
to Revintage
($ millions)
Upper NF Feather River $103.4 $0.0 $103.4
DeSabla-Centerville $27.7 $9.5 $18.2
McCloud-Pit $62.1 $0.0 $62.1
Drum-Spaulding $171.5 $0.0 $171.5
Kerckhoff 2 $30.3 $0.0 $30.3
Phoenix $8.9 $0.0 $8.8
Balch $23.7 $0.0 $23.7
Helms $89.2 $0.0 $89.2
Total $516.8 $9.5 $507.3
Total Hydro UCC $1,127.7
140 These figures are calculated based on the current inputs and assumptions to the 2026 PCIA currently in effect. Some of these inputs will change through the course of the 2027–2030 implementation of the PCIA through the ERRA proceedings, but this analysis provides a snapshot at a point in time of the magnitude of the revintaging impact.
141 These calculations are shown in Attachment RAM-3 and draw on the inputs developed in Attachments RAM-4, RAM-5, and RAM-6 which all are based on data provided by PG&E through DR responses. These attachments demonstrate the calculation of a facility-by-facility revenue requirement through time, tracking by facility all elements of plant as recommended in Section IV.B. The starting point for these calculations is the breakdown of plant by facility at the end of 2024 as provided by PG&E. The revenue requirement calculations do not include the impact of the Helms Uprate Investment, which will increase the hydro revenue requirement and therefore the magnitude of revintaging cost shift. The revenue requirement calculations do account for the impact of the CalCCA recommendation to remove plants no longer used and useful in Section II, but do not attempt to estimate the impact of any of the other recommendations presented herein which would reduce the overall Electric Generation revenue requirement. 142 The $0.0 in decommission costs shown in this table for some facilities does not indicate there will be no decommissioning costs, rather it reflects that PG&E has not yet estimated a decommissioning cost nor started recovering that estimated cost from customers.
50 1 IV. PG&E SHOULD FIX TWO SIGNIFICANT DEPRECIATION ACCOUNTING 2 ISSUES THAT UNNECESSARILY RAISE RATES FOR CUSTOMERS AND RISK 3 COST SHIFTS BETWEEN CUSTOMER GROUPS
PG&E performed a depreciation study in this case in order to determine the average service 4 life, net salvage rate, and applicable aging curve for all its electric and gas assets.143 Through this 5 study, PG&E determined a depreciation rate that allows for the capital cost of utility assets to be 6 recovered from customers in equal portions over the life of the asset, ensuring that all customers 7 who benefit from an asset share cost responsibility for that asset. The proposed depreciation rates 8 for the electric generation function that are a result of PG&E’s depreciation study would increase 9 depreciation expense by $28.5 million, from $278.7 million in 2024 to a forecast of $307.2 million 10 in 2027.144 11
There are two issues with PG&E’s accounting of accumulated depreciation for its hydro 12 assets. First, PG&E’s depreciation study omits approximately $507 to $562 million of accumulated 13 depreciation as compared to what PG&E separately reports on its books, the result of which is 14 15 hydro depreciation expense that is $16 million greater than it otherwise would be. Second, PG&E does not track accumulated depreciation at the hydro facility level, only at the FERC account level 16 for the entire hydro function. This accounting approach will result in inequitable cost shifts 17 between bundled and departed load customers in the event the Commission orders PG&E to 18 revintage any of its hydro facilities—either in response to CalCCA’s proposals in this case, or at 19 any point in the future. 20
21 A. PG&E’s Hydro Depreciation Expense Request Is Artificially Inflated PG&E’s requested depreciation expense for its hydro assets is artificially inflated by 22 approximately $16 million annually. This is the result of PG&E omitting approximately $507 to 23 143 Exh. PG&E-10 at 8-5 line 6 to 8-6 line 5.
144 Id. at 8-11 (Table 8-4).
51 $562 million of accumulated depreciation from its depreciation study calculations as compared to 1 what PG&E separately reports on its books. Revising PG&E’s depreciation study to account for 2 the actual book accumulated depreciation as reported by PG&E, allocated to each plant, remedies 3 this issue and results in an annual savings of $30 million. 4 PG&E’s depreciation study uses a remaining life method to calculate the depreciation rates 5 for electric production assets. The remaining life method depreciates the net plant balance (gross 6 plant less accumulated depreciation), adjusted for net salvage, over the estimated remaining life of 7 the asset.145 PG&E generally applies this method to its hydro facilities at the FERC license level. 8 That is because it is PG&E’s typical practice to set the depreciation end of life of a collection of 9 hydro facilities contained under a single FERC license to the end of that license period, typically 10 anywhere from 40 to 50 years from license issuance.146 11 In order to apply the remaining life method to each group of hydro facilities, PG&E must 12 determine the gross plant and accumulated depreciation for each hydro facility. PG&E only tracks 13 accumulated depreciation at the FERC account level,147 so in order to determine the remaining 14 NBV148 for each hydro facility, it must therefore allocate that book accumulated depreciation to 15 each facility in order to create a theoretical reserve for each facility for the purposes of determining 16 the annual depreciation expense for each facility.149 17 The theoretical reserve calculated for the purposes of the depreciation study is not 18 equivalent to the book reserve because of the impact of net salvage. However, it is possible to 19 145 Id. at 8-16 lines 22-33.
146 Exh. PG&E-10 Workpapers at WP 9-35.
147 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-026). 148 The NBV is the balance to be depreciated over the remaining life of the hydro facility. 149 Attachment RAM-2 (PG&E’s response to DR CalCCA 012-013). 52 translate the information presented in the depreciation study into a book reserve per hydro facility, 1 using the formula provided by PG&E:150 2
Accumulated Depreciation = Original Cost * (100% - Net Salvage %) - Future Accruals 3 w/ Net Salvage Adjustment 4
Using this method across all of PG&E’s hydro facilities results in a total calculated book 5 accumulated depreciation of $2.061 billion as of December 31, 2023.151 This book accumulated 6 depreciation for hydro facilities is markedly different from the book accumulated depreciation 7 8 reported by PG&E in multiple other locations, as shown in the table below.

Table RAM-12
2023 EOY Hydro Accumulated Depreciation Across Sources
Hydro
Accumulated
Depreciation as
of 12/31/2023
2027 GRC
Depreciation
Study152
Att. 1 to
PG&E’s
Response to DR
CalCCA 012-
009153
Att. 1 to PG&E’s
Response to DR
CalCCA 002-
001154
2023 FERC
Form 1155
Total $2.061 billion $2.625 billion $2.626 billion $2.571 billion Difference from
2027 GRC
Depreciation
Study
N/A $0.561 billion $0.562 billion $0.507 billion
9 The implication of the $507 to $562 million difference in PG&E’s reported hydro 10 accumulated depreciation as compared to the accumulated depreciation used in the depreciation 11 150 Attachment RAM-2 (PG&E’s response to DR CalCCA 012-009). 151 See Attachment RAM-4.
152 See id.
153 Attachment RAM-2 (PG&E’s response to DR CalCCA 012-009, Attach. 1). 154 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-001, Attach. 1). PG&E provided a revised response in DR CalCCA 012-007 stating “The balances provided in CalCCA002-001 Attachment 1 were prepared prior to the depreciation study and, as a result, have some differences.” PG&E’s Attachment 1 to DR CalCCA 012-007 revises its 2023 accumulated depreciation book balances downward by $169 million to $2.455 billion, which again highlights the mismatch between the accumulated depreciation on PG&E’s books and that used in the depreciation study to calculate depreciation rates. 155 PG&E 2023 FERC Form 1, Page 219, Lines 22–23.
53 study is that PG&E is proposing depreciation expense, and therefore depreciation rates, that are 1 too high. That is because PG&E is targeting a remaining accrual that is $507 to $562 million more 2 than is needed to fully depreciate its hydro facilities. In other words, in the hypothetical situation 3 where the underlying assumptions used in this depreciation study remain unchanged for the life of 4 the hydro assets, PG&E’s customers will have paid $507 to $562 million more than the original 5 investment in the plant once all hydro assets are retired. 6 Revising PG&E’s depreciation study to account for the actual book accumulated 7 depreciation as reported by PG&E, allocated to each plant, would result in an annual depreciation 8 expense using the remaining life method that is approximately $15.7 million less than PG&E’s 9 request in this case. 156 The Commission should therefore reduce the electric generation 10 depreciation expense in the 2027–2030 forecasted revenue requirement by $15.7 million annually. 11 12 B. PG&E’s Method of Tracking Accumulated Depreciation Hinders Its Ability 13 to Fairly Allocate Costs to Its Current Customer Base PG&E’s current practice of tracking accumulated depreciation only at the FERC account 14 level is no longer reasonable in light of the fact that PG&E’s customer base now primarily consists 15 of departed load customers. In this new landscape, not all customers are similarly situated when it 16 comes to cost responsibility for various PG&E generation assets. For certain hydro assets, PG&E’s 17 status quo method of tracking accumulated depreciation makes it difficult, if not impossible, to 18 modify ratemaking treatment—e.g., PCIA vintaging treatment—when appropriate. Unless PG&E 19 begins tracking accumulated depreciation and the other related components of rate base, at the 20 individual hydro facility level, it will not be able to accurately and efficiently account for future 21 changes to ratemaking for these generation assets. Therefore, the status quo approach risks 22 improper cost shifts between departed and bundled customers. 23 156 Attachment RAM-4.
54 PG&E generally depreciates its hydro facilities157 over the term of their FERC licenses, 1 which can be anywhere from 40 to 50 years.158 The depreciation expense for each generation unit 2 is calculated based on what is known as “life span” depreciation, in which the estimated remaining 3 life for that generation unit is largely based directly on the time remaining to the terminal 4 retirement date.159 Depreciation expense is then calculated with the goal of that unit’s value being 5 fully depreciated at the terminal retirement date. As the remaining life reaches zero, the unit’s 6 service value should be fully depreciated. 7
Observing an individual hydro facility’s service value approaching zero as it approaches 8 its terminal retirement date in the data provided by PG&E is challenging. That is because PG&E 9 does not track accumulated depreciation at the hydro facility level—it is tracked only at the FERC 10 account level. 160 In instances in which PG&E needs to determine the book accumulated 11 depreciation by hydro facility, the accumulated depreciation at the FERC account level is allocated 12 to the facility level in proportion to the calculated accrued depreciation based on the currently 13 adopted service life and net salvage estimates.161 14
The practical result of this allocation process is that the level of accumulated depreciation 15 allocated to a specific hydro facility can change dramatically when its probable retirement date 16 changes. The allocation of accumulated depreciation to the Helms facility over the last three GRCs 17 provides an example of this impact. Prior to the 2020 GRC, the probable retirement date for Helms 18 was 2026, the year in which its FERC license expires. This is 50 years after the license issuance 19 157 I will use the term “hydro facilities” to refer to the collection of dams, conveyances, powerhouses, turbines, and supporting equipment all contained under a single FERC license. It is the facilities within a FERC license to which PG&E applies the remaining life method in its depreciation study. 158 Exh. PG&E-10 Workpapers at WP 9-25.
159 The impact of interim retirements and net salvage is also taken into account, which I am ignoring at the moment for the sake of simplicity.
160 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-004). 161 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-026). 55 in 1976 and 42 years after the facility was placed into service in 1984. The plant was being 1 depreciated using the life span method so that it would be fully depreciated at the end of 2026. In 2 the 2023 GRC, the probable retirement date for Helms was extended to 2069 as the relicensing 3 process got underway.162 And PG&E has proposed to further extend Helm’s life in this GRC to 4 2079.163 The extension of Helms’ probable retirement date had the impact of increasing the total 5 future accruals164 needed to fully depreciate the plant, in the absence of significant new investment 6 at the time. This was due to PG&E’s accumulated depreciation allocation process and the fact that 7 the probable retirement date is one of the inputs.165 The following table shows the original plant, 8 total future accruals, and average remaining life for Helms across the past three GRCs. 9 Table RAM-13
Helms Original Plant and Total Future Accruals in 2027, 2023, and 2020 GRCs Original Plant
Total Future
Accruals
Avg. Remaining
Life
2027 GRC166 $1,057 $689 43.87
2023 GRC167 $1,064 $697 40.69
2020 GRC168 $1,011 $326 8.22
10 162 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-002, Attach. 1). 163 Exh. PG&E-10 Workpapers at WP 9-36.
164 An increase in future accruals would imply a decrease in the accumulated depreciation allocated to Helms.
165 Attachment RAM-2 (PG&E’s response to DR CalCCA 012-009). 166 See Attachment RAM-4.
167 See id.
168 PG&E 2020 GRC, Exh. PG&E-10 Workpapers at WP 11-42, WP 11-43, WP 11-90, WP 11-93, WP 11- 152, WP 11-157, WP 11-197, WP 11-199, WP 11-243, WP 11-245, WP 11-286, WP 11-289, WP 11-318, and WP 11-321.
56 In the three years between the 2020 GRC169 and the 2023 GRC,170 the Helms facility did 1 not actually have its accumulated depreciation decrease by $371 million. 171 The fact that a 2 probable retirement date extension results in an increase to the amount of depreciation needed until 3 the facility is fully depreciated results in it being impossible to precisely determine when the 4 original investment in a hydro facility has been fully depreciated, and therefore when the cost 5 responsibility of departed load customers for that facility should cease. 6 Without tracking the accumulated depreciation at a facility level, it is extremely difficult 7 to determine the amount of depreciation that has been accrued on any given facility. In order to 8 accurately estimate the amount of accrued depreciation, one would need to retroactively calculate 9 the accrued depreciation on each unit based on the accurate rates applied to each account within 10 that unit year by year, back through the full life of the asset. Since it is challenging and 11 impractical—if not impossible—to do this calculation, rather than allocating accumulated 12 13 depreciation at specific points in time across a facility’s lifecycle, CalCCA proposes that PG&E begin tracking accumulated depreciation by plant with the 2024 net plant data.172 14 The Commission should order PG&E to begin tracking all components of plant by hydro 15 facility. The practice of tracking accumulated depreciation at the FERC account level and 16 allocating it to hydro facilities may be reasonable for a vertically integrated utility with a monopoly 17 service territory in which all its customers are similarly situated and bear comparable cost 18 169 The depreciation study was based upon December 31, 2017, plant balances. 170 The depreciation study was based upon December 31, 2020, plant balances. 171 This would be necessary to generate the change from $326 million to $697 million in total future accruals. For that to happen, the facility would need to appreciate by $123.7 million each year from 2018– 2020. Given that the proposed annual depreciation for Helms from the 2017 GRC (in effect in 2018 and 2019) was to depreciate the facility by $29.0 million annually and from the 2020 GRC (in effect in 2020) was to depreciate the facility by $39.7 million, that is impossible. 172 Note that CalCCA uses this methodology to produce the going forward calculations and recommendations for revintaging in Section III herein as well as the detailed calculations in Attachment RAM-3.
57 responsibility for the full fleet of hydro resources over the life of those resources. In that situation, 1 to which specific facility the accumulated depreciation is apportioned would not change the overall 2 electric generation revenue requirement, nor customer rates. 3 That is no longer the case for PG&E. The fact that a significant share of PG&E’s service 4 is provided to departed load customers means that not all customers are similarly situated. CalCCA 5 proposes in Section III herein that the Commission order PG&E to revintage a portion of PG&E’s 6 hydro assets. Should the Commission agree and order PG&E to place some hydro assets into a 7 new vintage, then accumulated depreciation must be tracked along with all related components of 8 plant over time at the facility level. Otherwise, revenue requirements will be improperly shifted 9 between departed load and bundled customers. CalCCA therefore recommends that the 10 Commission order PG&E to begin tracking plant, accumulated depreciation, depreciation expense, 11 and Accumulated Deferred Income Tax (“ADIT”) by hydro facility. 12 13 V. RATEMAKING ISSUES
14 A. PG&E’s Improper Calculation of Deferred Tax Assets for the Diablo Canyon 15 Power Plant Increases the Electric Generation Revenue Requirement Despite Senate Bill 846’s (SB 846) explicit prohibition on PG&E earning returns on 16 extended operations investments at DCPP, 173 and despite the Commission’s clear directive to track 17 DTA revenue requirements in a memorandum account pending IRS guidance,174 PG&E now seeks 18 to include a separate DTA balance for pre-extension investments in its rate base calculation in this 19 General Rate Case—effectively circumventing regulatory restrictions and forcing customers to 20 pay inflated revenue requirements for a plant that was supposed to operate under a unique, no-21 173 SB 846, 2021-2022 Reg. Sess. (Cal. 2022); see Cal. Pub. Util. Code § 712.8(h)(1). 174 A.24-12-033, Application of Pacific Gas and Electric Company to Recover in Customer Rates the Costs to Support Extended Operation of Diablo Canyon Power Plant from September 1, 2023 through December 31, 2025 and for Approval of Planned Expenditure of 2025 Volumetric Performance Fees, at 36, section

6.5. 58
return ratemaking structure. This improper calculation violates the statutory framework, ignores 1 Commission orders, and unjustifiably increases the electric generation revenue requirement that 2 ratepayers must shoulder. 3
PG&E’s inclusion of the DCPP DTA in the 2027 GRC rate base is improper for four 4 reasons. First, because the DCPP investments should have been fully depreciated for tax purposes 5 by 2025/26 retirement dates, there should not be a DTA in the first place. Second, PG&E could 6 have avoided the creation of a DTA by taking advantage of recent tax laws allowing accelerated 7 depreciation option for nuclear plants. Third, if there is any amount of DTA (or deferred tax 8 liability) rightly accumulated, it would be a violation of the Internal Revenue Code’s (“IRC”) 9 Consistency Rule to include it in rate base in this GRC. Fourth, PG&E does not need to place the 10 DTA into the rate base since it could avail itself of the IRS’s Safe Harbor for Inadvertent 11 Normalization Violations provision applies here. Accordingly, the Commission should order the 12 DTA balance be removed from the 2027 – 2030 revenue requirement, order PG&E to submit a 13 PLR on this issue prior to the processing of this case. and require PG&E to provide balancing 14 account treatment to the DTA revenue requirement so while PG&E seeks IRS guidance. 15 16 1. Overview of Accumulated Deferred Income Taxes
ADIT arises from the difference in depreciation expense incurred by utility customers 17 through rates and the depreciation expense claimed by the utility on its federal taxes. For rate 18 making purposes, utilities use straight-line depreciation. Under straight-line depreciation (referred 19 to as book depreciation), the initial capital cost of the asset is depreciated in equal portions over 20 the life of the asset. On the other hand, IRS rules allow a utility to depreciate capital investments 21 for tax purposes using accelerated depreciation methods resulting in faster depreciation schedules, 22 59 allowing a utility to write off the cost of the asset more quickly and thus reducing a utility’s tax 1 expense in the years directly following the investment. 2 The use of different depreciation methods for rate making and tax purposes results in a 3 difference between the tax expense a utility collects from customers and the amount it pays to the 4 IRS. Because accelerated depreciation methods reduce the amount of taxes the utility pays in the 5 years immediately following a capital investment, the utility will typically collect from its 6 customers a higher amount to cover its tax liability than the amount the utility pays to the IRS. 7 This difference between the tax expense amount collected from customers and the actual amount 8 paid to the IRS is known as Deferred Income Tax (“DIT”). The DIT generated annually is 9 accumulated as a balance (ADIT) that the utility places into rate base. When the ADIT balance is 10 positive, meaning that the utility has paid more taxes to the IRS than it has collected from 11 customers, it is a Deferred Tax Asset—an addition to the rate base. When the ADIT is negative, 12 meaning that the utility has collected more of its tax expense from customers than it has paid to 13 the IRS, it is a Deferred Tax Liability—a reduction from the rate base. 14 15 2. DCPP Pre-Extended Operations Investments Should Have Been Fully 16 Depreciated by 2026
There should not be any remaining book to tax timing differences from DCPP investment 17 prior to the extended operations period, and therefore no DTA, since it was planned to be fully 18 retired by 2026 prior to the rates for the 2027 GRC going into effect. This subsection will walk 19 through why PG&E’s assumptions have resulted in a DTA remaining on their books, and how 20 those assumptions should be modified. 21
Because the utility will, in the years following a capital investment, collect more in tax 22 expense from customers than is due to the IRS, plant-related ADIT balances are typically negative 23 60 and are a reduction to rate base. The DCPP DTA in this case is positive—an addition to rate base.175 1 This is because that the DCPP was fully depreciated for book purposes through the 2023 GRC 2 period as DCPP was to retire by the end of 2026. The 2027 GRC maintains that DCPP is fully 3 depreciated for book purposes,176 thus there is no rate base investment nor depreciation expense 4 for DCPP included in the 2027 GRC, which is also consistent with the extended operations 5 required rate-making treatment. 6
However, a tax depreciation life is still assumed in the 2027 GRC. PG&E confirmed this 7 when it stated “DCPP assets placed in service during . . . 2023 through 2026. . . have a federal tax 8 life of 15-years that continues past 2026 (final attrition year) but which originate from the 2023 9 GRC.”177 PG&E went on to state “The DTA should be reduced to zero by 2041 at the latest.”178 10 Using a 15-year federal tax life for investments made from 2023-26 is not reasonable. The plant 11 was originally scheduled to retire in 2026, and its pre-extended operations capital investment 12 should have been fully depreciated by 2026 for tax purposes. 13 14 3. Had PG&E Taken Advantage of Accelerated Depreciation Options 15 Available to It, There Would Likely be No DCPP DTA
PG&E could have also avoided the creation of a DTA had it taken advantage of recently 16 enacted tax provisions. Recent tax law changes in the Inflation Reduction Act (“IRA”) of 2022 17 and the One Big Beautiful Bill Act (“OBBBA”) of 2025 specifically address the Depreciation of 18 Nuclear Plant Extensions in order to significantly boost nuclear power, especially from existing 19 plants. These recent tax laws allow usage of the 5-Year MACRS schedule with the 200% declining 20 175 Attachment RAM-2 (PG&E’s response to DR CalCCA 003-14(b)). 176 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-023(b)); Attachment RAM-2 (PG&E’s response to DR CalCCA 003-007(a)).
177 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-023(a)). 178 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-023(c)). 61 balance method for the nuclear power plant extensions.179 Additionally, for these nuclear power 1 plant extensions, the MACRS depreciation rates could be combined with bonus depreciation which 2 could be up to a 100 percent first-year bonus.180 PG&E has not stated whether it has been able to 3 leverage the recent tax laws and the possibility of taking 100 percent first-year bonus depreciation 4 for the Diablo Canyon nuclear plant 2023-26 additions. But the presence of a DCPP DTA indicates 5 it did not. Had it taken advantage of these tax provisions, there would likely be not DTA or the 6 DTA would be significantly smaller than its current forecasted balance. 7 8 4. The Inclusion of the DCPP DTA in the GRC 2027 Violates the 9 Consistency Rule
The “consistency rule,” established by IRC section 168(i)(9)(B)(i), states that if PG&E 10 uses an estimate or projection for any one of (1) tax expense, (2) depreciation expense, or (3) the 11 deferred tax reserve, it must also use consistent estimates or projections for all three of those items 12 and for the rate base itself.181 In this case, PG&E has projected the ADIT derived from DCPP but 13 has not included projections of any of the other three plant-related rate components associated with 14 DCPP: gross plant balances, accumulated depreciation balances, and depreciation expense. 15 PG&E’s treatment of the DTA for the DCPP, therefore, violates the consistency rule. 16 Even if there is a properly generated DTA balance, restoring the DTA to rate base does 17 not, in and of itself, protect PG&E from a normalization violation. That is because normalization 18 requires PG&E to recover the tax timing differences over the life of the investment and therefore 19 requires customers to be credited for any ADIT as a credit to rate base. It also requires consistency 20 when updating any element of a plant in rates. In an attempt to avoid a normalization violation by 21 179 26 U.S.C. § 168(e)(3)(B).
180 25 U.S.C. § 168(k).
181 IRS, PLR-101961-21 at 7 (2021).
62 including the DCPP DTA in rate base, PG&E may have created a separate potential normalization 1 violation of the consistency rule. 2
3 5. PG&E Does Not Need to Include the DCPP DTA in the 2027 GRC Rate 4 Base Because IRS’s Safe Harbor Provision Protects PG&E from 5 Inadvertent Normalization Violations
The IRS also has a safe harbor provision associated with its normalization rules. In plain 6 language, the safe harbor provision provides a utility or regulator that discovers a potential 7 normalization violation the ability to cure (or fix) that error at its next available opportunity to 8 avoid the penalty of losing access to accelerated depreciation.182 Given the facts and circumstances 9 surrounding the DCPP DTA, PG&E and the Commission the safe harbor provision applies here. 10 IRS Revenue Procedure 2017-47, Safe Harbor for Inadvertent Normalization Violations (“Safe 11 Harbor”) provides in Section 4.07(3) the following: 12
13 If, at the conclusion of a Rate Proceeding, the taxpayer has a private letter ruling 14 request pending before the Service to address whether or not a practice or 15 procedure addressed in the Rate Proceeding is a Consistent Practice or Procedure, and the Taxpayer’s Regulator later establishes or approves rates subject 16 to adjustment from the effective date of the unadjusted rates in order to conform to 17 the Service’s ruling, the taxpayer shall have corrected its Inconsistent Practice or 18 Procedure at the Next Available Opportunity. 19
PG&E requested a PLR for a determination related to PG&E’s extended operations 20 DTA.
183 On January 30, 2026, the IRS issued a PLR stating that “excluding these revenue 21 requirements is not considered a normalization violation under this ruling.”184 This determination 22 strongly suggests that the IRS would not find the removal of the DCPP DTA to be a normalization 23 violation. Accordingly, PG&E should similarly submit a PLR on the DCPP DTA issue prior to the 24 conclusion of the 2027 GRC, in order to access the safe harbor provision so long as it adjusts rates 25 182 IRS Rev. Proc. 2017-47 § 4.07(3).
183 Attachment RAM-2 (PG&E’s response to DR CalCCA 006-03); Attachment RAM-2 (PG&E’s response to DR CalCCA 005-023(c)); Attachment RAM-2 (PG&E’s response to DR CalCCA 014-011). 184 PG&E Advice Letter (AL) 7835-E (Feb. 12, 2026).
63 pending a PLR decision. This will allow PG&E to remove the DCPP DTA from rates without 1 risking a normalization violation. 2
3 6. PG&E’s Should Remove the DCPP DTA from Rates or Provide 4 Balancing Account Treatment to Allow for a Future Refund.

