Application: 19-11-019 (U 39 M) Exhibit No.: (PG&E-2) Date: July 16, 2020 Witness(es): Various

PACIFIC GAS AND ELECTRIC COMPANY

2020 GENERAL RATE CASE PHASE II

EXHIBIT (PG&E-2)

COST OF SERVICE

JULY 2020 ERRATA

(PG&E-2)

-i-

PACIFIC GAS AND ELECTRIC COMPANY 2020 GENERAL RATE CASE PHASE II

EXHIBIT (PG&E-2) COST OF SERVICE

ERRATA TESTIMONY

TABLE OF CONTENTS

Chapter Title Witness

1 INTRODUCTION TO COST OF SERVICE ANALYSIS PROPOSALS

Amitava Dhar

Attachment A MARGINAL COST TABLE SCENARIO Amitava Dhar

2 MARGINAL GENERATION COSTS Jan Grygier 3 GENERATION ENERGY AND CAPACITY COST

OF SERVICE Louay Mardini

Attachment A GENERATION PEAK CAPACITY ALLOCATION

FACTOR ANALYSIS Louay Mardini

4 DEFERRABLE TRANSMISSION CAPACITY

PROJECTS Marco Rios

5 MARGINAL TRANSMISSION CAPACITY COSTS

AND COST CAUSATION STUDY Brian M. Lubeck

6 DISTRIBUTION EXPANSION PLANNING

PROCESS AND PROJECT COSTS Satvir Nagra

7 MARGINAL DISTRIBUTION CAPACITY COSTS Trevor Bergero 8 DISTRIBUTION CAPACITY COST OF SERVICE Amitava Dhar 9 MARGINAL CUSTOMER ACCESS COSTS Louay Mardini

Annette Taylor

Attachment A NEW CUSTOMER ONLY METHOD Louay Mardini

10 MARGINAL COST LOADERS AND FINANICAL FACTORS

Trevor Bergero Annette Taylor

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY 2020 GENERAL RATE CASE PHASE II

EXHIBIT (PG&E-2) COST OF SERVICE

ERRATA TESTIMONY

TABLE OF CONTENTS (CONTINUED)

-ii-

Chapter Title Witness

11 TIME-OF-USE PERIOD ASSESSMENT AND ANALYSIS

Jan Grygier

Attachment A PG&E SUPPLEMENTAL ADVICE

LETTER 5037-E-A (PG&E PROPOSED DEAD BAND TOLERANCE RANGE)

Jan Grygier

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 1

INTRODUCTION TO

COST OF SERVICE ANALYSIS PROPOSALS

(PG&E-2)

1-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 1

INTRODUCTION TO COST OF SERVICE ANALYSIS PROPOSALS

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 1-1

B. The Economic Theory of Marginal Costs .......................................................... 1-4

C. The CPUC Has Long Used Marginal Cost Based Ratemaking ........................ 1-5

D. PG&E’s Proposed Marginal Cost Approach ..................................................... 1-9

1. Marginal Generation Costs ...................................................................... 1-11

2. Marginal Transmission Capacity Costs .................................................... 1-13

3. Marginal Distribution Capacity Costs ....................................................... 1-14

4. Marginal Customer Access Costs ............................................................ 1-15

E. Summary of Improvements Made................................................................... 1-16

1. Generation Marginal Cost Model Improvements ...................................... 1-16

2. Distribution Cost of Service Model Improvements ................................... 1-16

3. Identifying NEM Customers for Cost of Service Analysis ......................... 1-18

F. Summary of What Remains the Same as in PG&E’s 2017 GRC Phase II Filing ............................................................................................................... 1-18

G. Importance and Relevance of Marginal Cost Applications ............................. 1-19

1. Marginal Cost-Based Ratemaking ........................................................... 1-19

2. Price Floors for Flexible Pricing ............................................................... 1-19

H. Organization of the Cost of Service Exhibit .................................................... 1-20

I. Conclusion ...................................................................................................... 1-22

(PG&E-2)