PG&E’s position in its direct case in the 2027 GRC is that, should the IRS determine after 5 the close of this GRC that the DCPP DTA should not be included in base rates, it can make any 6 necessary adjustments using the electric generation revenue requirement adjustment mechanisms 7 during the Annual Electric True-Up.185 However that position shifts risks unnecessarily onto 8 customers if for any reason resolving the issue extends past future true-up windows. That risk can 9 be mitigated simply by taking the DTA out of base rates or moving the DTA related revenue 10 requirement to a balancing account. 11
The Commission should order PG&E to remove the DCPP DTA from rate base. Doing so 12 would reduce the Electric Generation revenue requirement by $4.3 million annually.186 To protect 13 against any potential normalization violations, the Commission should also order PG&E to submit 14 a Private Letter Ruling to the IRS for guidance on the appropriate rate-making treatment of the 15 DCPP DTA prior to the end of 2026, which would give PG&E access to safe harbor from a 16 potential normalization violation on this issue. The Commission should also order PG&E to 17 subject the DCPP DTA to balancing account treatment, so that any future IRS determinations can 18 be properly implemented and resulting in customers neither overpaying or underpaying for the 19 appropriate treatment of this asset. 20
185 Exh. PG&E-10 at 10-8 lines 6-13.
186 Attachment RAM-2 (PG&E’s response to DR CalCCA 003-014, Rev 1, Attach. 1). 64 1 B. PG&E’s Failure to Implement Changes to the Uniform System of Accounts 2 Required Under FERC Order No. 898 Overstates the Revenue Requirement 3 by $18.8 Million
PG&E’s failure to comply with FERC Order No. 898—nearly six months after the order’s 4 January 1, 2025 effective date—is passing millions in incorrectly allocated costs directly onto 5 CCA customers’ bills. Although PG&E completed the required plant account reclassifications for 6 book purposes in April 2025, it chose to exclude these mandatory reclassifications from its revenue 7 requirement proposal in this GRC, continuing to rely on obsolete functional allocations that inflate 8 the distribution of general and intangible plant costs to the Electric Generation function by $7.2 9 million. This unjustified non-compliance effectively subsidizes PG&E’s operations at the expense 10 of CCA customers, who are forced to pay their share of improperly allocated overhead costs while 11 PG&E avoids the adjustment that FERC accounting standards now require. 12 FERC Order No. 898 modified the Uniform System of Accounts to create new FERC 13 accounts specific to renewable generation, a new function with new FERC accounts specific to 14 energy storage, and new FERC accounts within each function to allow for the functionalization of 15 intangible and general plant, where possible.187 Prior to FERC Order 898, all intangible and 16 general plant were booked to specific intangible and general accounts, which PG&E functionally 17 allocated based on a labor factor in previous GRCs and a labor plus capitalized labor factor in the 18 current GRC.188 FERC Order 898 became effective January 1, 2025. 19 In its application, filed May 15, 2025, PG&E indicated that it had not yet complied with 20 FERC Order 898 because it was still compiling the data it allegedly needed to translate to the new 21 FERC accounts.189 While PG&E confirmed that plant account reclassifications were made in April 22 187 Accounting and Reporting Treatment of Certain Renewable Energy Assets, Order No. 898, 183 FERC ¶ 61,205 (2023).
188 Exh. PG&E-10 at 2-5.
189 Id. at 7-23.
65 2025 for book purposes, PG&E’s proposed revenue requirement in this case does not implement 1 the account reclassifications required by FERC.190 2
In response to data requests, PG&E provided its actual plant balances subject to 3 reclassification under Order 898 as of June 30, 2025.191 The following table summarizes the gross 4 plant from FERC Accounts 303 (intangible software), 391 (general office furniture), and 397 5 (general communication equipment) subject to reclassification to new functional hardware, 6 software, and communication equipment FERC accounts. 7
Table RAM-14
Gross General and Intangible Plant Balances Subject to FERC Order 898 Reclassification
FERC Account
Pre-Order 898
Balance FERC Account
Post-Order 898
Balance
303 – Misc
Intangible Plant $1,129,578,784
334.2 – Hydraulic
Production Software
$18,716,245
351.2 – Transmission
Software
$175,729,103
363.2 – Distribution
Software
$928,365,221
387.9 – Energy Storage
Computer Software
$2,436,496
397.2 – General Plant
Software
$4,331,719
Subtotal Software $1,129,578,784
391 – Office
Furniture Equip. $55,157,839
334.1 – Hydraulic
Production Hardware
$909,264
351.1 – Transmission
Hardware
$698,299,530
363.1 – Distribution
Hardware
$20,085,501
Subtotal Hardware $719,294,295
315.3 – Steam Production
Communication Equip.
$913,258
334.3 – Hydraulic
Production Communication
Equip.
$27,621,047
190 Attachment RAM-2 (PG&E’s response to DR CalCCA 002-003). 191 Attachment RAM-2 (PG&E’s Response to DR CalCCA 004-007, Attach. 1). 66 397 –
Communication
Equip.
$1,490,753,097
338.11 – Solar
Communication Equip.
$42,607
351.3 – Transmission
Communication Equip.
$210,909,254
363.3 – Distribution
Communication Equip.
$585,703,338
387.10 – Energy Storage
Communication Equip.
$448,563
397.3 – General Plant
Communication Equip.
$978,575
Subtotal Communication
Equip.
$826,616,642
Total $2,675,489,720 Total $2,675,489,720
Note: There is not a one-to-one translation between the total Communication Equipment in FERC account 397 pre-Order 898 and the total Communication Equipment in FERC Accounts 315.3, 334.3, 338.11, 351.3, 363.3, 387.10, and 397.3 post-Order 898 because some Communication Equipment was reclassified to the Hardware subaccounts.

Comparing PG&E’s allocation of General and Intangible plant prior to the direct-assigned plant accounting required by FERC Order 898 shows that PG&E’s proposed revenue requirement by business function in this case is quite different from what it would be if PG&E’s filing included the effects of FERC Order 898. The following table illustrates the change in net plant allocated to the Electric Generation function from the implementation of FERC Order 898.

Table RAM-15
Change in Electric Generation Allocated Net Plant from FERC Order 898192 CGI Net Plant in
FERC Accounts
303, 391, and 397
Electric Generation
Allocated
Prior to FERC Order 898 $103,717,420
Post-FERC Order 898
Direct Assigned Production $31,658,125
Direct Assigned Energy Storage $2,743,307
Remaining CGI Plant $4,557,438 $271,168
Total Post-FERC Order 898 $34,672,599
Total Change in CGI Plant Allocated
to Electric Generation from FERC
Order 898
($69,044,821)
192 Attachment RAM-9.
67 Attachment RAM-9 contains detailed calculations of the impact of FERC Order 898 on the 1 Electric Generation revenue requirement. Reclassifying general and intangible plant balances to 2 the new functional FERC accounts would reduce the Electric Generation revenue requirement by 3 an estimated $6.9 million based on the June 30, 2025 balances provided. 193 The revenue 4 requirement would also be reduced by the associated depreciation expense that would also be 5 reallocated to the new functional FERC accounts, reducing the revenue requirement an additional 6 $11.9 million, for a total of $18.8 million.194 7
The Commission should require PG&E to update its revenue requirement for the 2027–8 2030 GRC period to include the impact of FERC Order 898. To the extent that FERC Order 898 9 requires certain General and Intangible plant to be directly assigned to functional FERC accounts, 10 it should be reflected in PG&E’s proposed revenue requirement. It would be unjust and 11 unreasonable to set rates through 2030, 5 years after FERC Order 898 was required to be 12 implemented, without reflecting the impacts of the FERC decision. 13 14 VI. PRODUCTION O&M EXPENSE FORECAST
PG&E is claiming a bloated 35 percent spike in hydro production O&M expenses for 15 2027—jumping from $199.1 million in 2024 to $267.8 million195—yet ignores the substantial 16 operational savings that its own strategic investments in advanced asset management systems 17 should deliver. While PG&E justifies this dramatic expense increase by citing investments in 18 additional staffing and enhanced asset and risk management capabilities,196 the company makes 19 no attempt to quantify or pass through the efficiency gains and cost reductions these investments 20 are designed to generate. This one-sided treatment inflates ratepayers’ bills: customers are forced 21 193 Id.
194 Attachment RAM-9.
195 Exh. PG&E-5 at 1-23, Table 1-1 (Nov. 15, 2025).
196 Id. at 3-1 lines 16–27.
68 to absorb the full upfront cost of PG&E’s operational improvements while the company pockets 1 the downstream savings, turning what should be prudent investment into a hidden rate increase. 2 Figure RAM-11: Production O&M Expense, 2020–2024 Actuals, 2025–2027 Forecast
3 A. PG&E’s Forecasts Overestimate Its Expense for Hydro Operators PG&E’s proposed increased staffing of hydro operators results in a $10.3 million increase 4 in 2027 forecasted expense compared to 2024 from both the direct cost of those operators and the 5 indirect cost of additional training and support staff. 197 PG&E plans to hire 48 additional 6 hydroelectric operators by 2027 and has already begun hiring such that as of June 30, 2025, it has 7 55 operators, or 6 more than on December 31, 2024.198 Those 48 additional operators would be an 8 addition to the 49 existing operators as of the end of 2024, effectively doubling the number of 9 hydro operators on staff. Yet, although these hirings are intended to address staffing attrition and 10 197 Id. at 3-42, Table 3-14.
198 Attachment RAM-2 (PG&E response to DR CalCCA 005-013). 69 overtime,199 PG&E fails to account for the savings that would come from reduced overtime hours, 1 2 existing staff attrition, and future facility sales and retirements.

Table RAM-16
Existing and Additional Hydro Operators by Location
Location DeSabla Drum Helms
Kings-Crane
Valley Motherlode Shasta Total
Existing
Operators
9 13 3 5 8 11 49
Additional
Operators
12 9 3 7 6 11 48
Total
Operators
21 22 6 12 14 22 97
3 PG&E’s O&M expense does not reflect the staff attrition it is attempting to solve. PG&E describes that it takes three years to turn a new hire into a qualified hydro operator, and due to that 4 long lead time it is necessary to maintain a pipeline of hiring in order to maintain staffing levels in 5 the face of existing operator attrition.200 PG&E has a history of doing just that; from 2015 through 6 2024 it has hired an average of 8.1 hydro operators per year, with a low of one hire in 2019 and 7 2021 and a high of 13 hires in 2020.201 Over that same time period, PG&E experienced attrition of 8

9.1 hydro operators per year, on average, which resulted in its year-end head count decreasing 9 from 58 hydro operators in 2015 to 49 operators in 2024.202 Just because attrition is a factor that 10 necessitates new hires does not mean that it results in an operational expense increase. Since the 11 expense from the new hire to backfill for attrition replaces the expense that is reduced from the 12 operator who departed, there should be no net change to the forecast as a result of attrition. PG&E 13 199 Exh. PG&E-5 at 3-5 lines 7–15.
200 Id. at 3-43 lines 24–31.
201 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-014). 202 Attachment RAM-2 (PG&E’s response to DR CalCCA 014-007). 70 did not adjust its forecast for future hydro operator attrition. Therefore, the forecast overestimates 1 PG&E’s expense for hydro operators. 2
Although PG&E’s hiring plans are intended to address its significant overtime expense, 3 PG&E’s proposed O&M expense does not show reflect those savings. PG&E states that it “must 4 often ask operators to work overtime to continue to staff its switching centers and be able to 5 conduct necessary surveillance activities to ensure public safety”203 and that the additional staffing 6 will address fatigue management challenges and therefore reduce overtime. 204 PG&E then 7 contradicts itself to say that it does not expect fewer overtime hours as a result of hiring but that 8 there could be overtime reduction on nights and weekends in the specific circumstance when a 9 second operator could respond to a remote event without having to call out a roving operator on 10 overtime.205 However, the estimated cost reduction from this specific circumstance is only $0.18 11 million out of a total 2024 hydro operator overtime expense of $7.4 million. It seems highly 12 implausible that the planned doubling of the number of hydroelectric operators, with the stated 13 intention to reduce overtime, could not significantly reduce the $7.4 million in overtime expense 14 incurred by PG&E in 2024. 15
Further, PG&E has not made an adjustment to the number of planned hydro operators due 16 to the planned sale or retirement of the Phoenix, Kerckhoff 1, Potter Valley, Hamilton Branch, and 17 Battle Creek hydro facilities.206 The fact that these plants are no longer planned to be operated 18 through the GRC period should result in a downward adjustment to operator as well as 19 maintenance expense. 20
203 Exh. PG&E-5 at 3-43 lines 6–8.
204 Id. at 3-5 lines 9–10.
205 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-012). 206 Attachment RAM-2 (PG&E’s response to DR CalCCA 014-010). 71 In addition to failing to pass on to customers the expected savings resulting from its hiring 1 plans, there appears to be a more fundamental flaw in PG&E’s plan: it has not demonstrated that 2 it has the ability to actually hire all the operators it has forecasted. PG&E hired six net new hydro 3 operators in the first six months of 2025.207 To achieve its planned headcount of 97 operators, 4 PG&E would need to hire 42 new operators (net of attrition) over an 18-month period. Given 5 PG&E’s average annual hydro operator attrition rate of 9.1 operators per year,208 it would also 6 need to hire an additional 14 hydro operators to backfill for likely attrition As a result, PG&E must 7 hire 56 operators over an 18-month period, which is a rate of 37 operator hires/year. Given that 8 the most operators PG&E has hired in a year over the last 10 years is 13,209 its plans do not seem 9 achievable. 10
CalCCA recommends that the Commission reject the proposed $10.3 million increase to 11 operational expense from PG&E’s plans to nearly double its number of hydro operators. PG&E 12 has not justified the need for the substantial increase in operators it proposes, particularly in light 13 of its shrinking hydro fleet; it has not accounted for the full amount of overtime savings its plans 14 should produce; it has not accounted for existing operator attrition; and it has not demonstrated 15 that it can even reach the levels of forecasted costs. 16 17 B. PG&E Must Include in Its Revenue Requirement Cost Savings Realized from 18 Its Asset Management System Investments
PG&E acknowledges that its Asset Management System generates significant cost savings 19 through waste elimination and efficiency improvements—yet fails to pass these benefits on to 20 ratepayers. While seeking to recover the full cost of the AMS investment in its O&M expenses, 21 PG&E makes no offsetting adjustment for the measurable savings these investments are designed 22 207 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-013). 208 Attachment RAM-2 (PG&E’s response to DR CalCCA 014-007). 209 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-014). 72 to produce, effectively allowing the company to pocket windfalls from continuous improvement 1 while customers bear the full expense burden.210 2
Since the 2023 GRC, PG&E has adopted an asset and risk management practice with 3 objectives of continuous improvement, compliance with industry standards, and implementation 4 of risk management best practices. Its approach involves the implementation of an Asset 5 Management System (“AMS”) to integrate strategic planning, risk management, and reliability 6 optimization throughout the asset lifecycle. Power Generation received ISO 55001 certification in 7 March 2022 for its AMS.211 8
The planned continued investment in its AMS has been identified by PG&E as an O&M 9 expense increase driver. PG&E has identified enhanced asset and risk management as the reason 10 for the increase in O&M expense between 2024 actuals and the 2027 forecast of $16.3 million in 11 the Major Work Category KH – Maintain Hydro Generating Equipment;212 $1.1 million in the 12 MWC KI – Maintain Hydro Buildings, Grounds, and Infrastructure;213 and $13.3 million in the 13 MWC AX – Maintain Reservoirs, Dams, and Waterways.214 In total, enhanced asset management 14 is a driver for approximately $30.7 million of O&M expense increases from 2024–2027. 15 The investment in the AMS is necessary to achieve the desired objectives: improved safety 16 and reliability, reduced operational risk, and improved regulatory compliance. An additional 17 benefit of the implementation of AMS comes in the form of cost reductions from the elimination 18 of waste.215 PG&E provides an example of these savings in the form of an innovative construction 19 technique that saved an estimated $3.5 million in labor costs in 2024 on the Canyon Dam spillway 20 210 See Attachment RAM-2 (PG&E’s response to DR CalCCA 010-22). 211 Exh. PG&E-5 at 1-15 lines 4–21.
212 Id. at 3-48, Table 3-19.
213 Id. at 3-53, Table 3-22.
214 Id. at 3-57, Table 3-26.
215 See Attachment RAM-2 (PG&E’s response to DR CalCCA 010-24). 73 project.216 These benefits would be expected to recur on a regular basis. PG&E quantified that the 1 total cost savings achieved across all of Power Generation through its efforts to eliminate waste 2 and adopt lean operating principles was $24.5 million in 2024.217 In the absence of a specific 3 forecast for planned cost savings from AMS, the historical cost savings achieved can be a 4 reasonable proxy as they represent the level of savings possible from the presence of a waste 5 reduction, continuous improvement program. 6
If increased costs associated with AMS are included in the revenue requirement, offsetting 7 savings must also be included; otherwise, PG&E is overestimating its 2027 expense. If increased 8 investment in AMS does not continue to achieve year-over-year operational cost savings, then 9 PG&E should reevaluate the value of that investment. Absent any more specific information about 10 potential cost savings, PG&E should reduce its 2027 O&M expense by $24.5 million, the level of 11 achieved savings in 2024. 12
13 VII. COST ALLOCATION ISSUES
14 A. PG&E’s Allocation of Common, General, and Intangible Plant Violates 15 the Requirements of FERC Order No. 898 and Results in Improperly 16 Allocated Costs to the Functional Revenue Requirements PG&E’s selective and piecemeal changes to how it allocates Common, General, and 17 Intangible (“CGI”) costs is artificially shifting expenses between customer classes, forcing some 18 ratepayers to pay far more than their cost to serve while unfairly subsidizing others. By cherry-19 picking the direct assignment of hydro license costs to generation while failing to investigate or 20 adjust the allocation of any other CGI plant costs despite FERC Order 898’s clear requirements, 21 PG&E is creating an inconsistent cost allocation that distorts rates across all functions. 22 216 Exh. PG&E-5 at 3-6 lines 3–11.
217 Attachment RAM-2 (PG&E’s response to DR CalCCA 010-023). 74 FERC Order 898 requires PG&E to functionally allocate intangible plant booked to FERC 1 Account 303 and general plant booked to FERC Accounts 391 and 397.218 2 Here instead, PG&E proposes to functionally allocate all CGI plant based on a labor allocator, consistent with past 3 GRCs.219 PG&E maintains that approach in this case with one notable exception: it proposes to 4 direct assign the intangible hydro license costs to the Electric Generation function.220 PG&E has 5 not proposed to direct assign any other category of CGI plant, despite the requirements of FERC 6 Order 898, nor has it indicated that it investigated the potential to direct assign any other category 7 of CGI plant.221 8
If PG&E wants to depart from the historical precedent of allocating all CGI plant based on 9 a labor factor, it should not perform such changes piecemeal. Departing from an allocation method 10 for one portion of CGI plant without even investigating the direct assignment or more precise 11 allocation of other CGI plant would result in cost allocations, and therefore rates, that are 12 inconsistent across all functions, leading to some customers paying more than their cost to serve 13 and other customers paying less than their cost to serve. Instead, PG&E should conduct a 14 comprehensive study of all CGI cost allocations to determine their appropriateness and should 15 refrain from making any adjustments until that study is complete, with the exception of the 16 reclassifications originating from FERC Order 898 as it is a regulatory requirement that those be 17 reclassified. Even with the FERC Order 898 reclassification, there remain balances in the 18 intangible, general, and common accounts that would require further investigation to determine 19 218 Order No. 898, 183 FERC ¶ 61,205 at PP 99-100, 110. 219 Exh. PG&E-10 at 7-11 lines 2–3. In prior GRCs, that labor allocator was based solely on O&M labor, excluding capitalized labor. In this case, PG&E proposes to modify the labor allocator to instead be based upon the sum of O&M and capitalized labor. This change is due to the requirements of SB 846 that the Diablo Canyon investment not be capitalized, and as a result PG&E no longer incurs capitalized labor to support DCPP operations.
220 Exh. PG&E-10 at 7-12 lines 7–25.
221 Attachment RAM-2 (PG&E’s response to DR CalCCA 014-001). 75 more precise functional allocations. PG&E can then propose a more refined and fair allocation 1 method for CGI plant that considers all CGI plant in the next GRC. 2 Should the Commission not order PG&E to refrain from any CGI allocation changes prior 3 to the completion of a comprehensive study aside from FERC Order 898 requirements, then 4 CalCCA recommends that PG&E still complete a comprehensive study prior to the next GRC but 5 also make the following CGI allocation changes in addition to direct assigning the intangible hydro 6 licenses to the extent that it is not already required to be reclassified by FERC Order 898: 7 • Electric General (“EGP”) and Electric General Land (“EGPL”)222 – These should 8 only be allocated to Electric UCCs. These assets are specific to the electric utility 9 by definition, and therefore it is not reasonable to allocate a portion of their costs 10 to the gas utility. 11
• Gas General – Line 401 (“GGE”), Gas General-Land – Line 401 (“GGEL”), Gas 12 General (“GGP”), Gas General Land (“GGPL”), Gas Intangible-Line 401 (“GIE”), 13 and Gas Intangible (“GIP”)223 – These should only be allocated to Gas UCCs. 14 These assets are specific to the gas utility by definition, and therefore it is not 15 reasonable to allocate a portion of their costs to the electric utility. 16 In order to achieve these more specific allocations, PG&E should develop new labor 17 allocators to allocate costs only within the Electric utility and the Gas utility. The following table 18 shows a recommendation for that allocator.224 19
222 Exh. PG&E-10 at WP7-1, Table 7-5; id. at WP7-138, Table 7-45. 223 Id.
224 See Attachment RAM-10 for the derivation of the allocators. 76 Table RAM-17
Recommended Allocation for Electric General and Gas General & Intangible Plant Electric
Generation
Electric
Distribution
Gas
Distribution
Gas
Transmission
& Storage
Non-GRC Total
PG&E
Proposed

5.95% 57.59% 18.28% 8.09% 10.09% 100.00%
Electric
General

8.08% 78.22% 0% 0% 13.70% 100.00%
Gas
General &
Intangible
0% 0% 69.32% 30.68% 0% 100.00%
The following table shows the net change in WAVG 2027 general plant balances by 1 function. This table assumes that none of the EGP, EGPL, GGE, GGEL, GGP, GGPL, GIE, and 2 GIP are impacted by FERC Order 898. 3
Table RAM-18
Net Change in CGI Plant Allocation
Total
Investment
Electric
Generation
Electric
Distribution
Gas
Distribution
Gas
Transmission
& Storage Non-GRC
Change from
Electric
General
Reallocation
$0 $30,939 $299,502 ($259,419) ($114,797) $43,776
Change from
Gas General
& Intangible
Reallocation
$0 ($58,399) ($565,328) $502,599 $222,408 ($101,280)
Total
Allocation
Change
$0 ($27,460) ($265,826) $243,180 $107,611 ($57,504)
77 The impact of these reallocations is estimated to reduce the Electric Generation revenue 1 requirement by $2.7 million, reduce the Electric Distribution revenue requirement by $26.6 2 million, increase the Gas Distribution revenue requirement by $24.3 million, and increase the Gas 3 Transmission & Storage revenue requirement by $10.8 million. To the extent that any of the 4 balances are already impacted by FERC Order 898, the impact of the more precise allocation of 5 any remaining balances on the functional revenue requirement will be smaller. 6 7 B. PG&E’s Outdated Fees for Performing Billing Services on Behalf of CCAs 8 Are Resulting in CCA Customers Paying More to PG&E than the Cost to 9 Serve Those Customers
PG&E is charging CCA customers metering and billing fees locked in at 2015 cost levels, 10 creating a systematic overbilling scheme that forces CCA customers to pay inflated rates that no 11 longer reflect PG&E’s actual cost to serve them. By refusing to reconcile what PG&E actually 12 spends to provide billing services against what it charges CCA customers for those services, the 13 utility creates a hidden cost-shift, overbilling CCAs and allowing bundled customers to benefit 14 from the surplus. 15
PG&E provides billing services to CCA customers. CCAs are directly charged for metering 16 and billing costs by PG&E on a per meter or per customer basis to provide that service. Those fees 17 include the Meter Data Management (“MDM”) Fee, Rate-Ready Billing (“RRB”), and Bill-Ready 18 Billing (“BRB”).225 These rates, like all other rate schedules, are determined in a Phase 2 GRC 19 based on the cost to provide those services as determined during the Phase 1 GRC. The currently 20 applicable MDM, RRB, and BRB rates were determined based on a cost study included in PG&E’s 21 225 PG&E Electric Schedule E-CCA, Sheets 4 & 6 (effective March 1, 2018). 78 2017 GRC which is based upon 2015 costs and billing units.226 These fees must be updated as they 1 no longer reflect PG&E’s cost to serve CCA customers. 2 Based upon data provided by PG&E in response to CalCCA’s data requests, from 2020 3 through 2024, the revenue from DA/CCA billing services received by PG&E was greater than the 4 DA/CCA billing services expense by $10.3 million.227 The resulting impact of that imbalance is 5 that a net credit is embedded as a customer cost which is allocated by PG&E to the Gas and Electric 6 Distribution functions.228 Therefore, the net credit, or the amount of revenue paid by DA/CCA 7 customers that is greater than their costs to serve, gets allocated out to bundled customers as well 8 as to DA/CCA customers. PG&E’s bundled customers have therefore been receiving a benefit 9 from DA/CCA customers as opposed to being held indifferent. 10 Updating PG&E’s metering and billing fees based on costs from the ongoing Phase 2 GRC 11 proceeding would almost certainly result in a reduction in fees for CCA customers. Since 2017, 12 the number of DA/CCA customers has grown dramatically.229 DA/CCA billing revenue scales 13 linearly with the number of DA/CCA customers; in other words, each incremental DA/CCA 14 customer brings in the same amount of incremental revenue. It is not likely, however, that 15 incremental DA/CCA customers result in PG&E incurring the same amount of incremental cost 16 226 Attachment RAM-2 (PG&E’s response to DR CalCCA 001-020 from the 2023 Phase 2 GRC, A.24-09- 014, Attach. 1). See also Attachment RAM-2 (PG&E’s response to DR CalCCA 012-015). 227 Attachment RAM-2 (PG&E’s response to DR CalCCA 005-016, Attach. 1). The attachment shows a net credit in most years, which is consistent with PG&E’s stated practice of reducing costs by the revenue received, and therefore the sum of the total expense shown is equal to the net cost or net revenues from CCA/DA billing services. See also Attachment RAM-2 (PG&E response to DR CalCCA 002-022). 228 Attachment RAM-2 (PG&E’s response to DR CalCCA 014-002). 229 National Renewable Energy Laboratory, Community Choice Aggregation: Challenges, Opportunities, and Impacts on Renewable Energy Markets at 7 (Feb. 2019) which shows an estimated 5% of state sales served by CCAs in 2017; California Public Utilities Commission, Community Choice Aggregation and Energy Service Provider Formation Report at 7 which shows 38% of load statewide served by CCAs in 2017, and 49% of PG&E’s load.
79 for each customer added. That is because there are economies of scale in meter reading and billing 1 functions, so each new DA/CCA customer should result in incrementally lower marginal costs. 2 The most reasonable way to fix the current mismatch between CCA billing service revenue 3 and CCA billing service cost is for PG&E to complete an updated cost study to base the billing 4 service rate on more accurate and current costs. That would result in the revenue and expense being 5 much more closely aligned moving forward. However, that update is required to be performed in 6 a GRC Phase 2 case. PG&E should update its billing service rates in the ongoing GRC Phase II, 7 which would result in billing revenues and costs being aligned by the first quarter of 2027. 8 Until the billing services rates can be updated through the GRC Phase II proceeding, the 9 Commission should order PG&E to cap its revenues associated with CCA/DA billing services at 10 the cost included in its forecast for such services. Any revenue received above that cap within the 11 GRC should be tracked and refunded to CCA/DA customers through the ERRA proceeding. 12 13 VIII. HYDRO LICENSING BALANCING ACCOUNT EXPANSION
14 A. PG&E’s Expansive Use of Its Hydro Licensing Balancing Account Is 15 Unwarranted and Reduces the Incentive for PG&E to Control Costs PG&E’s use of the Hydro Licensing Balancing Account (“HLBA”) has morphed from a 16 limited true-up mechanism into a blank check that eliminates any financial incentive for the 17 company to control or efficiently manage hydro-related costs. By allowing PG&E to recover every 18 dollar of hydro licensing, regulatory fees, and miscellaneous costs through automatic balancing 19 account adjustments—rather than requiring the company to forecast and manage these expenses 20 within a fixed rate base like it must do with other operational costs—PG&E has eliminated the 21 fundamental ratemaking discipline that compels cost control. When a utility knows that any 22 expense variance will be automatically recovered through a balancing account, it loses the 23 powerful financial incentive to make tough resource allocation decisions, negotiate aggressively 24 80 with contractors, or implement efficiency improvements. The result is a moral hazard that invites 1 cost escalation, with ratepayers bearing the full burden of every overage, regardless of whether the 2 increases stem from prudent necessity or preventable inefficiency. 3 Balancing account treatment is appropriate for costs which are unpredictable, variable, and 4 generally outside of the utility’s control. That treatment ensures that the utility is not financially 5 harmed by factors beyond its direct control and allows for customers to pay the exact costs 6 associated with the underlying expense. On the other hand, base rate treatment is appropriate for 7 costs which are largely under the utility’s control and therefore can be forecasted and managed. It 8 is up to the utility to manage its costs relative to that forecast to ensure safe and reliable service to 9 customers while maintaining its financial health and achieving its authorized return. The risk of 10 achieving the earned rate of return rests with the utility, which provides an incentive to manage 11 capital and operational spending on an ongoing basis. Balancing account treatment removes that 12 incentive. 13
The HLBA is a two-way balancing account that was established in 2014 to capture 14 operational and capital costs associated with FERC relicensing activities for licenses issued 15 beginning on January 1, 2012. Its use has since been expanded to also include federal and state 16 regulatory fees, the costs of implementing the 2003 Crane Valley Recreation Settlement 17 Agreement, and costs stemming from the 2017 Oroville spillway incident.230 The 2023 GRC 18 further modified the HLBA to include a reasonableness review for all costs greater than 20 percent 19 above the authorized revenue requirement.231 20
While balancing account treatment is appropriate for hydro licensing costs, particularly 21 expenses, that are unpredictable and beyond the utility’s control, PG&E has expanded the amount 22 230 Exh. PG&E-5 at 7-7 lines 9–16.
231 Id. at 7-7 lines 17-19.
81 of costs proposed to be passed through the HLBA far beyond a reasonable interpretation of this 1 definition. The following graph shows actual costs incurred through the HLBA from 2020–2024 2 as well as forecasted costs from 2025–2030. 3
4 Figure RAM-12: Capital Costs in HLBA, Actual 2020–2024, Forecasted and Proposed 5 2025–2030
6 7
PG&E has expanded the amount of capital expenditure proposed to receive balancing 8 account treatment to a substantial share of its total forecasted capital expenditure on its hydro 9 facilities, specifically 43 percent in 2027, 55 percent in 2028, 57 percent in 2029, and 49 percent 10 in 2030. 232 Effectively, PG&E is stating that approximately half of its hydroelectric capital 11 expenditures in 2028, 2029 and 2030 are unpredictable, variable, and generally outside of the 12 utility’s control. Yet, many of the projects proposed for each category of the HLBA do not warrant 13 balancing account treatment. 14
The following table shows a categorization of the capital costs proposed for inclusion under 15 the existing scope of the HLBA,233 with a more detailed table presented in Attachment RAM-11. 16 232 Attachment RAM-11.
233 This categorization is based on the groupings shown in the electronic version of Exhibit PG&E-5 WP 3- 64 to 95.
82 Table RAM-19
Forecasted Capital Expenditure by Category for Inclusion in HLBA, 2027–2030 (‘000s of $)
Category 2027 2028 2029 2030
Security $34,117 $41,643 $26,430 $25,394
Spillway $64,836 $151,564 $146,686 $136,720
License
Conditions
$37,165 $54,255 $105,481 $112,658
HLBA
Expansion
$78,816 $100,512 $87,442 $59,074
Total $214,935 $347,974 $366,040 $333,846
1 The total costs of $127.6 million across the 2027–2030 time period for security investments 2 consist primarily of upgrades to or installation of fencing and security cameras across PG&E’s 3 hydro facilities.234 While those investments come as the result of FERC requirements, that alone 4 does not make these expenses unpredictable, variable, and outside of the utility’s control. In fact, 5 for some of these investments PG&E has been aware of the requirements since 2011.235 Given that 6 this program also implements risk mitigation strategies developed in the 2024 Risk Assessment 7 and Mitigation Phase (“RAMP”) report236 and consists largely of the same types of investment 8 across multiple facilities, it is the type of investment well suited to recovery in base rates. 9 The spillway category includes projects intended to mitigate the risk of large uncontrolled 10 water release and to maintain the spillways as required by state and federal regulations.237 The 11 largest project in this category, the McCloud Spillway Improvement, represents 72 percent of the 12 total 2027–2030 forecasted spillway investment of $499.8 million.238 The need for this project 13 originated in the wake of the 2017 Lake Oroville Dam spillway incident, when PG&E assessed its 14 234 Exh. PG&E-5 at WP3-90 to 95, lines 1,634–1,635, 1,709–1,711, 1,714, 1,716–1,718, 1,720–1,739, 1,741–1,743, 1,748, 1,752–1,757, 1,763–1,795, 1,797–1,851, 1,856–1,864, 1,866, and 1,890–1,918. 235 Id. at WP 3-452.
236 Id. at WP 3-453.
237 Id. at WP 3-382.
238 Id. at WP 3-381 to 384.
83 inventory of similar spillways. While this project originated from safety-related regulatory 1 requirements, in its implementation it appears to be comparable to any other large capital project. 2 PG&E has provided no evidence on why it is reasonable to provide balancing account treatment 3 for $358.1 million of forecasted capital investment for the McCloud Spillway project. Providing 4 guaranteed recovery of the forecasted cost plus 20 percent239 reduces the incentive for PG&E to 5 control costs, an incentive that is a critical element of ratemaking in this case given PG&E’s 6 substantial proposed generation capital investment and operating expense increases. 7 The category of license conditions refers to specific projects that are required as a result of 8 the FERC relicensing process. While there’s often uncertainty because the timing and scope of 9 those projects are unknown until that license is issued, projects totaling $62.5 million in forecasted 10 capital expenditure from 2027–2030 are the result of license conditions stemming from the 11 issuance of a FERC license prior to the filing of this GRC.240 For those projects, the license 12 conditions are already known, and the scope and timing uncertainty that would justify balancing 13 account treatment no longer exists. Therefore, those costs should not be subject to balancing 14 account treatment. The following table details the forecasted $62.5 million in capital expenditures 15 for licenses already issued prior to the filing of this case. 16 239 Costs more than 20% above forecast are subject to a reasonableness review, in effect guaranteeing costs expended up to 120% of forecast.
240 Attachment RAM-11.
84 Table RAM-20
Capital Expenditure Forecast for License Condition Projects by Hydro Facility Hydro
Facility
FERC
License
Date
Capital Expenditure Forecast (‘000s of $)
2027 2028 2029 2030
Mokelumne 2001 $5,098 $4,654 $2,311 $2,554
Rock Creek 2001 $285 $665 $1,541 $4,966
Hat Creek 1
& 2
2002 $778 $2,557 $2,371 $1,091
Pit 1 2003 $571 $23 $0 $0
Crane Valley 2003 $2,568 $2,294 $1,845 $469
Poe 2018 $890 $327 $2,080 $586
Bucks Creek 2022 $3,185 $4,145 $7,270 $7,392
Total Pre-2025 $13,376 $14,666 $17,418 $17,057
1 PG&E’s proposed expansion of the HLBA to include LGUWR projects is also not 2 reasonable based on the types of capital costs included. The LGUWR includes capital and 3 operational expenses related to vegetation management and rodent abatement, erosion mitigation, 4 spill way remediations, seismic retrofits, outlet refurbishments, and physical security.241 None of 5 these categories present the level of uncertainty and variability in cost to warrant balancing account 6 treatment. They represent the standard course of business for a utility of PG&E’s size and capacity. 7 Costs for capital projects and operational expenses of these types should be recovered through 8 base rates to continue to provide an incentive for PG&E to control costs. 9 PG&E argues that these projects take years to plan, design, and construct; that construction 10 windows are uncertain and narrow; and that the scope of the projects is dependent on government 11 agencies.242 But those factors are also likely present in other hydro investments that do not receive 12 241 Exh. PG&E-5 at 2-17, Table 2-2.
242 Id. at 1-18; id. at 3-94.
85 HLBA treatment. And the fact that projects take place over multiple years, even multiple years to 1 simply design, is a reason that they should be included in base rates set in a GRC. PG&E should 2 be developing specific cost estimates over those multiple years that it can manage in order to meet 3 during project implementation. Given these long time frames, there is plenty of time to include 4 those costs in a GRC forecast. 5
In summary, the Commission should limit the use of the HLBA to costs that are truly 6 unpredictable, variable, and generally outside of the utility’s control. In rigorously applying that 7 definition, the Commission should not allow HLBA recovery of the LGUWR Security and 8 Spillway categories of costs, and it should limit the License Conditions projects only those 9 facilities that have not yet received their FERC license at the time of filing a GRC. Adopting this 10 recommendation would result in the recovery of the following forecast capital expenditures 11 moving from the HLBA to base rates: 12
Table RAM-21
Capital Expenditure from HLBA to Base Rates243
(’000s of $) Year 2027 2028 2029 2030 Total
Capital Expenditure
from HLBA to Base
Rates
$191,146 $308,385 $277,977 $238,245 $1,015,753
243 Attachment RAM-11.

Flexible Connections All-Party
Workshop
R.21-06-017
February 20, 2026
1 Who Are Community
Choice Aggregators?
2 CalCCA Interactive CCA Map & Address Lookup:
https://cal-cca.org/cca-map/
CCA Launch Timeline
California CCAs: By the Numbers
Number of California communities served by CCAs: 229
Number of Counties with CCA: 21 of 58 counties (36%)
Number of Cities/Towns with CCA: 208 of 483 Cities (43%) California Population served by CCAs: 16 Million+ (38%) 4 2025 and 2026 CCA Expansions
CCA New/Expansion Cities/Counties Start of Service
Central Coast Community Energy Expansion County of SLO, City of Atascadero January 2025 Ava Community Energy Expansion City of Stockton and Lathrop April 2025 Marin Clean Energy Expansion City of Hercules April 2025 Clean Power Alliance Expansion Cities of La Canada Flintridge, Lynwood, Port Hueneme
October 2025
2025
2026
CCA New/Expansion Cities/Counties Start of Service
Ava Community Energy Expansion County of San Joaquin May 2026 Orange County Power Authority Expansion City of Fountain Valley October 2026 5 2024 Percentage of CA Retail Load by Energy Provider
(Electricity deliveries GWh to End Users)
47%
37%
16%
2024 Percentage of LSE Retail Load In IOUs Territories
IOU CCA Direct Access
29%
33%
26%
12%
2024 Percentage of CA Retail Load by Energy Provider
POU IOU CCA Direct Access
Source: 2023 IEPR Forecast. California Energy Demand 2023-2040 Forecast - Planning Forecast 6 Why Are CCAs Interested
in Flexible Service
Connections?
7 Climate Impact
Local, transparent decision-making/accountability
Rate competition/rate stability
Green energy choices
Community-based programs
Economic Development, Jobs
8 CCA APPEAL
9 "CCAs can design and deploy innovative initiatives and community-centered programs that provide financial and
environmental benefits and can respond to communities’
needs.”
-UCLA Luskin Center for Innovation
CCA Programs
• As CCAs continue to grow and
flourish in California, they are
advancing innovative, industry-leading projects and programs • Many are now focusing on DERs,
including Virtual Power Plants
• Several CCAs now operate DERMS
platforms to directly control
customer DERs and smart
appliances
10 CalCCA Positions on
Flexible Connection Topics
11 CalCCA Positions on
Flexible Connection Topics
• CCAs are open to engaging with IOUs to accelerate deployment of grid-edge capabilities • OpenADR is complementary to IEEE 2030.5, but more cost-effective, scalable and better-suited for demand flexibility events or pricing signals • The definition of ‘aggregator’ should not be limited to an IEEE 2030.5 cloud service provider
• DERs and grid-edge DERMS could serve as non-bridging solutions, subject to cost-effectiveness and equity considerations
• The IOUs should establish flexible connection pilot programs for secondary system customers
• Customers with variable operating envelopes should not be defaulted to dynamic rates
12 Optimizing Existing Grid Capacity Requires
Coordination Between IOUs and Aggregators
• Without grid-edge DERMS capabilities, DERs could inadvertently exacerbate local (circuit-, line-segment-, or substation-level) capacity constraints
• Leveraging all DERs, including non-IOU DERS, is crucial to optimizing existing grid capacity and helping lower costs
• CCAs have a unique role as not-for-profit LSEs that can both help reduce systems costs (e.g., RA, grid investments) and share value with ratepayers • However, CCAs need timely access to grid data and/or signals from IOUs to accomplish these grid-level objectives
13 OpenADR and IEEE 2030.5 are
Compatible and Complementary
• IEEE 2030.5 allows for precise control of individual inverter-based DER, but is not efficiently or economically scalable to millions of devices • IEEE 2030.5 is designed for inverter-based resources, but not smart appliances • OpenADR is better suited to delivering price- or event-based signals to aggregators and end-users, making it scalable and reducing the burden on DSOs
• OpenADR can convey precise device-level signals via IEEE 2030.5, SunSpec Modbus, Matter, or other interoperability standards.
• OpenADR can control inverter- and non-inverter-based resources • OpenADR offers greater flexibility and is more cost-effective to deploy at scale • Most IOUs already use OpenADR for their DR programs
• The use of OpenADR and IEEE 2030.5 is compatible and complementary 14 Aggregator Should Not Be Defined as
Merely IEEE 2030.5 Cloud Services
• Limiting ‘Aggregators’ to only IEEE 2030.5 cloud service providers could limit the participation of grid-edge DERMS providers using OpenADR or other communications protocols
• Many aggregators use other communications standards to control on-site inverters • Rule 21 does not preclude the use of other protocols
• While Rule 21 states the default application-level protocol shall be IEEE 2030.5, it also states that “other application-level protocols may be used by mutual agreement of the parties”
15 Dynamic Flexible Connections Should Be
Considered for Non-Bridging Solutions
16 • Non-bridging solutions should consider IOU and non-IOU solutions • CCAs are particularly well-positioned to provide non-bridging flexible connection services as community-led, not-for-profit load-serving entities • Any proposed solutions should be subject to cost-effectiveness evaluations and equity considerations
The IOUs Should Establish Flexible Connection
Pilots for Secondary Service Customers
• Grid-edge DERMS could enable participation of customers on the secondary system to participate in flexible connections • Grid-leveraging using OpenADR may prove to be more cost-effective than enabling solely IEEE 2030.5 options
• CCAs are open to exploring joint CCA-IOU pilot program opportunities
17 Customers Should Not Be Opted Into
Dynamic Rates
• Customers with variable operating envelopes should not be opted into dynamic rates
• Customers should be given information regarding available dynamic rate pilots and offerings, but should not be automatically enrolled, regardless of the ability to opt out
• Dynamic rates are still in pilot phases and pending evaluation on performance and cost-effectiveness
• Earlier evaluations did not find evidence of consistent or large changes in customer energy usage in response to hourly prices
• Requiring customers to opt-out of dynamic rates could create confusion and discourage participation in flexible service options 18 1
Comments of Marin Clean Energy, Pioneer Community Energy, San Diego Community Power, and San José Clean Energy on Proposed Revision Requests 1658 and 1659 Dear CAISO,
Marin Clean Energy, Pioneer Community Energy, San Diego Community Power, and San José Clean Energy (collectively the “Joint CCAs”) appreciate the opportunity to comment on CAISO’s Proposed Revision Requests (“PRRs”) 1658 and 1659. The Joint CCAs respectfully urge CAISO to withdraw PRR 1658, related to business processes for the Off Grid Charging Indicator (“OGCI”), a biddable parameter that permits co-located storage resources to elect not to “grid charge,” and its companion proposed revision, PRR 1659, for resources offering Flexible Resource Adequacy (Flexible “RA”). These revisions would undermine the purpose of a tariff amendment CAISO proposed and FERC approved in 2023; they reflect a policy proposal with significant ramifications for market participants that should not be contemplated through the Business Practice Manual Change Management Process via PRRs. Instead, these changes should be explored, if at all, through CAISO’s Policy Initiative process to ensure adequate and meaningful stakeholder engagement and feedback.