1-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 1 2

INTRODUCTION TO 3

COST OF SERVICE ANALYSIS PROPOSALS 4

A. Introduction 5

Cost of Service is regularly considered in Phase II of General Rate Case 6

(GRC) proceedings at the California Public Utilities Commission (CPUC or the 7

Commission). Exhibit (PG&E-2) presents Pacific Gas and Electric Company’s 8

(PG&E or the Utility) proposed marginal costs associated with electric 9

generation, transmission, and distribution, as well as generation and distribution 10

cost of service analyses. In this GRC Phase II filing, PG&E has made 11

improvements to its marginal cost-based cost of service methodologies to better 12

reflect increasing technology adoption in PG&E’s service territory in 13

recent years. 14

Specifically, technology adoption such as Solar Photovoltaic (PV),1 Storage, 15

Electric Vehicle Charging, and Energy Efficiency (EE) have led to significant 16

changes in PG&E’s total customer load as well as the customers’ load shapes 17

including the peak demand hours, which has already caused a shift of historic 18

mid-day Time of Use (TOU) periods into the evening. TOU Periods for the 19

Residential customers were addressed in PG&E’s 2015 Rate Design Window 20

and for Non-Residential customers in its 2017 GRC Phase II, where the summer 21

and winter on-peak periods were found to have shifted from daytime into the 22

evening.2 Consequently, the average cost of service and underlying unit 23

marginal costs have changed for PG&E’s customers. Therefore, it is more 24

important than ever to adopt an updated and more accurate cost of service 25

metric for the generation energy, generation capacity, and distribution capacity 26

costs to better reflect these changing circumstances. 27

PG&E has taken an important step towards analyzing cost of service 28

impacts due to changes in customer’s load shapes. This has required use of 29

“delivered” and “received” flows of electricity as may be the case for Net Energy 30

1 PV cells that convert light energy directly into electricity. 2 The Commission’s methodology for evaluating new revised TOU periods was set forth

in Decision (D.) 17-01-006 in Rulemaking (R.) 15-12-012.

(PG&E-2)

1-2

Metering (NEM)3 and storage customers.4 Due to the effect of NEM customers’ 1

solar generation during the sunlight hours of the day, NEM customers’ net load 2

profiles decrease significantly during the day whereas this does not happen with 3

Non-NEM customers. Additionally, the energy that is “received” from customers 4

can cause peak or near-peak demand that needs to be considered for allocation 5

of capacity costs. It is, therefore, important to improve capacity cost allocation 6

methodologies, such as Peak Capacity Allocation Factor (PCAF) coincident 7

demand and Final Line Transformer (FLT) non-coincident demand 8

methodologies used in this proceeding, by accounting for the direction of the 9

electricity flow and its impact on customers’ peak demand. PG&E’s improved 10

cost of service analysis quantifies the differences in costs between NEM and 11

Non-NEM customers. 12

To illustrate the results of its generation and distribution cost of service 13

analyses, PG&E utilizes Marginal Cost Revenue (MCR) per kilowatt-hour (kWh) 14

as the metric to compare cost of service across different load profiles for 15

different customer classes. 16

While average MCR per kWh has been traditionally used in revenue 17

allocation and rate design, it is also powerful in assessing cost of service since it 18

provides a standard unit of measure which reflects the effect of the relative unit 19

costs across the hours of the day and captures a higher average MCR per kWh 20

for a customer class with relatively higher load during the costlier hours. For 21

example, the peak load hours at the system level, which align with peak 22

Marginal Distribution Capacity Costs (MDCC) (4 p.m. – 10 p.m.)5 also generally 23

have higher hourly energy and generation capacity costs in comparison to the 24

hours with relatively lower load.6 Therefore, if the hourly load of a customer 25

class (e.g., Residential class) peaks during these system peak load hours, while 26

the hourly load of another customer class (e.g., Small Commercial class)7 peaks 27

3 NEM customers that have in-house generation, the vast majority of whom have

solar PV. 4 PG&E uses the term “delivered” to designate the flow of electricity from Utility grid to

customer premises, and the term “received” to designate the flow of electricity generated at a customer’s premises to the Utility grid.

5 As demonstrated in Figure 11-1 of Chapter 11 on TOU Period Review and Analysis. 6 As demonstrated in Chapter 2 on Marginal Generation Costs. 7 PG&E Rate Schedules A-1 and A-6.

(PG&E-2)

1-3

during other hours, then the average MCR per kWh of the Residential class will 1

be higher than that of the Small Commercial class. 2

Another example that illustrates this point is the hourly load shape of a 3

Residential class customer who has adopted solar PV. Generation from solar 4

PV causes the metered load to decrease significantly during daylight hours 5

(approximately 8 a.m. – 4 p.m.), but the hourly load during the system peak 6

hours (4 p.m. – 10 p.m.) does not change much due to the generation from solar 7

PV. Because the load decreases during the significantly less expensive hours of 8

a day due to the adoption of solar PV, the average MCR per kWh for the 9

customer group increases, since their usage is more heavily weighted in the 10

more expensive hours. This effect applies to both generation and distribution 11

cost of service and is illustrated in the cost of service analysis described in of 12

Exhibit (PG&E-2), Chapters 3 and 8. In summary, the standardized nature of 13

the MCR per kWh metric lends itself to comparing the effects of different load 14

profiles on varying marginal costs; it is therefore an effective metric for 15

comparing energy and capacity costs by time of day for each customer class. 16

This testimony demonstrates why the Commission should approve PG&E’s 17

improved marginal cost-based cost of service methodologies. These 18

improvements provide more accurate and granular marginal cost data for use in 19

revenue allocation and rate design that includes separate determination of the 20

costs of not only the “delivered” flow of electricity to customer service points, but 21

also the “received” flow of electricity back to the grid caused by customers’ 22

generation resources. This testimony presents, for the first time, segregated 23

data that allows comparative analysis of cost of service among PG&E’s 24

customers whose load shapes are now more diverse due to NEM adoption. 25

PG&E’s proposed enhanced cost of service methodologies provide both a better 26

framework and more accurate input data that enables a deeper understanding of 27

the cost of service particularly of the NEM and Non-NEM customers. 28

Although PG&E is not, at this time, proposing separate customer classes 29

based on whether a customer is NEM or not, PG&E does request that the CPUC 30

approve the methodologies presented in Exhibit (PG&E-2), including the unit 31

costs and cost drivers, as well as the marginal cost table8 used in revenue 32

8 The marginal cost table is being submitted as Attachment A of this chapter.

(PG&E-2)

1-4

allocation and rate design. This will also enable more accurate revenue 1

allocation and rate design in future proceedings. 2

The remainder of this chapter has the following sections: 3

Section B – The Economic Theory of Marginal Costs; 4

Section C – The CPUC has Long Used Marginal Cost Based Ratemaking; 5

Section D – PG&E’s Proposed Marginal Cost Approach; 6

Section E – Summary of Improvements Made; 7

Section F – Summary of What Remains the Same as in the 2017 GRC 8

Phase II Filing; 9

Section G – Importance and Relevance of Marginal Cost Applications; 10

Section H – Organization of the Cost of Service Exhibit; and 11

Section I – Conclusion. 12

B. The Economic Theory of Marginal Costs 13

As commonly defined by economists, the marginal cost of a particular 14

service measures the change in the total cost of providing that service caused 15

by a corresponding unit change in quantity of output.9 Long-established 16

economic theory holds that an additional unit of consumption should occur if, 17

and only if, the value of that consumption to the customer exceeds the marginal 18

cost. Thus, as a general principle, marginal cost-based pricing maximizes 19

economic efficiency. An important principle in developing marginal costs is that 20

they should be forward-looking and causally linked to a change in demand. 21

Given the regulated utility’s obligation to serve, an electric utility’s marginal 22

cost is the change in total cost to provide the designated service resulting from a 23

corresponding change in demand. The demand for electric generation is 24

measured in kilowatts (kW) for capacity and kWh for energy; the demand for 25

transmission and distribution (T&D) is measured in kW; and the demand for 26

customer access is measured in numbers of customers. 27

In most modern utility systems, including PG&E’s, the additional cost 28

incurred to serve an additional demand varies substantially and often depends 29

on location and time of demand. Aggregations across time intervals and 30

9 Mathematically, marginal cost is the change in total cost divided by the change in

output: MC = ΔTC / ΔQ. Ref: “Principles of Economics 2e” by Steven A. Greenlaw, University of Mary Washington, David Shapiro, Pennsylvania State University, (OpenStax, Rice University), Chapter 7, p. 166.

(PG&E-2)

1-5

geographic areas mask important costing detail. The consequence is less 1

accurate results for current and future utility marginal cost applications. 2

The marginal costs and methodological enhancements proposed in 3

Exhibit (PG&E-2) are designed to minimize these types of estimation 4

inaccuracies. 5

The principle that marginal cost estimates should be forward-looking, 6

especially for load growth-related costs, should be consistently applied. There 7

should be a causal link between a change in demand and any given expenditure 8

that goes into the calculation of marginal cost. To strengthen this link, it follows 9

that these forward-looking costs should be developed based on future resource 10

and investment plans, to the extent practicable. In addition, the length of time 11

over which these costs are estimated should be sufficiently long to encompass 12

the planning process for adding capital investments to address changes 13

in demand. 14

C. The CPUC Has Long Used Marginal Cost Based Ratemaking 15

The Commission’s use of marginal costs for the purpose of electric revenue 16

allocation and rate design dates back to 1981, when the CPUC transitioned from 17

the use of embedded costs to a marginal-cost-based approach: 18

We have chosen marginal costs as our foundation for [electric cost] 19 allocation and rate design. We have used marginal costs to promote 20 economic efficiency and to provide the greatest good for the greatest 21 number. (D.93887, 7 CPUC 2d 349, 492 (1981), emphasis added.) 22