CAISO’s Energy Storage Enhancements – Phase 2 Initiative culminated in a Tariff Amendment to create a “Charging Constraint,” known as OGCI. OGCI created a biddable parameter reflecting a storage resource’s election not to bid beyond the output of its co-located Variable Energy Resource.1 FERC accepted the Tariff revisions on September 29, 2023.2 As CAISO explained at the time, local and federal taxes that incentivize renewable development discourage storage resources from “grid charging,” i.e., charging from sources other than a co-located renewable resource.3 This resulted in power purchase agreements that limit grid charging by storage resources. The stakeholder engagement process surfaced the delicate balance faced by storage resources with contractual grid charging limitations – must-offer obligations require them to bid economically or self-schedule, potentially resulting in a schedule to charge during a real-time interval, and resources are also required to follow dispatch instructions.4

CAISO therefore created a “new functionality for storage resources to help address some of the concerns voiced by stakeholders.”5In particular, “[t]his functionality will be electable on an hour by hour basis and offered to any co-located storage resources and will prevent ‘grid charging.’”6 CAISO implemented this new functionality by amending the Tariff to allow a 1 Aug1-2023-TariffAmendment-EnergyStorageEnhancements-Phase2-ER23-2537.pdf (“August 1, 2023 Tariff Amendment”).
2 184 FERC ¶61,209.
3 August 1, 2023 Tariff Amendment at page 4; Energy Storage Enhancements, Final Proposal (Oct 27, 2022) (“Final Proposal” available at California ISO - Energy storage enhancements) at page 20. 4Final Proposal at 21.
5 Id. at 21.
6 Id.; see also August 1, 2023 Tariff Amendment at page 5 (“The CAISO proposes to create a “Charging Constraint” co-located storage resources can use to help avoid grid-charging.”). 2 storage resource to “include Charging Constraints in its Bids.”7 “The Charging Constraints will reflect an election not to charge beyond the output of the co-located Variable Energy Resource(s) in a given hour.”8 At the same time, CAISO also amended the Tariff “to expressly allow co-located storage resources to deviate from any dispatch instructions that would conflict with their charging constraints.”9 As CAISO stated and FERC acknowledged, “the storage resources would still be subject to imbalance energy charges for the deviation but would not be subject to the other penalties in the CAISO Tariff.”10
In adopting the Tariff amendment, CAISO also acknowledged the limitations of OGCI. It made clear that OGCI “will not apply in operating intervals where the storage resources receive an award to provide regulation.”11 And it made clear that even where a storage resource elects OGCI, “the CAISO may still use its authority to issue exceptional dispatches to manage reliability.”12
FERC accepted the Tariff amendment based on its finding “that establishing the Charging Constraint bidding parameter improves the ability of co-located resources to avoid grid-charging and improves the accuracy of dispatch instructions to co-located storage resources. Moreover, by allowing co-located resources to avoid grid-charging, the proposed Tariff revisions will enhance the ability of co-located resource owners to manage grid-charging.”13 PRR 1658, however, now proposes to add language to the BPM for OGCI that would materially limit the application of OGCI and undercut CAISO’s stated purpose of OCGI. PRR 1658 states:

It would raise serious compliance questions [related to tariff section 37.3.1 for resources to submit feasible bids] if the co-located storage resource submitted charging bids with the parameter selected for hours after sunset. This is particularly so where the co-located storage resource is a Resource Adequacy Resource and may have submitted infeasible bids to meet the Resource Adequacy must-offer obligation.

The PRR strongly suggests that submission of bids during non-solar hours, with OGCI elected, would violate the Tariff’s feasible bid requirements. The PRR materially undercuts CAISO’s stated purpose in creating OGCI, which was “to prevent grid charging.”14 In amending the Tariff, CAISO carefully signaled to stakeholders the limitations of OGCI – it does not apply when a storage resource receives a regulation award and it does not preclude issuance of an 7 Tariff Section 30.5.6.3.
8 Id.
9 August 1, 2023 Tariff Amendment at page 6; see also Tariff Section 34.13.3. 10 184 FERC ¶61,209 at ¶12.
11 August 1, 2023 Tariff Amendment at page 6.
12Id.
13 184 FERC ¶61,209 at ¶25.
14 Final Proposal at 21.
3 Exceptional Dispatch.15 In proposing OGCI and updating its Tariff there was no indication that OGCI would not apply in non-solar hours. Both CAISO in proposing and FERC in approving the Tariff amendment spoke of OGCI in broad terms as allowing storage resources to avoid grid charging. Now, through a PRR CAISO is proposing a significant retreat from the purpose underlying the Tariff amendment. The proposed revisions amount to material policy changes, not mere clarifications, which should be addressed through the Policy Initiative process.

Specifically, if adopted, PRR 1658 would now appear to limit OGCI’s use to solar hours, and presumably to apply solely to the narrow circumstance when a storage resource is dispatched to charge based on forecast but actual generation deviates from forecast. This narrow application, while important, hardly addresses the broader needs of storage resources contractually restricted from grid charging, needs that CAISO stated OGCI was intended to address.

Particularly troubling is the suggestion that the election of OGCI during non-solar hours violates the Tariff’s feasible bid requirement. CAISO’s Final Proposal recognized that storage resources have must-offer obligations to submit bids.16 The 2023 Tariff amendment was carefully drafted to create OGCI as a “biddable parameter.”17 In particular, CAISO adopted Section

30.5.6.3 to permit a storage resource to “elect” not to grid charge and designate the election as a “Charging Constraint in its Bids.”18 By specifically including OGCI as a parameter in the bid, a charging bid with the parameter selected transparently conveys to CAISO that the resource is not available to grid charge. The feasible bid provision of the Tariff requires market participants to submit bids “from resources that are reasonably expected to be available and capable of performing at the levels specified in the Bid.”19 Under the plain language of the feasible bid provision, a charging bid with OGCI elected is feasible because the charging constraint is specified in the bid and the market participant is reasonably conveying that limitation. At the same time, the charging bid is not pointless; it still signals to CAISO the resource’s availability for an Exceptional Dispatch, which CAISO expressly indicated OGCI would not override.

In addition, the Tariff amendment authorizes deviations from dispatch instructions when OGCI is elected.20 As CAISO and FERC explained, where resources deviate from dispatch, “the storage resources would still be subject to imbalance energy charges for the deviation but would not be subject to the other penalties in the CAISO Tariff.”21 The Tariff amendment therefore clearly contemplated that storage resources would submit bids to charge that include an election not to grid charge, or there would be no need to authorize deviations from dispatch. And, the Tariff amendment created an explicit consequence when a resource that submits a bid with OGCI elected deviates from dispatch – uninstructed imbalance energy charges. It would be illogical to 15 August 1, 2023 Tariff Amendment at page 6.
16 Final Proposal at 21.
17August 1, 2023 Tariff Amendment at page 5 (emphasis added). 18 Tariff Section 30.5.6.3 (emphasis added).
19 Tariff Section 37.3.1.1 (emphasis added).
20 Tariff Section 34.13.3.
21 184 FERC ¶61,209 at ¶12 (emphasis added).
4 authorize storage resources to submit charging bids with OGCI elected, to deviate from dispatch, and to incur uninstructed imbalance energy charges under those circumstances, if those bids were actually infeasible bids that should not have been submitted at all.

For these reasons, PRR 1658 materially undercuts the CAISO’s 2023 Tariff amendment and represents a significant departure from the policy underlying that amendment to allow contractually restricted storage resources to avoid grid charging. Such a consequential policy shift is unsuited to a PRR and should instead be addressed in CAISO’s Policy Initiative process to ensure adequate and meaningful stakeholder feedback.

PRR 1659, the companion to PRR 1658, should be withdrawn for similar reasons. PRR 1659 appears to suggest that the mechanism for addressing the needs of storage resources contractually restricted from grid charging is not to use OGCI but instead to alter their Flexible RA showing. PRR 1659 also proposes to add language indicating that a resource’s Effective Flexible Capacity (“EFC”) value is its “maximum” value, suggesting that the MW quantity of flexible capacity a storage resource may qualify to provide is lower than its EFC. Disqualifying capacity from providing EFC based on when a Scheduling Coordinator could “reasonably foresee” an “operating limitation” is ambiguous. Operating limitations of OGCI resources are highly dependent upon CAISO market dispatches in real-time. The CAISO does not provide any examples of situations when a Scheduling Coordinator could “reasonably foresee” operating limitations at the time of the RA showing.

As discussed above, federal and state tax incentives resulted in power purchase agreements that contain contractual limitations on grid charging to incentivize development of renewable resources. CAISO observed in the course of the Energy Storage Enhancements Policy Initiative that such resources may face challenges in complying with Tariff provisions, referencing must-offer obligations and requirements to comply with dispatch instructions.22 CAISO’s Final Proposal did not suggest that the mechanism for addressing these compliance needs was for these resources instead to include in their RA Supply Plans a lower MW quantity of flexible capacity than their EFC. Such an approach would significantly upend market participants’ expectations, and like PRR 1658, PRR 1659 represents a highly consequential policy proposal that would make changes to rates, terms, or conditions of the Flexible RA product. It therefore is not a change that should be contemplated in the PRR process.

For the above stated reasons, the Joint CCAs respectfully request that CAISO withdraw PRRs 1658 and 1659. The changes proposed in these PRRs should be explored, if at all, through the CAISO’s Policy Initiative process to ensure adequate and meaningful stakeholder engagement and feedback.
22 Final Proposal at 21.
5 Respectfully Submitted,
Dated: February 18, 2026
/s/ Sabrinna Soldavini
Sabrinna Soldavini
VP of Policy

MARIN CLEAN ENERGY
1125 Tamalpais Avenue
San Rafael, CA 94901
Phone: (415) 464-6670
Email: [email protected]
/s/ Laura Fernandez
Laura Fernandez
Senior Director of Regulatory and Legislative Affairs
San Diego Community Power
PO Box 12716
San Diego, CA 92112-3716
Phone: 619-665-9296
Email: [email protected]
/s/ Lee Ewing
Lee Ewing
Legislative and Regulatory Manager

PIONEER COMMUNITY ENERGY
2510 Warren Drive, Suite B
Rocklin, CA 95677
Telephone: (916) 626-9909
Email: [email protected]
Heather Dauler
Deputy Director, Regulatory Compliance
and Policy
San José Clean Energy
200 E. Santa Clara St.

San José, CA 95113
Office: (408) 975-2619
Email: [email protected]
February 19, 2026
Via Electronic Email
California Public Utilities Commission – Energy Division Attention: Tariff Unit
505 Van Ness Avenue
San Francisco, California 94102
([email protected])
Re: Protest of Joint CCAs to Pacific Gas and Electric Company Advice Letter 7822-E – PG&E’s 2026 Low Carbon Fuel Standard Implementation Plan Dear Energy Division Tariff Unit:

Pursuant to the California Public Utilities Commission’s (“Commission”) General Order (“GO”) 96-B, Sonoma Clean Power (“SCP”), Peninsula Clean Energy Authority (“PCE”), San José Clean Energy (“SJCE”), and Marin Clean Energy (“MCE”) (“Joint CCAs”) hereby protest Pacific Gas and Electric Company’s (“PG&E”) Advice Letter (“AL”) 7822-E, 2026 Low Carbon Fuel Standard (“LCFS”) Implementation Plan.
1 AL 7822-E was submitted by PG&E on January
30, 2026. Pursuant to GO 96-B, the Joint CCAs submit this protest on the grounds that the “relief requested in the advice letter would violate statute or Commission order,” 2 and “is unjust,
unreasonable, or discriminatory[.]”3
AL 7822-E should not be approved as filed because PG&E’s 2026 LCFS Implementation Plan lacks the required implementation details in violation of prior Commission decisions.4 Furthermore, approval of AL 7822-E as filed would be unjust, unreasonable, and discriminatory to affected Community Choice Aggregators (“CCAs”) and their customers. PG&E’s proposed programs would operate on CCA-served electric load and would provide electric vehicle (“EV”) incentives to CCA customers, but the AL currently lacks the required implementation details to 1 References below to “General Rules” are to the general rules identified in GO 96-B. 2 General Rule 7.4.2(2).
3 General Rule 7.4.2(6).
4 Decision 20-12-027, Concerning Low Carbon Fuel Standard Holdback Revenue Utilization (December 17, 2020) (citing D.14-12-083)
https://docs.cpuc.ca.gov/PublishedDocs/Published/G000/M356/K223/356223853.PDF Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

ensure just, reasonable, and competitively neutral program implementation, including coordination with CCAs that operate EV programs affecting the same customers and load.

PG&E acknowledges5that dual participation in CCA managed charging programs is within the scope of the Commission’s Demand Response (“DR”) Order Instituting Rulemaking (“OIR”) (R. 25-09-004),6 however that proceeding does not have a timeline for addressing dual participation policy.7 Deferring a critical implementation process to a separate proceeding that explicitly lacks any near-term resolution is unjust, unreasonable, and risks conferring a competitive program advantage to PG&E.

Accordingly, the Joint CCAs request that Energy Division conditionally approve AL 7822-E, subject to the following conditions of approval:
• PG&E shall, in coordination with and approval by CCAs serving affected customers, develop and implement a CCA coordination process inclusive of an interim dual enrollment process that is applicable to all proposed enrollment channels, including any point-of-sale enrollment design; and
• Within ninety (90) days of approval of AL 7822-E, PG&E shall file a supplemental Tier 2 Advice Letter that includes the following:
o a description of the interim dual enrollment review process, which shall remain in effect until superseded by an applicable decision in the DR OIR; o a description of how the Residential Optimized Charging Program will be marketed in a manner competitively neutral to the customer’s LSE choice, including at the point-of-sale;
o updates to the Non-Residential Behind-the-Meter Infrastructure Incentive program identifying Level 1 charging as an eligible technology, including Level 1 charging in assigned MFH parking spaces.

BACKGROUND
On December 17, 2020, the Commission adopted Decision (“D.”) 20-12-027 Concerning Low Carbon Fuel Standard Holdback Revenue Utilization. The Decision requires utilities to file LCFS Holdback Implementation Plans, via a Tier 2 AL.8 The Decision requires Implementation Plans to include: (1) a proposal for at least one program, and (2) a description for how the large IOU plans to spend the rest of the funds, which shall include the status of the program 5 AL 7822-E, Attach. A, at pg. 28.
6 Rulemaking 25-09-004, Order Instituting Rulemaking to Enhance Demand Response in California (September 25, 2025).
https://docs.cpuc.ca.gov/PublishedDocs/Published/G000/M582/K072/582072320.PDF 7 R. 25-09-004, Assigned Commissioner’s Scoping Memo and Ruling (February 12, 2026), https://docs.cpuc.ca.gov/PublishedDocs/Efile/G000/M599/K038/599038857.PDF (scoping schedule through Q4 2026 does not include addressing dual participation policies). 8 D. 20-12-027, Conclusion of Law (“COL”) 17; Ordering Paragraph (“OP”) 3.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

development of the remaining program(s), an implementation timeline, and the approximate budget.9 Once a subsequent LCFS program is developed, utilities submit additional Implementation Plans, and as programs change or conclude over time, utilities submit revised Implementation Plans to reflect such updates.

Within each Implementation Plans, the utilities are required to specify:
1) How LCFS holdback expenditures are dedicated to equity projects or resiliency projects.
2) How each of its LCFS holdback expenditures and planned investments benefit current or future EV drivers in the state.
3) How the LCFS holdback expenditures comply with all other CARB regulations regarding the use of LCFS holdback funds (e.g., administrative cost caps and prohibited uses).
4) How any proposal for its LCFS holdback expenditure:
a. Demonstrates input from environmental justice groups and/or community-based organizations. b. Will address gaps in program design not already addressed through the utility’s transportation electrification (TE) expenditures or other publicly funded program, or in the alternative how the proposed expenditure will reduce costs to ratepayers.
c. Will address a barrier to TE, equity, and/or resiliency. d. Includes data collection requirements that allow for an evaluation of the effectiveness of the proposal in addressing TE, equity, and/or resiliency. 5) How any proposal for an equity project will be for the primary benefit of, or primarily serve, communities eligible for equity project expenditures. 6) How any proposal for a resiliency project is aligned with other TE-related utility resiliency efforts, including but not limited to, Public Safety Power Shutoffs (PSPS) and Wildfire Mitigation Plans, and reflects consultation with electric vehicle service providers, where appropriate.
7) How any proposal for a resiliency project aligns with Commission policy on vehicle-grid integration (VGI). 10 The Implementation Plans are also required to address general informational questions, as ordered by the Commission in D.14-12-083:
1) How will the large electrical corporation calculate the number of LCFS credits generated by each customer?
2) Who receives the revenue from the sale of LCFS credits? 3) How are LCFS revenue recipients identified?
4) How will the large electrical corporation calculate the amount of revenue to be distributed to each customer, if appropriate?
9 D. 20-12-027, at pg. 26.
10 D. 20-12-027, at pgs. 26 – 28.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

5) By what means is the revenue distributed to the customer and how frequently is revenue distributed?
6) How will vehicle ownership changes be identified, addressed, and tracked? 7) How will the large electrical corporation track and true-up revenues and disbursements from the program?
8) How will the program be marketed in a competitively neutral manner so that plug-in EV owners, regardless of their load serving entity, are aware that they are eligible to receive LCFS revenue?
9) How will the large electrical corporation receive and distribute credits generated by non-residential customers?
Pursuant to OP 3 of D.12-20-027, PG&E submitted its 2026 LCFS Implementation Plan via AL 7822-E, filed on January 30, 2026. Therein, PG&E proposes two new holdback programs:11
Proposed Program Description Program
Duration
Total Cost
Residential
Optimized Charging
Monthly incentive program to promote
customer enrollment in Optimized
Charging to reduce costs associated with
upgrading distribution grid assets
2027-2029 $31.30M
Non-Residential
Behind-the-Meter
Infrastructure
Incentive
Incentives to reduce the cost of EV charger
installation at non-residential sites
2027-2030 $31.99M
Residential Optimized Charging Program (i.e. managed EV charging) PG&E plans to market the Residential Optimized Charging program to new and existing EV customers with a residential level 2 (“L2”) charger (among other eligibility criteria, some of which has not yet been determined).12 PG&E plans to implement the program under a hybrid approach, which may include bundling “Original Equipment Manufacturer (OEM), third party, and/or EV retailer incentives and PG&E incentives (including optimized charging) into an offering available at the point of purchase or lease.” 13 PG&E states that if the Residential Optimized Charging program “proves successful, PG&E may request to extend the program in future LCFS filings, and potentially create a scaled, generally available program via the General Rate Case (GRC).”
14 PG&E acknowledges the current limitations for customers to participate in PG&E programs and CCA programs, and notes that this limits both the potential system value and the 11 AL 7822-E, Attach. A, at pg. 5, Table 1.
12 Id. at pg. 30.
13 Id. at pg. 26.
14 Id. at pg. 21.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

potential for customers to stack related incentives.15 PG&E’s Plan notes that while the Residential Optimized Charging program is focused on distribution system value, CCA programs are focused on generation value and resource adequacy, and that PG&E anticipates the DR OIR will address “some of the key issues to enable participating in multiple managed charging programs, including dual participation policy, valuation methodologies, and standardized data systems.”
16 Non-Residential Behind-the-Meter (BTM) Infrastructure Incentive PG&E plans to offer the Non-Residential Behind-the-Meter (“BTM”) Infrastructure Incentive to new or existing non-residential customers installing L2, DC fast chargers (“DCFC”), or Megawatt Charging System (“MCS”) chargers.17 PG&E’s Plan specifies that CCA customers are eligible to participate, and that the program is designed to work with other grant and incentive programs to maximize customer savings.18 PG&E also describes the relationship between the EV Advisory Services (“EVAS”) program, and the Commission’s recent authorization of the Technical Assistance program, stating that under this expansion, “EVAS will be able to support all customer segments that are eligible for the BTM incentive program.” Joint CCAs appreciate that the program design includes CCA customer eligibility and allows for incentive stacking. In the AL, PG&E provides a non-exhaustive list of other external relevant funding sources, however the Joint CCAs note that the example list is thus far limited to state-level incentives.19 The Joint CCAs ask that PG&E coordinate with CCAs as early as possible regarding CCA EV incentives and technical assistance programs, and propose doing so through the development of the communication channels and established practices envisioned for the Technical Assistance program in accordance with D. 22-11-040.20

JOINT CCA INTEREST
Joint CCAs are the default Load Serving Entities (“LSEs”) within PG&E’s service territory.
21 CCAs currently serve more electric load in PG&E’s service territory than PG&E does. As local public agencies, CCAs are also tasked with reducing GHG emissions associated with the electricity used by their customers. To that end, CCAs offer many programs aimed at achieving California’s transportation electrification goals including, but not limited to, the following EV programs:
• Managed EV charging programs
15 Id. at pg. 28.
16 Id.
17 Id. at pg. 54.
18 Id. at pg. 49.
19 Id. at pg. 50.
20 R. 18-12-006 Decision on Transportation Electrification Policy and Investment (November 17, 2022) at pg. 192 https://docs.cpuc.ca.gov/PublishedDocs/Published/G000/M499/K005/499005805.PDF 21 CCA customers receive generation services from their respective CCA, and receive transmission, distribution, billing, and other such services from PG&E.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

o Ava SmartHome Charging, launched in March 2025, is an EV managed charging program that automatically shifts home charging to off-peak times and when renewable energy is cheaper and more available. Eligible Ava customers who charge their EVs at home can earn both one-time rewards and monthly rewards per vehicle.22
o MCE Sync, launched in 2021, is an app-based managed EV charging and load shifting program that helps customers to reduce EV charging load during peak times, while also seeking to align EV charging load with high-solar daytime hours. The program provides eligible MCE customers with an enrollment bonus and monthly cash back for charging during low-carbon events.23 o SJCE's Peak Rewards for Smart Homes program, launched in May 2025, offers a Managed Charging option to automatically shift participant home charging to off-peak times. Customers that successfully shift their charging receive $3/month in rewards. In 2026, SJCE will add the Midday Charging Bonus to encourage customers to plug in during peak solar generation hours. Participants that maximize charging during these hours will be able to earn up to $160/year in rewards.24
• EV rebate programs
o PCE’s Used EV Rebate Program offers savings toward the purchase of a used plug-in hybrid or fully-electric vehicle for income-qualifying eligible PCE residents. In combination with PCE’s rebate, income-qualified residents may also qualify for additional incentives from regional and state programs. 25 o MCE’s EV Instant Rebate Program lowers the cost of purchasing or leasing EVs for income-qualified MCE customers. Launched in May 2023, the program offers savings on the purchase or lease of an eligible EV at participating dealerships.

This rebate may also be combined with other available incentives to further reduce the final vehicle costs.
26 o SJCE’s EV Instant Rebate Program provides income-qualified customers with discounts when purchasing or leasing a new or used EV from participating car dealerships. EV Instant Rebates can be stacked with other available incentives to make EVs accessible for all.27
• EV charger incentive programs
o MCE’s EV Charging program, launched in 2018, provides rebates for EV charging stations at multifamily properties, businesses, and community-serving public locations. The program also offers bonus incentives for stations powered 22 https://avaenergy.org/go-electric/electric-vehicles/smarthome-charging/ 23 https://mcecleanenergy.org/mce-sync/
24 https://sanjosecleanenergy.org/peak-rewards-for-smart-homes/ 25 https://www.peninsulacleanenergy.com/residential/rebates-offers/ev-rebate-program/ 26 https://mcecleanenergy.org/ev-rebate/
27 https://sanjosecleanenergy.org/ev-rebates/
Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

by 100% renewable energy service, and free technical assistance to help accelerate EV adoption.28
o PCE’s EV Ready program provides free technical support and rebates to help offset the costs of installing EV charging infrastructure at businesses and multifamily properties.
29 o PCE’s GovEV public fleet electrification program provides free technical assistance, project planning, and funding to public agencies for new EVs and charging infrastructure for their fleets.30
o SJCE’s Multifamily EV Charger Assistance Program, launched in 2024, provides incentives to multifamily properties to install Level 1 EV charging outlets, Level 2 EV charging outlets, and Level 2 EV charging ports31

PROTEST
The Joint CCAs submit this protest to AL 7822-E on the grounds that the requested relief is unjust, unreasonable, or discriminatory, and in violation of the Commission’s orders. Although the Joint CCAs appreciate PG&E’s efforts to support the growth of EV adoption, the Joint CCAs find that AL 7822-E, as filed, lacks the required implementation information required by D. 20-12-027 and D. 14-12-083. Additional implementation information is necessary to ensure that PG&E designs and implements the proposed LCFS holdback programs in a manner that is just, reasonable, and competitively neutral with respect to CCAs that operate EV programs affecting the same customers and load.

1. PG&E fails to provide the necessary implementation details regarding dual participation
In AL 7822-E, PG&E acknowledges that:
“[T]here are currently limitations for customers wanting to participate in PG&E and other managed charging programs (e.g., through a CCA). This limits the total system value that can be achieved (e.g., CCAs focus on generation value and resource adequacy and PG&E focus on value to the distribution system) and the ability for customers to stack value. PG&E anticipates that the CPUC Demand Response (DR) Order Instituting Rulemaking (OIR) (R.25-09-004) will address some of the key issues to enable participating in multiple managed charging programs, including dual participation policy, valuation methodologies, and standardized data systems.”32
28 https://mcecleanenergy.org/ev-charging/
29 https://www.peninsulacleanenergy.com/business/rebates-offers-business/ev-ready-program/ 30 https://www.peninsulacleanenergy.com/public-organization/govev-program-public-fleet-electrification/ 31 https://sanjosecleanenergy.org/ev-charger-assistance/ 32 AL 7822-E, Attach. A, at pg. 28.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

PG&E further states that it “will determine additional eligibility criteria during program development ... [which] could include…[meeting] any dual participation requirements (e.g., checks for participation in other PG&E programs and other managed charging programs).”33 While the Joint CCAs appreciate that PG&E acknowledges the potential for future value stacking of managed EV charging programs,
34 and the need for dual participation checks in the
meantime, the Joint CCAs submit this protest due to the lack of required implementation detail within AL 7822-E regarding CCA coordination and dual enrollment. PG&E has correctly noted that dual participation policy falls within the scope of the DR OIR. However, that proceeding has no defined timeline for addressing dual participation. The recently released Scoping Ruling outlines a schedule through Q4 of 2026, and it does not include dual participation. Thus, it is unreasonable to defer a critical program detail and implementation process to a separate proceeding that explicitly lacks any near-term resolution.

Since 2024, PG&E has been testing managed charging through the pilot-scale EV Charge Manager (“EVCM”) program, which is currently slated to run through 2026. To date, PG&E has coordinated with MCE and Ava in an informal manner to develop and implement a dual enrollment check to ensure that customers are enrolled in only one managed charging program of their choice. Ava and MCE are appreciative of PG&E’s past coordination on the dual enrollment review for the EVCM pilot. Due to the size of the EVCM pilot, an informal process has generally been administratively manageable.

However, the informal dual enrollment process developed for the pilot-scale EVCM program likely will not suffice for the size and scale of PG&E’s Residential Optimized Charging program as proposed in AL 7822-E. If successful, PG&E also hopes to further scale the program in future LCFS or GRC filings, which further highlights the need for additional implementation details on dual enrollment at the early stages of program design. Ava and MCE have endeavored to be good partners with PG&E given the clear benefits that EV managed charging programs provide to our customers. The Joint CCAs look forward to continuing that collaborative partnership to develop and implement a coordinated dual enrollment process for the Residential Optimized Charging program, to govern in the interim until superseded by a Commission decision on dual participation policy in the DR OIR.

2. PG&E fails to describe how the Residential Optimized Charging program will be marketed in a competitively neutral manner
Under current LCFS regulations, electric distribution utilities (“EDUs”) are the exclusive credit generator of base credits for residential loads, including the “holdback credits” which serve as the primary funding for the two new programs proposed by PG&E in AL 7822-E. 33 Id. at pg. 31.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

In prior comments to CARB, Joint CCAs encouraged CARB to revise the LCFS base credit provisions to define an LSE as the base credit generator, instead of just an EDU.35 Since the LCFS is intended to lower the GHG intensity of transportation fuels, CCAs argued that it is reasonable that an LSE should receive the base credit for displacing diesel and gasoline, and in turn, that residential base credits should be allocated to the LSE that procures the energy for and serves the credit-generating EV load. Unfortunately, these arguments have been unsuccessful to date. As such, PG&E alone receives LCFS credits for residential EV charging customers, and CCAs have no direct access to those credits even though a significant portion of those credits are generated by EV charging from CCA-served load.

This inequitable base credit structure stresses the importance of the requirement that all Implementation Plans shall describe how “the program be marketed in a competitively neutral manner so that plug-in EV owners, regardless of their load serving entity, are aware that they are eligible to receive LCFS revenue.”36
In AL 7822-E, PG&E proposes implementing an incentive bundling design for the Residential Optimized Charging program that would potentially enroll customers at the point of vehicle purchase or lease. However, the AL does not demonstrate how PG&E will market this program in a competitively neutral manner consistent with D. 14-12-083. Specifically, PG&E does not address dual enrollment or coordination with CCA programs when marketing its program at the point-of-sale (“POS”). Lack of coordination will lead to customer confusion, administrative challenges, and enrollment barriers. Moreover, a POS enrollment process that lacks the necessary CCA coordination safeguards runs the risk of conferring a competitive advantage onto PG&E by way of customer capture and lock-in at arguably the most influential customer decision point. Maintaining competitive neutrality between PG&E and CCA programs is particularly important as CCAs have interest in leveraging managed charging and other demand response strategies to provide value to the distribution system, 37 targeting the same
value stream as PG&E’s Residential Optimized Charging program.

The Joint CCAs recognize that PG&E plans to explore POS incentive bundling under the 2026 scope of the EVCM pilot, and that the potential implementation of bundled POS enrollment for the Residential Optimized Charging program depends on the success of those EVCM pilot efforts.38 However, the Joint CCAs are concerned that without modification, approval of AL 7822-E as filed would constitute approval of a POS enrollment program design that lacks the necessary implementation and coordination details and safeguards. A bundled POS enrollment structure materially differs from existing standard enrollment channels, and would significantly affect customer choice, coordination with CCAs, and operational outcomes. 32 Comments of the Joint CCAs on Potential Future Changes to the LCFS Program (January 7, 2022) https://sdcommunitypower.org/wp-content/uploads/2022/09/LCFS-Comments.pdf 36 D.14-12-083, pg. 32-33
37 See e.g., Ava Community Energy Opening Comments on DR OIR (November 13, 2025) https://docs.cpuc.ca.gov/PublishedDocs/Efile/G000/M587/K122/587122342.PDF 38 AL 7822-E, Attach. A, at pg. 20-21.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

3. PG&E fails to include Level 1 charging without a reasoned analysis PG&E’s proposed Non-Residential Behind-the-Meter Infrastructure Incentive program (“BTM Program”) identifies L2 and DCFC as eligible technologies but does not include Level 1 (“L1”) charging. In the Program Overview, PG&E describes the BTM Program as providing incentives “to reduce the cost of EV charger installation at non-residential sites.”39 The AL specifies the charger types that are eligible for incentives but does not list L1 chargers.40 The Implementation Plan does not provide any explanation for excluding L1 chargers.

This exclusion is consequential because the charging needs at MFH properties differ materially from other applications. While PG&E explains how the BTM Program supports fleets and public charging applications, it does not provide a comparable analysis of MFH charging needs or evaluate whether L1 may more efficiently address those needs.

At MFH properties, vehicles are typically parked for extended periods. Based on census data analyzed by PCE, most drivers park their vehicles for at least 12 hours per day.41When EVs are plugged into L2 chargers at MFH properties, vehicles often remain parked overnight by may only draw electricity for fewer than three hours. L1 charging delivers lower power over a longer dwell period, which is sufficient to meet driver needs and aligns with overnight residential charging patterns.

Excluding L1 from eligibility therefore disregards a technology that aligns with observed residential charging behavior and risks over-sizing infrastructure relative to actual demand.

Including L1 would advance the BTM Program’s stated objectives. PG&E identifies the BTM Program’s purpose as reducing installation costs and increasing access to charging.42 PG&E further proposes to provide incentives covering up to 75% of the eligible BTM project costs.43 PCE’s EV Ready program provides MFH charging incentives of up to 100% of costs, or $2,500 per L1 charger, and requires a minimum of a 25% cost share for the property for L2 chargers. To date, the average cost, including both equipment and installation, for MFH projects is $2,700 per L1 charger versus $8,000 per L2 charger. Because L1 chargers are materially less expensive, a fixed program budget would support a greater number of individual chargers if L1 were eligible. In other words, including L1 would allow the BTM Program to maximize charging access within the same budgetary limits.

The infrastructure implications are equally important. Higher-power infrastructure is more likely to trigger the need to upsize the electrical service to serve the new charging project. The costs of upsizing the services, as well as any upstream distribution work that they implicate, are recovered through electric rates. By contrast, L1 charging can often be deployed within the property’s existing service capacity. PCE completed a project in 2025 that delivered charging for all 143 residential units at 39 AL 7822-E, Attach. A, at pg. 41.
40 AL 7822-E, Attach. A, at pg. 54-55.
41 https://www.peninsulacleanenergy.com/wp-content/uploads/2021/09/Determining-the-Appropriate-Level-of-Power-Sharing-for-EV-Charging-in-Multifamily-Properties-1.pdf. 42 AL 7822-E, Attach. A, at pg. 41.
43 AL 7822-E, Attach. A, at pg. 48.

Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

a MFH property using smart L1 outlets and without needing to upsize the utility electrical service.44 The project illustrates charging access can be delivered at scale with infrastructure that is right-sized to meet customer needs rather than higher-power infrastructure that may require service upsizing.

L1 charging also advances the Commission’s stated focus on removing barriers to transportation electrification deployment and support timely energization.45 Projects that avoid service upsizing entirely can be energized more quickly and would enable faster deployment of MFH charging projects.

Allowing L1 as an eligible technology would therefore enable faster deployment of MFH charging projects and allow PG&E to prioritize more complex projects where service upsizing and distribution upgrades are unavoidable.

Finally, the BTM Program should provide incentives for L1 chargers installed in assigned parking spaces. PG&E suggests that the BTM Program will only be available for shared MFH chargers. “For public charging and shared private (i.e., MFH and workplace) charging, the program aims to increase access to EV charging in these locations by accelerating the widespread installation of charging through BTM incentives to reduce development costs.”46 However, assigned parking installations are the most effective way to ensure equitable and reliable access for MFH residents.

Dedicated charging spaces reduce conflicts among residents, limit the need for property owners to manage the residents’ use of the chargers, avoids triggering costly accessibility upgrades that often require reducing available parking for MFH residents, and provide predictable access to overnight charging.

The California Energy Commission’s (“CEC”) Communities in Charge program allows incentives for L1 installations in MFH settings, including installations serving assigned parking spaces.47 This demonstrates that assigned L1 charging in MFH properties is administratively feasible and consistent with state policy to expand EV charging access. Therefore, the BTM Program Implementation Plan should make it clear that incentives will be available for L1 chargers installed in assigned MFH parking spaces, not solely for shared configurations.

For these reasons, the exclusion of L1 charging from the BTM program is inconsistent with actual MFH charging behavior, undermines the Program’s stated goals of cost reduction and expanded access to charging, increases the likelihood of unnecessary and costly infrastructure upsizing, may slow deployment timelines, and contradicts recent developments in state CEC programs that are now encouraging L1 charging. Including L1 charging, and clarifying eligibility for assigned parking installations, would better align the BTM program with observed MFH charging needs and the Commission’s policy objectives.
44 https://www.peninsulacleanenergy.com/news-releases/local-leaders-celebrate-groundbreaking-bayview-condos-ev-charging-installation/ 45 D.25-12-005, Decision Adopting Revised Data Gathering And Reporting Requirements For Transportation Electrification Programs And Providing Clarification On Programs Adopted In Decision 22-11-040, at pg. 15.
46 AL 7822-E, Attach. A, at pg. 45.
47 https://thecommunitiesincharge.org/
Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

CONCLUSION
For the foregoing reasons, the Joint CCAs request that Energy Division conditionally approve AL 7822-E, subject to the following:
• PG&E shall, in coordination with and approval by CCAs serving affected customers, develop and implement a CCA coordination process inclusive of an interim dual enrollment process that is applicable to all proposed enrollment channels, including any point-of-sale enrollment design; and
• Within ninety (90) days of approval of AL 7822-E, PG&E shall file a supplemental Tier 2 Advice Letter that includes the following:
o a description of the interim dual enrollment review process, which shall remain in effect until superseded by an applicable decision in the DR OIR; o a description of how the Residential Optimized Charging Program will be marketed in a manner competitively neutral to the customer’s LSE choice, including at the point-of-sale;
o updates to the Non-Residential Behind-the-Meter Infrastructure Incentive program identifying Level 1 charging as an eligible technology, including Level 1 charging in assigned MFH parking spaces.

The Joint CCAs thank the Commission for its consideration of this protest.

Respectfully,
/s/ Jordyn Bishop
Jordyn Bishop
Policy Counsel
MCE
1125 Tamalpais Ave
San Rafael, CA 94901
[email protected]
/s/ Neal Reardon
Neal Reardon
Director, Regulatory Affairs
Sonoma Clean Power Authority
431 E. Street
Santa Rosa, CA 95404
[email protected]
/s/ Matthew DS Rutherford
Matthew DS Rutherford
Manager of Regulatory Policy

PENINSULA CLEAN ENERGY

AUTHORITY
2075 Woodside Road
Redwood City, CA 94061
[email protected]
/s/ Claire Huang
Claire Huang
Regulatory Analyst II
Ava Community Energy
1999 Harrison St
Oakland, CA 94612
[email protected]
Protest of Joint CCAs to PG&E’s Advice Letter 7822-E

/s/ Kayla Baum
Kayla Baum
Regulatory Compliance and Policy Specialist
San José Clean Energy
200 E Santa Clara Street
San Jose, CA 95113
[email protected]
cc: Sidney Bob Dietz II c/o Megan Lawson, PG&E ([email protected]) Service List of R.23-12-008

MARCH FILINGS
Docket No. R.25-02-005
Exhibit No.

Date March 2, 2026
Witness Brian Dickman
TRACK 2 DIRECT TESTIMONY OF BRIAN DICKMAN

ON BEHALF OF

THE CALIFORNIA COMMUNITY CHOICE ASSOCIATION

ORDER INSTITUTING RULEMAKING TO UPDATE AND REFORM ENERGY

RESOURCE RECOVERY ACCOUNT AND POWER CHARGE INDIFFERENCE

ADJUSTMENT POLICIES AND PROCESSES
i
Table of Contents
I. INTRODUCTION AND SUMMARY .................................................................................... 1 II. BACKGROUND ...................................................................................................................... 4 A. The Legislature Enabled Community Choice Aggregation to Promote Customer Retail Choice and Competition ..................................................................................................... 4 B. California Law Requires All Customers to Remain Indifferent as a Result of Customers Departing IOU Generation Service ..................................................................................... 4 C. The Commission Adopted the PCIA to Implement the Legislature’s Indifference Requirements ...................................................................................................................... 6 D. RPS-Eligible Resources Produce Value Through RECs the IOUs Use Immediately or Bank for Later RPS Compliance ......................................................................................... 8 III. LATER DEPARTING CUSTOMERS CURRENTLY RECEIVE NO BENEFIT FOR PRE-2019 BANKED RECS NOW BEING USED BY THE IOUS FOR CURRENT BUNDLED CUSTOMER RPS COMPLIANCE ..................................................................................... 10 IV. VALUATION OR ALLOCATION OF PRE-2019 BANKED RECS PREVENTS COST SHIFTS AND MAINTAINS INDIFFERENCE AS REQUIRED BY STATUTE ............. 16 A. Later Departing Customers Should Receive Value at the Current RPS MPB, Through a Vintage-Specific Credit to the PCIA ................................................................................ 17

1. The Best Measure of Value Accruing to Current Bundled Customers When the IOU Uses Pre-2019 Banked RECs for Compliance is the Current RPS MPB ...................... 17

2. Providing Vintage-Specific Value of Pre-2019 Banked RECs Ensures Only the Right Customers are Credited .................................................................................................. 20 B. Allocation of Pre-2019 Banked RECs to Later Departing Customers is an Alternative Valuation Method Consistent with Indifference Requirements ........................................ 21 V. CALCCA’S PROPOSAL IS CONSISTENT WITH APPLICABLE LAW AND COMMISSION PRECEDENT (SCOPING RULING ISSUE 1) ........................................ 24 A. Statutory Indifference Principles, the Mandate to Prevent Cost Shifts, and Prior Commission Decisions All Require that Later Departing Customers Receive Their Share of Value for Bundled Customer Use of Pre-2019 Banked RECs (Scoping Ruling Issue

1.a.) ................................................................................................................................... 25 B. No Policy Considerations or Other Downstream Consequences Negate the Commission’s Statutory Obligation to Maintain Indifference and Ensure Charges are Just and Reasonable (Scoping Memo Issue 1.b.) ............................................................................ 26 C. Differences Between Pre-2019 and Post-2018 Banked RECs Do Not Impact the Reasonableness of Assigning, or the Commission’s Obligation to Assign, Value to the Pre-2019 RECs (Scoping Issue 1.c.) ................................................................................. 26 D. Banking of RECs Under the Pre-2019 Rate Methodology Does Not Result in Forfeiture of Value by Later Departing Customers (Scoping Issue 1.d.) .......................................... 27 VI. THE IOUS SHOULD BE DIRECTED TO APPLY CALCCA’S VALUATION OR ALLOCATION PROPOSAL (SCOPING ISSUE 2) .......................................................... 28 ii
A. Using Current RPS MPB Data is Reasonable and Required to Maintain Indifference (Scoping Issue 2.a.) ........................................................................................................... 28 B. The Only Impact the Pre-2019 PCIA Methodology Has on How the Commission Should Value or Allocate Pre-2019 Banked RECS to Customers Now is That Such Valuation Must be Vintage-Specific to Avoid Double-Payment to Previously Departed Customers (Scoping Issue 2.b.) ........................................................................................................... 28 C. CalCCA’s Valuation Proposal Would Not Require Changes to PCIA Ratemaking, Balancing Accounts, and/or Tariffs (Scoping Issue 2.c.) ................................................. 31

ATTACHMENTS
Attachment A: Pre-2019 Banked RECs Numerical Example
Attachment B: Curriculum Vitae of Brian Dickman
Attachment C: Select Discovery Responses
1 1 I. INTRODUCTION AND SUMMARY
The California Community Choice Association (CalCCA)
1 presents this Track 2 2
3 direct testimony in the Order Instituting Rulemaking to Update and Reform Energy 4 Resource Recovery Account and Power Charge Indifference Adjustment Policies and Processes (OIR). This testimony was prepared on behalf of CalCCA by Brian Dickman, 5 Partner, NewGen Strategies and Solutions, LLC. Mr. Dickman’s qualifications are set forth 6 in Attachment B. 7
California law obligates the Commission to ensure that both investor-owned utility 8 (IOU) bundled customers and departed community choice aggregator (CCA) and electric 9 service provider (ESP) unbundled customers remain indifferent to retail customer choice. 10 That is, no one customer group should experience a shift of costs in their direction caused 11 by the other customer group. Track 2 is considering how to address a discrete cost shift 12 resulting from one cost and value stream within the Power Charge Indifference Adjustment 13 (PCIA): certain unbundled customers paying, and receiving no value in return, for 14 15 renewable energy credits (REC) generated and banked prior to 2019 (Pre-2019 Banked 16 RECs) that are being used in 2026 and later for bundled customer (Current Bundled Customer) Renewables Portfolio Standard (RPS) compliance. The unbundled customers 17 at issue are those that depart bundled service after the Pre-2019 Banked RECs were 18 generated and banked (Later Departing Customers), as discussed in detail below. 19 1 California Community Choice Association represents the interests of 24 community choice electricity providers in California: Apple Valley Choice Energy, Ava Community Energy, Central Coast Community Energy, Clean Energy Alliance, Clean Power Alliance of Southern California, CleanPowerSF, Desert Community Energy, Energy For Palmdale’s Independent Choice, Lancaster Energy, Marin Clean Energy, Orange County Power Authority, Peninsula Clean Energy, Pico Rivera Innovative Municipal Energy, Pioneer Community Energy, Pomona Choice Energy, Rancho Mirage Energy Authority, Redwood Coast Energy Authority, San Diego Community Power, San Jacinto Power, San José Clean Energy, Santa Barbara Clean Energy, Silicon Valley Clean Energy, Sonoma Clean Power, and Valley Clean Energy. 2 Commissioner John Reynolds’ February 3, 2026, Scoping Ruling determined the 1 following issues are to be considered in Track 2 of this proceeding:2 2

1. Is the proposal to value Pre-2019 Banked RECs at a value other than zero 3 dollars consistent with applicable law and Commission precedent? This 4 question includes but is not limited to the following sub-questions: 5 a. Do the indifference principles and mandate to prevent cost shifts contained 6 in Public Utilities Code Sections 365.2, 366.2, and 366.3 and prior 7 Commission Decisions require that when an [IOU] uses Pre-2019 Banked 8 RECs for compliance on behalf of its bundled service customers, Later 9 Departing Customers must receive a credit from bundled service customers 10 for the use of the Pre-2019 Banked REC? 11
b. Are there potential “downstream” consequences or other policy 12 considerations stemming from the proposal to value Pre-2019 Banked 13 RECs at a value other than zero dollars that would render the proposal 14 unreasonable pursuant to Public Utilities Code section 451, or any other 15 statute or Commission decision? 16
c. Are there characteristics of RECs generated prior to January 1, 2019, that 17 make them categorically different from RECs generated after January 1, 18 2019, impacting whether it is reasonable for the Commission to adopt a 19 valuation methodology for Pre-2019 Banked RECs that assigns them other 20 than a zero-dollar valuation? 21
d. Did bundled service customers who paid the cost of RECs at the time they 22 were generated and banked under the methodology in effect prior to 23 adoption of the methodology adopted in D.19-10-001 permanently give up 24 their claim to those RECs if they departed from bundled service while those 25 RECs were still retained by an IOU? 26
27 2. If the answer to Issue 1 is determined to be yes, should the Commission direct IOUs to apply a value other than zero dollars when Pre-2019 Banked RECs are 28 valued for ratemaking in the 2026 and later ERRA Forecast proceedings? If so, 29 how should such value be determined and allocated while adhering to 30 indifference principles for all customers? This question includes but is not 31 limited to the following sub-questions: 32
a. Is it reasonable for RECs generated prior to January 1, 2019, to be valued 33 using MPBs developed using data from later years? If not, is there another 34 value that would be appropriate? 35
b. Does the methodology used to allocate the value of RECs, expenses, or 36 2 Assigned Commissioner’s Scoping Memo and Ruling, Rulemaking (R.) 25-02-005 (February 3, 2026), at 3-5 (Scoping Ruling).
3 revenues from RPS-eligible resources prior to 2019 impact the manner or 1 extent to which the Commission should allocate the value of Pre-2019 2 Banked RECs to bundled and unbundled customers, and, if so, how? 3 c. If the Commission is to establish a valuation for Pre-2019 Banked RECs 4 other than zero dollars, how should the Commission modify the PCIA 5 ratemaking, balancing accounts, and/or tariffs to effectuate the valuation? 6 To ensure Later Departing Customers remain indifferent, CalCCA recommends 7 that Pre-2019 Banked RECs be valued at the current RPS market price benchmark (MPB) 8 when used for bundled customer compliance. The current RPS MPB best measures the cost 9 avoided when bundled customers rely on banked RECs rather than procuring additional 10 RECs in the market. That value should be provided to Later Departing Customers by 11 crediting the PCIA in the customer vintage corresponding to the year the RECs were 12 13 generated. Alternatively, Pre-2019 Banked RECs that were paid for by Later Departing Customers could reduce the Net RPS Procurement Need for the load serving entities (LSE) 14 that serve those customers when those Pre-2019 Banked RECs are used for compliance. 15 My testimony describes CalCCA’s recommendations in detail and addresses each 16 of the issues identified in the Scoping Ruling. Specifically, my testimony: 17 • Explains how the PCIA framework operates to ensure statutory indifference 18 between bundled and unbundled customers, including how REC value is 19 conveyed to customers; 20
• Demonstrates that Later Departing Customers paid for their share of Pre-21 2019 Banked RECs when those RECs were generated but have not received 22 the value of those RECs when they are later used for Current Bundled 23 Customer RPS compliance; 24
• Shows that failure to credit Later Departing Customers when Pre-2019 25 Banked RECs are used for Current Bundled Customer compliance results 26 in a cost shift from Current Bundled Customers to Later Departing 27 Customers; 28
• Recommends that Pre-2019 Banked RECs be valued at the current RPS 29 MPB – the avoided cost of procuring new RECs in the market – when used 30 for Current Bundled Customer compliance, with the full value recorded as 31 4 1 a vintage-specific credit to the PCIA corresponding to the year the RECs were generated; 2
• Explains why vintage-specific crediting ensures only the correct customer 3 vintages receive the correct values and prevents double compensation; 4 • Concludes that CalCCA’s proposal is consistent with Commission 5 precedent, maintains statutory indifference, and can be implemented within 6 the existing PCIA framework; and that 7
• In the absence of valuation, the Commission is required pursuant to Public 8 Utilities Code section 366.2(g) to provide an allocation to Later Departing 9 Customers for their share of RECs used by the IOU of Pre-2019 Banked 10 RECs, which could be achieved by reducing the Net RPS Procurement Need 11 as measured on the Commission’s Renewable Net Short Calculation 12 template for the LSEs serving those Later Departing Customers. 13 14 II. BACKGROUND
15 A. The Legislature Enabled Community Choice Aggregation to Promote 16 Customer Retail Choice and Competition
Through the passage of Assembly Bill (AB) 117 in 2002, the California Legislature 17 broadened opportunities for retail electric competition previously allowed only through 18 Direct Access (DA) service via ESPs by authorizing Community Choice Aggregation. 19 Instead of receiving all services (i.e., generation, transmission, distribution, and other 20 services) from their IOU (Bundled Customer), CCA customers receive generation service 21 from their local CCA, and continue receiving transmission, distribution, and all other 22 services from the IOU (Unbundled or Departed Load Customer).3 Today, 25 CCAs are 23 in operation in California serving more than 15 million customers. 24 25 B. California Law Requires All Customers to Remain Indifferent as a Result of 26 Customers Departing IOU Generation Service
California law requires customer indifference when customers depart IOU 27 generation service. In other words, no customer – either bundled or unbundled – should 28 3 DA/ESP customers are also Unbundled or Departed Load Customers. 5 have additional costs shifted to them as a result of retail choice. California statutes require 1 CCA customers to continue paying the above-market costs of IOU resources originally 2 procured on their behalf, even after they depart.4 In return, CCA customers are entitled to 3 receive the value of the resources’ attributes that remain with bundled customers or an 4 allocated share of the attributes, to compensate the unbundled customers for the payments 5 they continue to make for those resources.5 These two components of indifference are 6 shown in Figure 1 below: 7
Figure 1: Statutory Indifference Requirements 8
9 4
Cal. Pub. Util. Code §§ 365.2 (“The commission shall ensure that bundled retail customers of an electrical corporation do not experience any cost increases as a result of retail customers of an electrical corporation electing to receive service from other providers. The commission shall also ensure that departing load does not experience any cost increases as a result of an allocation of costs that were not incurred on behalf of the departing load.”); 366.2(a)(4) (“The implementation of a community choice aggregation program shall not result in a shifting of costs between the customers of the community choice aggregator and the bundled service customers of an electrical corporation.”); 366.3 (“Bundled retail customers of an electrical corporation shall not experience any cost increase as a result of the implementation of a community choice aggregator program. The commission shall also ensure that departing load does not experience any cost increases as a result of an allocation of costs that were not incurred on behalf of the departing load.”).
5 Cal. Pub. Util. Code § 366.2(g) (“Estimated net unavoidable electricity costs paid by the customers of a community choice aggregator shall be reduced by the value of any benefits that remain with bundled service customers, unless the customers of the community choice aggregator are allocated a fair and equitable share of those benefits.”).
6 1 C. The Commission Adopted the PCIA to Implement the Legislature’s 2 Indifference Requirements
The PCIA is the tool the Commission adopted consistent with the statutes 3 referenced above “to equalize cost sharing” between these two groups of customers and 4 ensure indifference.6 Therefore, when customers of IOUs depart from bundled service and 5 receive their electricity from a non-IOU provider, such as a CCA, “those customers remain 6 responsible for costs previously incurred on their behalf by the IOUs — but only those 7 costs.”7 Consistent with these indifference statutes, the PCIA also conveys to departed load 8 customers the value of resource attributes retained by the IOU for bundled customers. 9 10 PCIA rates are established to recover the IOU’s Indifference Amount, or the above-market costs related to PCIA-eligible resources. The Indifference Amount is the difference 11 between the cost of the IOU’s supply portfolio and the market value of the IOU’s supply 12 portfolio as demonstrated in Figure 2. 13
Figure 2: Indifference Amount Calculation 14
15 The PCIA therefore tracks both: (a) the costs of resources; and (b) the value of the 16 benefits provided by the IOUs’ PCIA-eligible portfolios and retained by the IOU for 17 bundled customers. Since D.11-12-018, the value of the IOUs’ supply portfolio has been 18 derived from administratively determined proxy values that are intended to estimate the 19 6 See Decision (D.) 18-10-019, at 3.
7 Scoping Memo and Ruling of Assigned Commissioner, R.17-06-026 (Sept. 25, 2017), at 2; see also D.18-10-019, Decision Modifying the Power Charge Indifference Adjustment Methodology, R.17-06-026 (Oct. 11, 2018), at 3.
7 market value of the resource portfolio attributes: Energy Value, RPS Value, and Resource 1 Adequacy (RA) Value. 8 These values are calculated by multiplying eligible resource 2 output (quantities) by the relevant MPBs (prices). 3
Each generation resource and customer is assigned a “vintage.” A distinct portfolio 4 of generation resources is identified for each vintage year based on when a commitment to 5 procure each resource was made. Customers are assigned vintage years according to the 6 date the customer departed bundled IOU service. Customers continuing to receive bundled 7 service from the IOU are included in the latest vintage (e.g., bundled customers are 8 currently vintage 2026). An Indifference Amount is calculated for each vintage (summing 9 portfolio costs and values), and all customers are responsible for the cumulative 10 Indifference Amount for years prior to and including their vintage. Thus, when a cost is 11 included in a particular vintage, customers in that vintage and all later vintages are 12 responsible for that cost. Customers in prior vintages are not responsible for that cost. 13 Conversely, when a credit is attributed to a particular vintage, customers in that vintage 14 and all later vintages receive the credit. Customers in prior vintages do not receive the 15 credit. 16
17 This can be seen in Figure 3 below, which provides a summary of the latest Indifference Amount forecast for Pacific Gas and Electric Company (PG&E) by vintage: 18 8 See, e.g., D.11-12-018, at 9-10, Ordering Paragraph (OP) 5, OP 8; D.19-10-001, at 6 (“Market Value is the estimated financial value, measured in dollars, that is attributed to a utility portfolio of energy resources for the purpose of calculating the Power Charge Indifference Adjustment for a given year.”).

D.23-06-006 subsequently also required Energy Division to calculate a separate adder for Greenhouse-Gas Free (GHG-Free) energy.
8 1 Figure 3: PG&E 2026 Indifference Amount by Vintage ($000) 2 For the 2015 vintage, for example, a market value credit of $11.5 million is subtracted from 3 a portfolio cost of $6.1 million, producing a negative indifference amount for the 2015 4 vintage of ($5.4 million). However, customers in the 2015 vintage do not pay PCIA rates 5 based solely on that negative indifference amount. They pay PCIA rates based on the sum 6 7 of that indifference amount plus the vintage indifference amounts for the years before 2015, i.e., the vintage indifference amounts to the left of the 2015 column in Figure 3 above. 8 Customers in the 2015 vintage do not pay the vintage indifference amounts for any year 9 after 2015, i.e., those amounts to the right of the 2015 column in Figure 3 above. Thus, if 10 a credit is made to the 2016 vintage, customers in the 2015 vintage (and earlier) do not 11 receive that credit; only customers in the 2016 vintage and later will receive the credit. 12 13 D. RPS-Eligible Resources Produce Value Through RECs the IOUs Use 14 Immediately or Bank for Later RPS Compliance
RPS Value, as one of the components of Utility Portfolio Value, estimates the 15 financial value attributable to RECs generated by PCIA-eligible resources and retained by 16 the IOU for bundled customer RPS compliance obligations.9 This RPS Value is determined 17 by applying the RPS MPB to the quantity of RECs used by the IOU.10 To the extent an 18 IOU uses RECs to meet the needs of bundled customers, the PCIA design requires that 19 bundled customers ‘purchase’ departed load customers’ share of the RECs to convey the 20 9 D.19-10-001, at 6, 26; D.11-12-018, at 8, 10, Conclusion of Law (COL) 3. 10 RPS Value may also include revenue received when the IOU’s RECs are sold to third parties.

UOG Legacy 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 Total Portfolio Costs $910,406 $1,439,267 $432,238 $127,428 $160,483 $49,165 $3,672 $6,121 $1,775 $9,342 $482 $37,053 $496 $155,405 $495 $19,156 $478 $593 $3,354,055 Portfolio Market Value $665,282 $917,396 $372,308 $126,736 $168,059 $84,511 $6,713 $11,514 $3,744 $23,724 $0 $56,144 $0 $238,759 $0 $14,387 $662 $1,104 $2,691,042 Vintage Indifference Amount $245,125 $521,872 $59,930 $692 ($7,576) ($35,346) ($3,041) ($5,393) ($1,969) ($14,382) $482 ($19,091) $496 ($83,354) $495 $4,769 ($184) ($512) $663,013 Cumulative Indifference Amount $245,125 $766,996 $826,927 $827,619 $820,042 $784,696 $781,655 $776,262 $774,293 $759,911 $760,393 $741,302 $741,798 $658,444 $658,940 $663,709 $663,525 $663,013 9 value of the RECs to departed load and ensure indifference according to California law. 1 That ‘purchase’ is priced at the RPS MPB. 2
The issue being considered in this Track 2 arises from a unique characteristic of 3 RECs, which makes them different from other generation resource attributes (e.g., energy 4 and RA): RECs can be banked and stored by the IOUs for later use.11 This leads to the 5 question of how and when value should be provided to unbundled customers for these 6 RECs for which they pay but that the IOU stores for later use. 7 Prior to 2019, bundled customers paid for the cost of IOU generation resources 8 9 through generation rates, including RPS-eligible resources that generated RECs whether they were used for compliance in the year generated or banked for later use. Departed load 10 customers also paid for a share of these IOU generation resources on a vintaged basis 11 through PCIA rates. Each year, the IOUs would forecast the quantity of RECs from PCIA-12 eligible resources that would not be sold to third parties, and these RECs were valued in 13 the PCIA for departed load customers using the RPS MPB in the year of generation. The 14 value was credited to the PCIA, providing a partial offset to the costs paid by then departed 15 load customers. In essence, all customers who were bundled customers at the time the REC 16 was generated pre-paid for RECs banked for later use. 17 In 2019, the Commission refined D.11-12-018’s requirement to compensate 18 departed load for the renewable value of RPS-eligible PCIA resources via more specific 19 PCIA accounting requirements.12 These new requirements recognized that customers do 20 21 not realize the value of the IOUs’ REC banks until those RECs are used for bundled 11 See Cal. Pub. Util. Code § 399.13(a)(5)(B).
12 See generally D.19-10-001.
10 customer RPS compliance (or sold):13 “Any of the offered [RPS resource] quantity that is 1 not sold will be considered as Actual Unsold RPS and should not be assigned credit in 2 PABA until the value of the RPS product, if any, is known. If previously unsold RPS is 3 sold in a future year, it should be valued at the actual transacted price. If previously unsold 4 RPS is used by the IOU for compliance in a future year, it should be valued at the applicable 5 future year’s RPS Adder [MPB]. If Unsold RPS is never used, it should not be assigned 6 credit.”14 7
The Commission required the IOUs to treat banked RECs as Retained RPS when 8 those banked RECs are later used for compliance for bundled customers. This is reflected 9 in the IOU accounting as charging bundled customers for the cost of those RECs at the 10 then-current market value (i.e., the RPS MPB) and including an offsetting credit in the 11 PCIA. This credit to the PCIA for the value of RECs compensates departed load customers 12 for the value of their share of the RECs. As discussed in more detail below, these 13 refinements did not upset the original accounting for RECs generated before 2019. They 14 merely created a more granular accounting for such RECs, which, until recent years, had 15 not been used for bundled customers’ RPS compliance. 16 17 III. LATER DEPARTING CUSTOMERS CURRENTLY RECEIVE NO BENEFIT FOR 18 PRE-2019 BANKED RECS NOW BEING USED BY THE IOUS FOR CURRENT 19 BUNDLED CUSTOMER RPS COMPLIANCE
Four categories of customers exist in the context of Pre-2019 Banked RECs (i.e., 20 RECs generated by PCIA resources from 2011 through 2018 that were not needed for RPS 21 compliance and were stored in the ‘bank’ for later use.) At the time the Pre-2019 Banked 22 RECs were generated, there were two categories of customers. First, there were customers 23 13 Id., at 35.
14 Id., at 30.
11 1 receiving bundled service at the time the RECs were generated (Then-Bundled Customers). Second, there were customers that had already departed bundled service 2 (Previously Departed Customers). Figure 4, below, demonstrates these two categories of 3 customers: 4
5 Figure 4: Customer Groups When Pre-2019 Banked RECs Were Generated 6 Both Then-Bundled Customers and Previously Departed Customers paid for their 7 share of the RECs when they were generated. Because Previously Departed Customers had 8 9 paid the cost of their share of the RECs, but those RECs were retained by the IOU for bundled customers, Previously Departed Customers received a credit against their PCIA 10 obligations for the value of their share of the RECs. That credit was calculated using the 11 value of the RPS MPB at the time. These Previously Departed Customers therefore were 12 fairly compensated for their share of the Pre-2019 Banked RECs and no longer have any 13 claim to further value or allocation for those RECs. 14 The benefits Then-Bundled Customers received at that time depended on whether 15 the RECs were needed for compliance. If the RECs were counted at that time toward Then-16 Bundled Customer compliance, Then-Bundled Customers benefited immediately. 17 However, if the RECs were not needed at that time for bundled customer compliance, the 18 IOUs placed the RECs in the ‘bank’ for later use. Then-Bundled Customers received no 19 12 credit at the RPS MPB at that time for the share of the RECs for which they paid.15 1 2 Therefore, the only customers that received a credit at the RPS MPB for the RECs for which they paid were Previously Departed Customers. The only ‘value’ provided to Then-3 Bundled Customers for banked RECs was the deferred ability to use the RECs at some 4 future point. 5
Figure 5 below provides a graphic illustration of: (1) the immediate benefit to 6 Previously Departed Customers when RECs were generated; and (2) the deferred benefit 7 to Then-Bundled Customers when RECs were banked. 8
9 Figure 5: REC Benefit Conveyed to Customer Groups When RECs Were Generated 10 and Banked
11 In fact, Then-Bundled Customers paid for the value conveyed to Previously 12 Departed Customers for the RECs retained by the IOU by virtue of the PCIA framework, 13 as shown in Figure 6. 14
15 See PG&E Response to CalCCA DR 3.04 and 3.05. All discovery responses referred to herein are included in Attachment C.
13 1 Figure 6: REC Benefits Were Conveyed to Previously Departed Customers 2 After the Pre-2019 Banked RECs were banked, some of the Then-Bundled Customers 3 departed bundled service to be served by CCAs or ESPs (i.e., Later Departing Customers). 4 This split the category of Then-Bundled Customers in two, creating two further categories 5 of customers: Current Bundled Customers and Later Departing Customers. Current 6 Bundled Customers are those customers that continue to take bundled service after the 7 Later Departing Customers leave bundled service.16 8
Later Departing Customers paid for the Pre-2019 Banked RECs when they were 9 Then-Bundled Customers, but upon their departure from bundled service were not paid for 10 those RECs which remained in the bank, ready for future use. These Later Departing 11 Customers number in the millions, constituting between 29 percent and 66 percent of 12 16 There is arguably a fifth category of customers within the pool of “Current Bundled Customers” who were Later Departing Customers that then returned to bundled service. They are a very small portion of customers that I do not address in this opening testimony. 14 customers in each IOU’s service territory, depending on the IOU and the REC generation 1 year.17 2
Recently, the Joint IOUs began to use Pre-2019 Banked RECs to meet their RPS 3 compliance requirements for Current Bundled Customers, thus avoiding the purchase of 4 additional RECs at current market prices. In doing so, Current Bundled Customers receive 5 the deferred value of the Pre-2019 Banked RECs. However, Later Departing Customers 6 still have not received any value for the Pre-2019 Banked RECs for which they also paid 7 when they were Then-Bundled Customers – neither the ‘avoided cost’ value Current 8 Bundled Customers enjoy nor the ability to use those banked RECs. These Later Departing 9 Customers will not otherwise receive any benefit when the IOUs use the RECs for Current 10 Bundled Customer compliance needs. Absent a credit to the PCIA or an allocation of these 11 Pre-2019 Banked RECs, Later Departing Customers will not receive a share of the value 12 of those RECs despite paying for them, as illustrated in Figure 7 below. 13 14 Figure 7: Customer Groups When Pre-2019 Banked RECs Are Later Used for 15 Compliance
16 17
See SCE response to CalCCA 3.14, PG&E response to CalCCA 3.14, and SDG&E response to CalCCA 3.14.
15 The following is a hypothetical example that will be utilized throughout my 1 testimony to illustrate how each category of customers discussed above bears the costs and 2 receives the values of Pre-2019 Banked RECs: 3
• Assume in 2015: 4
o 90 percent of customers were bundled; 5
o 10 percent of customers were unbundled; 6
o The IOU generated and banked 100,000 MWh of RECs; 7
o The RPS MPB was $50 per MWh. 8
• Assume in 2026: 9
o An additional 60 percent of the IOU’s customers have departed 10 bundled service (i.e., 60 percent of Then-Bundled Customers have 11 departed and become Later Departing Customers, with the 12 remaining 40 percent constituting Current Bundled Customers); 13 o The RPS MPB is $60 per MWh. 14
In 2015, the value of the banked RECs was credited to the PCIA and Previously 15 Departed Customers would have received 10 percent of that value ($500,000 in this 16 example18) through PCIA rates in exchange for the IOU retaining the RECs. Then-Bundled 17 Customers (which includes Later Departing Customers that will depart IOU service after 18 2015) paid for all of the RECs to be generated and banked but received no benefit in 2015 19 other than the deferred ability to use the banked RECs in the future. 20 In 2026, when an IOU uses the banked RECs for Current Bundled Customer RPS 21 compliance, it does not provide any credit to the Later Departing Customers, even though 22 they paid the same total amount as those Current Bundled Customers back in 2015 (when 23 both categories were Then-Bundled Customers). At the same time, the IOU can use the 24 18 100,000 MWh * $50 / MWh * 10% = $500,000.
16 Pre-2019 Banked RECs for Current Bundled Customer RPS compliance, resulting in the 1 realization of the value previously deferred when the RECs were banked in 2015. Because 2 60 percent of the banked RECs were paid for by Later Departing Customers, they should 3 receive 60 percent of the value of the banked RECs ($3.6 million in this example19) for 4 their share of the RECs when they are used by Current Bundled Customers for RPS 5 compliance. Failure to convey this value to Later Departing Customers results in a cost 6 shift of $3.6 million in this example from bundled to unbundled customers. 7 While that may seem like a small cost shift, it is enormous when applied to all of 8 the IOUs’ Pre-2019 Banked RECs. For example, if all Pre-2019 Banked RECs that remain 9 in the IOUs’ REC banks are valued at the 2026 RPS MPB of $62.45/MWh, over $1.5 10 billion in costs would be shifted from bundled customers to departed customers. 11 12 IV. VALUATION OR ALLOCATION OF PRE-2019 BANKED RECS PREVENTS 13 COST SHIFTS AND MAINTAINS INDIFFERENCE AS REQUIRED BY 14 STATUTE
Valuation or allocation of Pre-2019 Banked RECs is necessary to ensure 15 indifference and prevent a cost shift from Current Bundled Customers to Later Departing 16 Customers. To ensure Later Departing Customers remain indifferent to an IOU’s use of 17 Pre-2019 Banked RECs for Current Bundled Customer RPS compliance, CalCCA provides 18 the following proposal to ensure either the value or allocation of attributes is provided in 19 accordance with section 366.2(g): 20
• Value: The Commission should provide value to Later Departing 21 Customers for their share of RECs used by the IOU at the current RPS MPB 22 through a vintage-specific credit to the PCIA. The credit should be recorded 23 to the PCIA vintage of the year in which the RECs were generated. 24 • Allocation: Alternatively, Pre-2019 Banked RECs should reduce the Net 25 RPS Procurement Need as measured on the Commission’s Renewable Net 26 19 100,000 MWh * $60 / MWh * 60% = $3,600,000.
17 Short Calculation template for the LSEs that serve customers who paid for 1 the Pre-2019 Banked RECs when they were generated and then later 2 departed bundled service. 3
These proposed frameworks ensure indifference for Later Departing Customers and are 4 described more fully below. 5
6 A. Later Departing Customers Should Receive Value at the Current RPS MPB, 7 Through a Vintage-Specific Credit to the PCIA
8 1. The Best Measure of Value Accruing to Current Bundled Customers 9 When the IOU Uses Pre-2019 Banked RECs for Compliance is the 10 Current RPS MPB
Current Bundled Customers are responsible for the cost of RPS compliance on their 11 behalf. Value is derived from RECs based on the ability to count those RECs toward an 12 LSE’s RPS compliance obligation; in using banked RECs for compliance, Current Bundled 13 Customers avoid the cost of procuring an equivalent amount of RECs at current market 14 prices. Because the IOUs can bank and store RECs, their value may change over time as 15 the IOUs’ need for RPS procurement changes. Southern California Edison Company 16 (SCE) and PG&E both confirmed that, regarding RPS, there is no difference between 17 RECs generated before or after 2019.20 SCE explained, “For RPS compliance purposes, 18 there is no difference between RECs banked prior to 2019 and those banked in 2019 or 19 later, provided they meet standard RPS eligibility. Both types of banked RECs may be used 20 to satisfy SCE’s RPS obligations.”21 21
The IOUs routinely use banked RECs as a tool to manage their RPS procurement 22 as approved in the annual RPS Procurement Plans filed with the Commission. For example, 23 in D.24-12-035 the Commission describes that according to San Diego Gas & Electric 24 20 See SCE Response to CalCCA 3.09. See also PG&E Response to CalCCA 3.09. 21 See SCE Response to CalCCA 3.09.
18 Company’s (SDG&E) Draft 2024 RPS Plan, “SDG&E plans to meet its compliance 1 requirements through 2034 by utilizing its bank and/or, after considering the benefits to 2 customers, holding a solicitation or entering into bilateral agreements to procure long-term 3 and/or short-term resources to meet its Procurement Quantity Requirements.”22 In D.24-4 12-035, the Commission approved SDG&E’s request for the ability to procure short-term 5 and long-term RECs as an alternative to utilizing its banked RECs for compliance.23 In 6 other words, if an IOU does not have banked RECs, or decides not to use them, it can seek 7 to procure RPS resources from the market as needed to meet a shortfall in its compliance 8 obligation.24 9
The RPS MPB is designed to reflect the price at which the RPS resource attributes 10 can be purchased and sold in the market. 25 Consequently, the RPS MPB is the best 11 available estimate of the current market value of a REC, and Current Bundled Customers 12 should be required to pay that price to Later Departing Customers when the Later Departing 13 Customers’ share of the Pre-2019 Banked RECs are used for compliance. Notably, the 14 Commission recognized this fact in D.19-10-001. When the Commission adopted PG&E’s 15 recommendation to assign a zero value to Unsold RPS in the year of generation, it 16 explained the following: 17
The IOUs’ RPS procurement planning takes place in formal proceedings 18 subject to stakeholder review. The Commission may authorize the IOUs to 19 retain RPS resources based on their procurement planning needs. If the 20 IOUs use RECs in the future based on their approved procurement plans, 21 the value in the year of generation may be different from the value at the 22 time of the future transaction. To value all RECs in the year of generation, 23 as TURN notes, could conflict with the Commission-approved procurement 24 22 D.24-12-035, at 42.
23 Id., at 43.
24 See SCE Response to CalCCA 3.07 (confirming that IOUs can either use RECs from the bank, produce more RECs, or purchase RECs on the market).
25 See D.18-10-019, at 73.
19 plans. It effectively shifts market risks and opportunities associated with 1 changing REC prices to bundled customers even though the resources 2 generating those RECs were not procured solely on their behalf.26 3 As a result, the Commission directed that excess RPS that is not sold should be considered 4 as Unsold RPS and not assigned credit in the PCIA until some point in the future when the 5 RECs are either sold or used for bundled customer compliance.27 The Commission also 6 directed that when previously Unsold RPS is used by the IOU for compliance in a future 7 year, it should be valued at the applicable future year’s RPS MPB. 28 8
It is important to reiterate that, unlike Previously Departed Customers that were 9 already departed when the Pre-2019 Banked RECs were generated, Later Departing 10 Customers – those that departed bundled service after the RECs were generated – have 11 never been compensated for the value of Pre-2019 Banked RECs that are being used by 12 the IOU for Current Bundled Customer RPS compliance. When Current Bundled 13 Customers benefit from using the Pre-2019 Banked RECs for compliance, they should be 14 required to purchase the share of Pre-2019 Banked RECs originally paid for by Later 15 Departing Customers, thus ensuring Later Departing Customers receive a share of the 16 benefit. 17
Returning to the 2015 example to demonstrate these points, Then-Bundled 18 Customers paid for the RECs the IOU banked in 2015 but did not benefit from those RECs 19 20 at that time, i.e., the value of the RECs was deferred to a future date. Sixty percent of those banked RECs were paid for by Later Departing Customers, e.g., customers that departed 21 bundled service after 2015 but prior to 2026. In other words, if 100,000 MWh of 2015 22 26 D.19-10-001, at 35.
27 Id., at 30.
28 Id.
20 banked RECs are used for compliance on behalf of Current Bundled Customers in 2026, 1 60 percent of those banked RECs were originally paid for by Later Departing Customers. 2 3 These Later Departing Customers were part of the group of Then-Bundled Customers that 4 originally paid for the 100,000 MWh of RECs in 2015 and are waiting to receive their share of the benefits. Absent a credit to the PCIA, Current Bundled Customers will retain all the 5 benefits from these RECs even though Later Departing Customers paid for a share of the 6 RECs when they were generated. Figure 8 illustrates how such a credit should be conveyed 7 to Later Departing Customers. 8
9 Figure 8: REC Benefit Conveyed to Customer Groups When Pre-2019 Banked RECs Are Used for Compliance 10
11 12
13 2. Providing Vintage-Specific Value of Pre-2019 Banked RECs Ensures 14 Only the Right Customers are Credited
As noted above, the PCIA framework in place prior to 2019 ensured that Previously 15 16 Departed Customers who had already departed when the RECs were generated received 17 the benefit of the RECs through the RPS Value credit embedded in the PCIA. Those Previously Departed Customers were fully compensated in the year the RECs were 18 generated. Then-Bundled Customers receiving bundled service when the RECs were 19 21 generated, but who later departed (i.e., Later Departing Customers), paid for the RECs but 1 have not yet received the benefit of those RECs. Those customers, and not the Previously 2 Departed Customers, should benefit from the Pre-2019 Banked RECs when they are used 3 for bundled customer compliance. 4
Conveying the value of Pre-2019 Banked RECs through a vintage-specific credit 5 to the PCIA ensures only the Later Departing Customers (and not the Previously Departed 6 7 Customers) receive the value of their share of the RECs used by the IOU. In my 2015 REC example, the value of RECs banked in 2015 but used for compliance in 2026 would be 8 measured using the RPS MPB for 2026 and recorded as a credit to PCIA vintage 2015. 9 This method ensures that customers departing bundled service in 2015 or later receive 10 credit for the share of RECs they paid for when they were generated in 2015; the credit will 11 convey the same value received by Current Bundled customers, the avoided cost of 12 additional procurement measured by the RPS MPB. This method also ensures that 13 customers who had already departed bundled service (Previously Departed Customers) 14 prior to 2015, and who received credit for RECs through the PCIA in 2015, are not given 15 additional credit for the same RECs. 16
17 B. Allocation of Pre-2019 Banked RECs to Later Departing Customers is an 18 Alternative Valuation Method Consistent with Indifference Requirements Allocating the Pre-2019 Banked RECs to Later Departing Customers rather than 19 20 charging Current Bundled Customers for the REC value is an alternative method that is also consistent with the indifference requirements. As noted above, Public Utilities Code 21 section 366.2(g) entitles departed load customers to receive either the value of resource 22 attributes or an allocated share of the attributes. Specifically, using the same formulaic 23 approach to determine which PCIA vintages should receive a monetary credit, the 24 22 Commission could determine the share of each IOU’s Pre-2019 Banked RECs that was 1 paid for by Later Departing Customers in each PCIA vintage. That share of RECs could 2 then be inserted as a reduction to the relevant LSE’s Net RPS Procurement Need as 3 measured on the Commission’s Renewable Net Short Calculation template and an 4 offsetting increase to each IOU’s Net RPS Procurement Need.29 5 The idea is similar to the allocation of RA capacity from Cost Allocation 6 Mechanism (CAM) resources for the Commission’s RA compliance program. 7 Jurisdictional LSEs are required to demonstrate that they have procured enough RA 8 capacity to meet their RA requirement as prescribed by the Commission. Under the CAM 9 framework, when the IOU procures a resource on behalf of bundled and unbundled 10 11 customers (as directed by the Commission), each LSE in the IOU service territory is allocated a load ratio share of the RA provided by the resource. Rather than transfer 12 ownership of the allocated portion of RA, however, the Commission increases the IOU RA 13 requirement and reduces each LSE’s RA requirement by its allocated share of the CAM 14 resources. 15
16 A similar process could be followed for Pre-2019 Banked RECs. All retail sellers are required to submit Renewable Net Short calculations to the Commission each year with 17 their RPS Procurement Plans. The Commission’s Renewable Net Short calculation 18 19 template quantifies each LSE’s Gross RPS Procurement Quantity Requirement as a percentage of retail sales volume. Adjustments are then made to determine the LSE’s Net 20 RPS Procurement Need and Net RPS Position after considering available resources. To 21 22 treat Pre-2019 Banked RECs similar to CAM, the relevant IOU would retain ownership of 29 See, e.g., 2023 Renewable Net Short Template, https://webtraininga.cpuc.ca.gov/-/media/cpuc-website/industries-and-topics/documents/energy/rps/2023/renewable-net-short-template-2023.xlsx. 23 the banked RECs, but the Commission would adjust an individual LSE’s Net RPS 1 Procurement Need for its allocated share of the RECs (i.e., a reduction to non-IOU Net 2 RPS Procurement Need, offset by an increase to IOU Net RPS Procurement Need).30 3 Figure 9 and Figure 10 below show an excerpt from the Commission’s Renewable Net 4 Short template and how an adjustment for Pre-2019 Banked RECs could be made in the 5 template. 6
7 Figure 9: Current Renewable Net Short Template Excerpt 8 9 Figure 10: Renewable Net Short Template Excerpt With Proposed Pre-2019 Banked REC Adjustment 10
11 The quantity of Pre-2019 Banked RECs to be allocated each year would be 12 determined by an IOU’s need to use its share of the Pre-2019 Banked RECs for Current 13 Bundled Customers. For example, if Later Departing Customers paid for 60 percent of the 14 IOU’s 100,000 MWh of Pre-2019 Banked RECs, when the IOU needs to use 40,000 MWh 15 of RECs for Current Bundled Customers, it would also allocate 60,000 MWh of additional 16 30 The offsetting increase would not increase the IOU’s RPS obligation; it would simply reflect the idea that the RECs remain on the IOU’s books but should not be counted towards their RPS procurement.

Variable Calculation Item 2025 Forecast 2026 Forecast 2027 Forecast 2025-2027 Forecast Year 3 4 5 CP 5
Annual RPS Requirement
A Total Retail Sales (MWh) -
B RPS Procurement Quantity Requirement (%) 46.7% 49.3% 52.0% NA C A*B Gross RPS Procurement Quantity Requirement (MWh) - - - - D Voluntary Margin of Over-procurement (MWh) -
E C+D Net RPS Procurement Need (MWh) - - - -
RPS-Eligible Procurement
Variable Calculation Item 2025 Forecast 2026 Forecast 2027 Forecast 2025-2027 Forecast Year 3 4 5 CP 5
Annual RPS Requirement
A Total Retail Sales (MWh) -
B RPS Procurement Quantity Requirement (%) 46.7% 49.3% 52.0% NA C A*B Gross RPS Procurement Quantity Requirement (MWh) - - - - D Voluntary Margin of Over-procurement (MWh) -
E Pre-2019 Banked REC Allocation (MWh)
F C+D + E Net RPS Procurement Need (MWh) - - - -
RPS-Eligible Procurement
24 RECs among the LSEs serving Later Departing Customers and would no longer be able to 1 use those RECs for its own compliance. Each LSE’s share of the Pre-2019 Banked RECs 2 allocated to Later Departing Customers would be based on the relevant vintage’s load share 3 of all non-exempt LSEs in the IOU service territory. The IOUs’ need to use Pre-2019 4 Banked RECs, and the quantities allocated to Current Bundled Customers and Later 5 Departing Customer LSEs, would be determined in the IOUs’ annual ERRA proceedings. 6 7 V. CALCCA’S PROPOSAL IS CONSISTENT WITH APPLICABLE LAW AND 8 COMMISSION PRECEDENT (SCOPING RULING ISSUE 1)
The Scoping Ruling asks two overall questions. First, is the proposal to value Pre-9 10 2019 Banked RECs at a value other than zero consistent with applicable law and Commission precedent? Next, if the answer to Issue 1 is determined to be yes, should the 11 12 Commission direct IOUs to apply a value other than zero dollars when Pre-2019 Banked RECs are valued for ratemaking in the 2026 and later ERRA Forecast Proceedings? If so, 13 how should such value be determined and allocated while adhering to indifference 14 principles for all customers.31 15
As demonstrated in detail throughout my testimony above, CalCCA’s proposal to 16 value Pre-2019 Banked RECs at the current RPS MPB is consistent with California statutes 17 applying to bundled and unbundled customers, and is consistent with Commission 18 decisions and precedent related to the PCIA. To ease the Commission’s review of 19 CalCCA’s proposal, the remainder of my testimony takes each sub-question in the Scoping 20 Ruling and provides a specific response to it. 21
31 Scoping Ruling, at 3-4.
25 1 A. Statutory Indifference Principles, the Mandate to Prevent Cost Shifts, and 2 Prior Commission Decisions All Require that Later Departing Customers 3 Receive Their Share of Value for Bundled Customer Use of Pre-2019 Banked 4 RECs (Scoping Ruling Issue 1.a.)
As demonstrated in my testimony, ensuring Later Departing Customers receive 5 value for Pre-2019 Banked RECs prevents a cost shift from Current Bundled Customers to 6 Later Departing Customers that would otherwise occur if such value is not provided when 7 the IOUs use the Pre-2019 Banked RECs for Current Bundled Customer RPS compliance. 8 Valuing the Pre-2019 Banked RECs as proposed by CalCCA ensures Later Departing 9 Customers are indifferent to the IOUs using RECs originally paid for by those customers 10 for Current Bundled Customers’ RPS compliance obligations. On the other hand, giving 11 Later Departing Customers no value for Pre-2019 Banked RECs used for Current Bundled 12 13 Customers’ RPS compliance obligations would not leave Later Departing Customers indifferent. As a policy matter, customers would not be indifferent if they were forced to 14 15 leave behind these valuable banked RECs and get nothing in return when departing IOU bundled service. 16
Not only would Later Departing Customers remain indifferent under CalCCA’s 17 proposal, Current Bundled Customers also would remain indifferent. As shown in my 18 examples above, and in the numerical example in Attachment A to my testimony, Then-19 Bundled Customers paid for the RECs generated and banked in 2015. Thus, a portion of 20 the Pre-2019 Banked RECs was paid for by Current Bundled Customers, and a portion of 21 those RECs was paid for by Later Departing Customers (customers who departed bundled 22 service since 2015). Valuing the 2015 Banked RECs at the RPS MPB when they are finally 23 used for Current Bundled Customers in 2026 does not result in Current Bundled Customers 24 paying twice for the same RECs. Rather, it results in Current Bundled Customers 25 26 purchasing the portion of the 2015 RECs that were paid for by Later Departing Customers 1 (who were Then-Bundled Customers in 2015 and paid for a share of the REC bank). That 2 is in line with the costs Current Bundled Customers avoid by using RECs paid for by 3 another group of customers. It is also in line with the Commission’s existing indifference 4 framework, and not contrary to it. 5
6 B. No Policy Considerations or Other Downstream Consequences Negate the 7 Commission’s Statutory Obligation to Maintain Indifference and Ensure 8 Charges are Just and Reasonable (Scoping Memo Issue 1.b.) The Scoping Ruling asks whether there are potential “downstream” consequences 9 or other policy considerations stemming from the proposal to value Pre-2019 Banked RECs 10 at a value other than zero dollars that would render the proposal unreasonable pursuant to 11 Public Utilities Code section 451, or any other statute or Commission decision. My 12 understanding of Public Utilities Code section 451 is that it requires rates to be just and 13 reasonable. As stated throughout my testimony, because Later Departing Customers paid 14 for their share of Pre-2019 Banked RECs, it is just, reasonable, and required by the 15 indifference principle for Current Bundled Customers to compensate Later Departing 16 Customers when those RECs are used by the IOU to meet RPS compliance needs of 17 Current Bundled Customers. 18
19 C. Differences Between Pre-2019 and Post-2018 Banked RECs Do Not Impact the 20 Reasonableness of Assigning, or the Commission’s Obligation to Assign, Value 21 to the Pre-2019 RECs (Scoping Issue 1.c.)
Distinguishing RECs as either ‘Pre-2019’ or ‘Post-2018’ is a consequence of D.19-22 10-001 that created the PCIA category of Unsold RPS effective January 1, 2019. In 23 response to discovery, PG&E and SCE both confirmed that for RPS compliance there is 24 27 no distinction between RECs generated before and after 2019.32 The need to determine the 1 value of the Pre-2019 Banked RECs arises because the IOUs plan to use the RECs to meet 2 RPS compliance needs for Current Bundled Customers. If an IOU did not have banked 3 4 RECs or decided not to use them, it would procure RPS resources from the market as needed to meet a shortfall in its compliance obligation. Consequently, the current RPS 5 MPB best measures the cost avoided when bundled customers rely on banked RECs rather 6 than procuring additional RECs in the market, and Current Bundled Customers should be 7 required to pay that price to Later Departing Customers when the Pre-2019 Banked RECs 8 9 are used for compliance.
10 D. Banking of RECs Under the Pre-2019 Rate Methodology Does Not Result in 11 Forfeiture of Value by Later Departing Customers (Scoping Issue 1.d.) As a matter of policy, customer groups that pay the cost of IOU resources should 12 receive the benefit of those resources. The Commission has enshrined this policy in a PCIA 13 framework that requires departing load to continue to pay for IOU generation resources 14 even after their departure from bundled IOU service. In fact, many of the resources that 15 generated the Pre-2019 Banked RECs are still included in the PCIA rates paid by 16 unbundled customers today. My testimony demonstrates that a portion of the Pre-2019 17 Banked RECs was paid for by Later Departing Customers. By banking those RECs for 18 future compliance needs, the IOUs deferred conveying the value of the RECs to customers. 19 When the benefits of the banked RECs are realized, the benefits should follow the customer 20 groups that paid for the RECs to be generated. 21
I am not aware of any Commission directive that would preclude customers from 22 receiving the value of the RECs for which they previously paid. I am also not aware of any 23 32 See PG&E Response to CalCCA 3.09; see also SCE Response to CalCCA 3.09. 28 Commission directive that would require Later Departing Customers to forfeit the value 1 they are owed to ensure indifference. 2
3 VI. THE IOUS SHOULD BE DIRECTED TO APPLY CALCCA’S VALUATION OR 4 ALLOCATION PROPOSAL (SCOPING ISSUE 2)
5 A. Using Current RPS MPB Data is Reasonable and Required to Maintain 6 Indifference (Scoping Issue 2.a.)
Current Bundled Customers should be responsible for the cost of RPS compliance 7 on their behalf. The value of Pre-2019 Banked RECs is based on the IOU’s ability to count 8 those RECs toward its RPS compliance obligation for Current Bundled Customers rather 9 than procure additional RECs at current market prices. Because RECs can be banked and 10 11 stored by the IOUs, their value may change over time as the IOUs’ need for RPS procurement changes. The RPS MPB is designed to reflect the price at which the resource 12 attributes can be purchased and sold in the market. Consequently, the current RPS MPB is 13 the best available estimate of the cost Current Bundled Customers avoid by not having to 14 procure at the current market value of a REC, and Current Bundled Customers should be 15 required to pay that price to Later Departing Customers when the Pre-2019 Banked RECs 16 are used for compliance. 17
18 B. The Only Impact the Pre-2019 PCIA Methodology Has on How the 19 Commission Should Value or Allocate Pre-2019 Banked RECS to Customers 20 Now is That Such Valuation Must be Vintage-Specific to Avoid Double-21 Payment to Previously Departed Customers (Scoping Issue 2.b.) As described in detail throughout my testimony, prior to 2019 Then-Bundled 22 23 Customers paid for the RECs that were generated and banked by the IOUs, but only Current Bundled Customers will receive the benefit of those RECs when the IOU uses them for 24 compliance. Following CalCCA’s recommended approach to valuing Pre-2019 Banked 25 RECs and crediting the PCIA resolves the unfair treatment of Later Departing Customers 26 29 and ensures three outcomes. First—and most importantly—customers that originally paid 1 for the RECs but later departed bundled service will finally receive the benefit of the RECs 2 for which they originally paid. Second, Current Bundled Customers only purchase Later 3 Departing Customers’ share of RECs used by the IOU, meaning they never pay twice for 4 the same RECs. Third, Previously Departed Customers are not credited twice for the same 5 RECs. 6
While ensuring these fair outcomes, CalCCA’s proposal does not upset the prior 7 8 PCIA methodology and ratemaking treatment. The IOUs suggest otherwise in their responses to CalCCA’s discovery. 33 However, the Commission’s prior ratemaking 9 treatment addressed the question of indifference between Previously Departed Customers 10 and Then-Bundled Customers. That is different than the question before the Commission 11 in Track 2, which asks how to ensure indifference between the two sub-groups of Then-12 Bundled Customers: Later Departing Customers and Current Bundled Customers. In other 13 words, the IOUs’ arguments focus on the wrong groups of customers. 14 Prior to the accounting changes adopted in D.19-10-001, the general framework of 15 the PCIA described above was already in place. Decision 19-10-001 did not change the 16 17 fact that: (1) Then-Bundled Customers paid for the RECs at the time they were generated; (2) only Current Bundled Customers benefit when the IOU uses the RECs to meet current 18 compliance obligations; and (3) Later Departing Customers will not receive any value for 19 20 the RECs unless the Commission adopts CalCCA’s recommendations in this proceeding.

CalCCA’s proposal does not constitute reopening past cases or partially refunding 21 22 generation rates paid for by Later Departing Customers in the years Pre-2019 Banked RECs 33 See SCE Response to CalCCA 3.09.
30 were generated. Those rates were paid and they remain paid. But when a resource is paid 1 for through those generation rates and the attributes are placed on the shelf for later use, 2 customers who originally paid those rates should receive the value of the attributes when 3 they are eventually taken off the shelf and used. 4
CalCCA also does not propose truing-up rates set in prior years. Using my prior 5 6 example, a true-up would update the value of the RECs when they were generated in 2015 ($50/MWh) with the value of the RECs when they were used for compliance in 2026 7 ($60/MWh). That approach is similar to the approach adopted in SDG&E’s ERRA Forecast 8 cases, which suffers from significant short-comings.34 As PG&E explained in response to 9 discovery,35 only Previously Departed Customers received a credit at the RPS MPB when 10 the RECs were originally generated so it would not make sense to true-up the prior MPB 11 when addressing a cost shift tied to Then-Bundled Customers that later departed. 12 At most, the prior ratemaking treatment necessitates avoiding double-compensating 13 Previously Departed Customers, i.e., it necessitates the vintage-specific credit or allocation 14 15 approach CalCCA proposes here (rather than spreading credits or allocations to all vintages) when the Pre-2019 Banked RECs are used for compliance. If the Pre-2019 16 Banked REC credit is applied to the PCIA vintage corresponding with the year the RECs 17 were generated, the value of the RECs will be allocated to customers that departed service 18 in the year the REC was generated and later. Under this approach, Later Departing 19 Customers will receive their proportionate share of the credit as will Current Bundled 20 Customers, while Previously Departed Customers that were already credited in the PCIA 21 34 See Attachment A. Another shortcoming to the SDG&E ERRA Forecast approach is that the 2015 RPS MPB was calculated using a different methodology than the 2026 RPS MPB, meaning the Commission would be truing up an apple with an orange.
35 See PG&E response to CalCCA 3.05.
31 when the RECs were generated will not receive any of the new credit. Beyond that need to 1 2 avoid a double payment, the prior ratemaking treatment should have no impact on ensuring 3 all customers remain indifferent by crediting all RECs used for Current Bundled Customer compliance at the RPS MPB. 4
5 C. CalCCA’s Valuation Proposal Would Not Require Changes to PCIA 6 Ratemaking, Balancing Accounts, and/or Tariffs (Scoping Issue 2.c.) Following CalCCA’s proposed approach fits within the existing PCIA framework 7 and would not require modification to balancing accounts or tariffs. In fact, this approach 8 is not new. PG&E has used the approach endorsed by CalCCA for valuing banked RECs 9 in its ERRA Forecasts since 2023, including its 2023 ERRA Forecast (A.24-05-009),36 10 2024 ERRA Forecast (A.23-05-012),37 and 2025 ERRA Forecast (A.24-05-009).38 In each 11 of these proceedings, PG&E’s forecasted RPS generation was less than the annual RPS 12 compliance obligation. In each proceeding, PG&E used banked RECs to make up the 13 difference and applied a credit to the PCIA vintage matching the year the banked RECs 14 were generated. 15
PG&E explained the logic of this approach in its 2023 ERRA Forecast testimony. 16 PG&E explained that in previous years its ERRA revenue requirement was calculated to 17 recover the full RPS market value associated with its annual RPS generation volumes, even 18 if the volume exceeded the annual RPS compliance requirement.39 As a result, all of 19 PG&E’s banked RECs through that point were paid for by customers who received bundled 20 36 A.22-05-029, PG&E 2023 ERRA Forecast Prepared Testimony, at 11-13 through 11-21. 37 A.23-05-012, PG&E 2024 ERRA Forecast Prepared Testimony, at 9-17 through 9-24. 38 A.24-05-009, PG&E 2025 ERRA Forecast Fall Update Testimony, at 9. See also A.25-05-011, Confidential Tr. Vol. 1, at 110:2-24.
39 A.22-05-029, PG&E 2023 ERRA Forecast Prepared Testimony, at 11-14, lines 10-21. 32 service in the year the RECs were generated.40 PG&E explained, “It is precisely those 1 customers that earlier procured surplus RPS generation who will benefit from an 2 accounting adjustment for 2023 ratesetting.”41 Accordingly, PG&E credited the PCIA 3 vintages corresponding to the years in which the banked RECs were generated by applying 4 the then-current forecast RPS Adder to the quantity of banked RECs utilized. 5 6 7
This concludes my testimony. 8
40 Id., at 11-16, lines 28-33.
41 Id., at 11-17, lines 1-4.

ATTACHMENT A
Pre-2019 Banked RECs Numerical Example
Attachment A-1
The following tables demonstrate the payments and credits for my 2025 example made through the PCIA framework when the RECs were generated, and show the credit required when banked RECs are used for compliance. Table 1 below illustrates the overall PCIA framework with generic assumptions for resource costs and portfolio market value as it would have occurred in 2015 when the RECs were generated. As previously noted, the assumption for this example is that 10 percent of the IOU’s customers had departed bundled service prior to 2015.

Table 1: Example PCIA Mechanics in 2015

As Table 1 demonstrates, Then-Bundled Customers would pay $45.5 million in generation rates, comprising their share of resource costs ($50 million * 90% = $45 million) plus the market value ($500,000) of Previously Departed Customers’ share of RPS attributes that remained with the IOU for bundled customers. Previously Departed Customers would pay for their share of expected resource costs, reduced by a credit for the expected value of the Previously Departed Customers’ share of the RECs that remained with Then-Bundled Customers. Through this framework, Then-Bundled Customers paid for all RECs generated and banked in 2015, and there was a value credit only to Previously Departed Customers in exchange for their share of RECs that remained with Then-Bundled Customers. Then-Bundled Customers did not receive a similar Total ($000)
Then-Bundled Customers
Pay: Generation Resource Cost $50,000
Less: Departed Load Share of Above Market Cost ($4,500) Net Paid by Then-Bundled Customers $45,500
Previously Departed Customers
Pay: Departed Load Share of Generation Resource Cost $5,000 Less: RPS Value ($500)
Departed Load Share of Above Market Cost $4,500
Total Cost Responsibility $50,000
Attachment A-2
credit, meaning they received $0 for their own share of the RECs.42 That treatment is intuitive because the RECs were banked for later use: Then-Bundled customers eventually realize the value of the RECs for which they paid when they are used for compliance.

Continuing with my example from above, the IOU generated and banked 100,000 MWh of RECs in 2015, and the RPS MPB was $50 per MWh in 2015. Table 2 demonstrates how Then-Bundled Customers (still 90 percent of all customers at the time) would have been attributed a 90 percent share of RECs generated in 2015 (i.e., 90,000 MWh) because they paid their 90 percent share of the resource costs. Then-Bundled Customers also purchased the 10 percent share of RECs belonging to Previously Departed Customers (i.e., 10,000 MWh). After the payments and credits, Then-Bundled Customers paid for all 100,000 MWh of these illustrative RECs generated in 2015 but received no value for the RECs as they were placed in the bank and the benefits (value realization) were deferred.

Table 2: Allocation of RECs, RPS Payments and Credits in 2015 Continuing my example from above, the IOU needs to use the 100,000 MWh of RECs banked in 2015 to meet its compliance obligations for Current Bundled Customers in 2026. Since 2015, 60 percent of the IOU’s Then-Bundled Customers have departed bundled service (meaning that in 2026, only 40 percent of the Then-Bundled Customers still receive bundled service as Current Bundled Customers). Consequently, 60 percent of the 2015 banked RECs were paid for 42 See PG&E Response to CalCCA DR 3.04 and 3.05.

Bundled Load (2015) Departed Load (Pre-2015) Total
MWh 90,000 10,000 100,000
Price ($/MWh) $50.00
RPS Value Realized $500,000
Share of RECs (MWh)
Share of RECs Generated 90,000 10,000 100,000
Share of RECs Paid for at Market 10,000 - (10,000)
Total RECs Banked - 100,000 100,000
Attachment A-3
by Later Departing Customers. Current Bundled Customers should receive the benefits from their own share of the 2015 banked RECs (i.e., 40 percent * 100,000 = 40,000 MWh). Later Departing Customers (who departed after 2015) should receive value for their share of the banked RECs (i.e., 60 percent * 100,000 = 60,000 MWh). As shown in Table 3, assuming the 2026 RPS MPB is $60 per MWh, the 2015 banked RECs have a total value of $6 million and Later Departing Customers should be given a credit of $3.6 million in exchange for their share of the banked RECs.

Table 3: Allocation and Value of 2015 Banked RECs in 2026

When the 2015 banked RECs are counted toward the IOU’s compliance obligation, Current Bundled Customers will receive value from: (1) their own RECs (the 40,000 MWh); and (2) the avoided cost of buying additional RECs (60,000 MWh) in the market. Later Departing Customers should receive the same value as that avoided cost through credit to the PCIA to reflect the fact that RECs paid for by Later Departing Customers will be retained for Current Bundled Customers.

This provision of value will avoid shifting the cost of Current Bundled Customer RPS compliance to the Later Departed Customers.

If the Pre-2019 Banked REC credit is applied to the PCIA vintage corresponding with the year the RECs were generated, the value of the RECs will be allocated to customers that departed service in the year the REC was generated and later (i.e., Later Departing Customers). As described Bundled Load (2015) Departed Load (Pre-2015)
Banked REC Share (MWh)
Bundled Customers (2026) - 40,000
Later Departing Customers (2015-2025) - 60,000
Total Banked RECs (MWh) - 100,000
RPS MPB $60.00
Banked REC Value
Current Bundled Share (40%) $2,400 -
Later Departing Share (60%) $3,600 -
Total Banked RECs Value ($000) $6,000 -
Attachment A-4
earlier, customers in a given vintage pay all costs and receive all credits recorded to their vintage and earlier. So if a credit for the value of banked RECs is recorded to PCIA vintage 2015, customers assigned to PCIA vintage 2015 and later (i.e., Later Departing Customers in this example) will receive a share of the credit, while customers assigned to PCIA vintage 2014 and earlier (i.e., Previously Departed Customers in this example) will not receive any of the new credit.

As shown in Table 4, crediting the total value of banked RECs (i.e., $6 million) to vintage 2015 results in $3.6 million returning through PCIA rates to Later Departing Customers that departed bundled service between 2015 and 2025.43 The remaining $2.4 million stays with vintage 2026, which is Current Bundled Customers.

Table 4: PCIA Rate Impact of Vintage-Specific Credit

In this example, customers in vintages 2014 and earlier would not receive any benefit since Previously Departed Customers received their benefit in 2015 when the RECs were generated.

Adopting the vintage-specific credit mechanism demonstrated above is required to avoid unintended outcomes of other allocation methods like the one adopted in SDG&E’s 2025 and 2026 43 Note that this example assumes customers depart at a linear rate although the model would work for any rate of customer departure.

PCIA Template (Simplified)
Vintage UOG Legacy 2009 # # # # # 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 Total Cost of Portfolio ($000)
Market Value of Portfolio ($000)
RPS Value ($6,000) ($6,000)
Indifference Amount $0 $0 # # # # # ($6,000) $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 ($6,000) IOU System Sales by Vintage (GWh) 70,000 70,000 63,000 59,564 56,127 52,691 49,255 45,818 42,382 38,945 35,509 32,073 28,636 25,200 PCIA Rates - - (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) Value to Bundled Load ($2,400) ($2,400)
Value to Later Departed Load $0 $0 # # # # # ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($3,600) Total Banked REC Value ($000) $0 $0 # # # # # ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($327) ($2,400) ($6,000) Attachment A-5
ERRA Forecast proceedings.44 The method adopted in those cases has two distinct components:
(1) the price used to value banked RECs is determined by the difference between the current RPS MPB and the RPS MPB when the REC was originally generated; and (2) the value of banked RECs is applied as a credit to all PCIA vintages.

Using the incremental change in the RPS MPB as the price of banked RECs assumes the value of RECs needs to be trued up to a previous inaccurate value. Using the hypothetical example, the SDG&E method would credit all departing load for the $10 per MWh increase in the RPS MPB (from $50 in 2015 to $60 in 2026). However, there is no previous inaccurate value that needs to be trued up in this case—Later Departing Customers never received the value of the REC at the prior $50 per MWh RPS MPB. That is why CalCCA’s proposal is not intended to be a true-up like the mechanism implemented in D.19-10-001 (i.e., one that adjusts a forecasted price or quantity of RECs to an updated, or actual, price or quantity). Rather, valuing Pre-2019 Banked RECs using the current RPS MPB and applying a vintage-specific credit for the full value will, for the first time, convey the current market value of the RECs to Later Departing Customers who have not yet received any value of the Pre-2019 Banked RECs they paid for.

Applying the credit to all PCIA vintages also results in the incorrect customer groups receiving the value of the Pre-2019 Banked RECs. If the credit is applied to all vintages, Previously Departed Customers who departed before the RECs were generated and who already received value for their share of the RECs retained by the IOU will receive additional value. In the 2015 example, some of the banked REC value will flow to customers in vintages 2014 and earlier. Those Previously Departed Customers already received a credit for RECs according to the proportion of their generation responsibility in 2015 and the market value at the time. To convey credit again 44 See D.24-12-040, at 28-29; see also D.25-12-008, at 29-30.

Attachment A-6
would double-compensate some customers. That would not be just or reasonable, and it would not leave all customers indifferent.

ATTACHMENT B
Curriculum Vitae of Brian Dickman

BRIAN DICKMAN
Chief Financial Officer
Economics | Strategy | Stakeholders | Sustainability
www.newgenstrategies.net

CONTACT
225 Union Boulevard, Suite 450
Lakewood, CO 80228
[email protected]
www.newgenstrategies.net

EDUCATION
Master of Business Administration,
Finance Emphasis, University of Utah
Bachelor of Science, Accounting, Utah
State University

KEY EXPERTISE
Cost of Service and Rates
Financial Analysis and Modeling
Power Charge Indifference Amount
Regulatory Strategy
Revenue Requirement
Mr. Brian Dickman is a partner in NewGen’s energy practice with over 20 years of utility industry experience. Mr. Dickman’s career includes over a decade working for PacifiCorp, a vertically integrated investor-owned utility, including senior-level positions in regulatory, financial, and commercial roles. He began consulting in 2017, assisting a wide array of clients across the United States and internationally, including utilities, large consumers, and private investment firms. Mr. Dickman has extensive experience preparing and evaluating utility revenue requirements and cost allocation studies, developing utility-avoided costs, and analyzing the impact of new initiatives and transactions on a utility and its customers. In addition to his extensive technical experience, Mr. Dickman understands the regulatory governance process, and he has personally testified as an expert witness before state public utility commissions in California, Idaho, Indiana, Oregon, Utah, Washington, and Wyoming.

Mr. Dickman advises numerous Community Choice Aggregator (CCA) clients in California, focusing on regulatory and rate issues such as the state-mandated exit fee known as the Power Charge Indifference Adjustment (PCIA). He also represents California CCAs as a member of the Cost Allocation Mechanism Procurement Review Groups for PG&E and Southern California Edison, which the California Public Utility Commission established to provide an independent review of the centralized procurement of local generation capacity requirements.

RELEVANT EXPERIENCE
Electric Cost of Service, Rate Design, and Regulatory
Analysis
Mr. Dickman leads projects developing utility revenue requirements, preparing cost of service and rate design studies, and performing financial and regulatory analyses for electric utilities. Mr. Dickman previously held leadership positions at a multi-billion-dollar utility. He interfaced with state regulatory agencies in support of revenue requirements, cost recovery mechanisms, avoided costs, valuations of potential asset acquisitions and other commercial opportunities, and financial impacts of utility initiatives. Mr. Dickman now works with clients and stakeholders to prepare pro forma financial models to determine revenue sufficiency, evaluate the cost of service studies and rate design proposals, and support such proposals before local and state governing bodies. Mr. Dickman’s experience also includes evaluating the financial and rate impact of proposed mergers and acquisitions, acquisition and divestiture of utility assets, negotiated retail service contracts, changing business models, and stranded costs due to exiting load. A sample of Mr. Dickman’s utility clients includes the following:
 Abu Dhabi Distribution
Company, UAE
 Central Coast Community
Energy, CA
 City and County of San
Francisco, CA
 Clean Power Alliance, CA
 Duke Energy, NC
 East Bay Community Energy, CA
 Hydro One, Ontario, Canada
 Liberty Utilities, CA

BRIAN DICKMAN
Partner
Thoughtful Decision Making for Uncertain Times 2
Electric Cost of Service, Rate Design, and Regulatory Analysis (cont.)  Lubbock Power and Light, TX
 Minnesota Power, MN
 New York Power Authority, NY
 Portland General Electric, OR
 San Diego Community
Power, CA
 San Jose Clean Energy, CA
 Silicon Valley Clean Energy
Authority, CA
 Vermont Gas Systems, VT
Non-Utility Clients
A sample of Mr. Dickman’s non-utility clients includes the following:
 Blackstone Group, NY
 California Community Choice
Association, CA
 Facebook, CA
 Hemlock Semiconductor, MI
 Newmont Mining, NV
 SABIC Innovative Plastics, IN
 Tri-County Metropolitan
Transportation District, OR
 Vistra Energy, TX
Expert Witness and Litigation Support
Mr. Dickman provides comprehensive expert witness testimony related to utility revenue requirements, cost of service, rate design, and other ratemaking issues before state and local regulatory bodies. He has provided litigation support in wholesale and retail jurisdictions, including California, Idaho, Indiana, Oregon, Washington, Wyoming, Utah, the Federal Energy Regulatory Commission, and Ontario Energy Board. Mr. Dickman offers expert witness testimony and litigation support in the following areas.

Revenue Requirement | Cost Allocation | Rate Design
Mr. Dickman prepared revenue requirements, inter-jurisdictional cost allocation, coincident peak allocation studies, and supporting testimony for PacifiCorp over many years. He now provides litigation support and expert testimony for clients wishing to review utility filings on revenue requirements, cost allocation, and rate design, including program-specific rate tariffs.

Power Supply Costs | Stranded Costs | Rate Adjustment Mechanisms Mr. Dickman has prepared and evaluated variable power supply cost forecasts, power supply cost balancing accounts and other rate mechanisms, stranded costs, and exit fees for departing loads. Since 2019, Mr. Dickman has actively participated in PCIA matters in California on behalf of CCA clients.

Avoided Costs | Resource Valuation
Mr. Dickman provided expert testimony for PacifiCorp on various components included in a proposed method for valuing solar generation resources, the calculation of Public Utility Regulatory Policies Act avoided costs for large resources and support of modifications to the avoided cost calculation for small resources.

BRIAN DICKMAN
Partner
NewGen Strategies and Solutions, LLC 3

WORKSHOPS AND PRESENTATIONS
Host organizations and the topics Mr. Dickman presented are displayed below.

Advanced Workshop in Regulation and Competition, Center for Research in Regulated Industries, 2018
Customer Choice at a Vertically Integrated Utility
Advanced Workshop in Regulation and Competition, Center for Research in Regulated Industries, 2018