The CPUC later reiterated that: 23

The theory behind adoption of marginal costs was that they would provide a 24 better price signal to customers of the impact of their consumption decisions 25 on the utility cost of providing service on a prospective basis and hopefully 26 would induce them to be more efficient. (D.92-12-057, 47 CPUC 2d 143, 27 276 (1992).) 28

Throughout the approximately 38-year history of Commission-adopted 29

marginal cost-based ratemaking, there has been a continual evolution in 30

marginal cost methodologies, with a general trend toward greater detail and 31

increased accuracy. Over time, Commission-adopted marginal cost 32

methodologies have become more reflective of key underlying factors, such as 33

the timing and location of growth-related investments that affect the cost of 34

service. Over the last 20 years, because the CPUC has adopted marginal cost 35

(PG&E-2)

1-6

settlements in PG&E’s GRC Phase II proceedings, the issue of marginal costs 1

has not been fully litigated since the 1990s. Significant CPUC developments 2

regarding marginal cost methodology during the mid- to late-1990s are 3

summarized below. 4

The seminal CPUC marginal cost decision was Decision (D.) 92-12-057, 5

deciding PG&E’s 1993 GRC Phase II proceeding. There, the Commission 6

adopted, with limited modifications, certain innovative PG&E proposals that the 7

CPUC hailed as “advancements” that make a “thorough alteration to our current 8

methodology of marginal costs.” (Id., mimeo, at p. 276). The CPUC called this 9

a “trial run” for PG&E electric proceedings only, with the goal being “to continue 10

to improve our methodology of sending the most accurate marginal cost price 11

signals to PG&E’s customers.” (Id.). The CPUC concluded: 12

[W]e agree with PG&E’s policy principle that marginal cost components 13 should be based on the design and operation of PG&E’s system, accurately 14 signal the cost of providing electrical service, be forward-looking, capture the 15 timing and magnitude of future investments, reflect geographic differences 16 where significant, reflect the value that PG&E’s customers place on electric 17 service, only include those costs actually incurred by PG&E for revenue 18 allocation purposes, and, finally, provide consistent signals in the evaluation 19 of supply and demand resources for planning purposes. Our goal is to more 20 fairly and equitably allocate responsibility to the several customer classes for 21 recovery of PG&E’s…revenue requirement…. We are committed that 22 marginal cost pricing, when refined sufficiently, will send price signals to 23 consumers which will guide resource planning for the future. In fact, a major 24 attraction of PG&E’s recommended changes is the forward-looking aspect of 25 its proposals. (Id., at 276-277, emphasis added.) 26

Included among the 1993 GRC’s advancements was, for the first time, 27

estimating marginal costs based on PG&E’s 13 operating divisions.10 The 28

CPUC agreed that “this substantially increases accuracy, thus sending price 29

signals which better reflect the differing costs customers cause PG&E to incur, 30

and furthermore provides the area-specific data necessary for future targeting of 31

customer energy efficiency programs.” (Id., at p. 303; Finding of Fact 32

(FOF) 158.) The CPUC went a step further and directed PG&E: 33

10 As a compromise between accuracy and manageability of data, in PG&E’s 1993 GRC,

the Commission adopted location-specific MDCCs based on PG&E’s then-existing 13 operating divisions. Division-level MDCC values are determined by aggregating the investments and loads of the Distribution Planning Areas (DPA) that make up the operating divisions.

(PG&E-2)

1-7

…to refine its original proposal of breaking down its area study to the TPA 1 [Transmission Planning Area] and DPA [Distribution Planning Area] levels in 2 its next General Rate Case because we endorse the concept that more 3 disaggregated data yields better and more equitable marginal costs for 4 different customer classes. (Id., at 303, emphasis added; FOF 159.) 5

In addition, the CPUC adopted PG&E’s proposed Present Worth Method 6

(PWM) for marginal cost estimation, because “it is the only method which 7

estimates the opportunity cost of deferring transmission and distribution 8

investments due to a change in load growth, taking into account both the timing 9

and magnitude of such changes.” (Id., FOF 160, p. 303.)11 10

In PG&E’s 1996 GRC, PG&E proposed marginal cost methodologies 11

seeking to continue and extend the advances12 adopted in the 1993 GRC 12

decision. In its decision on PG&E’s 1996 GRC Phase II (D.97-03-017, 13

71 CPUC 2d 212 (1997)), the Commission adopted PG&E’s proposed 14

location-specific marginal costs and New Customer Only (NCO) methodology, 15

with modifications. However, this time the CPUC rejected PG&E’s PWM based 16

on location-specific T&D costs,13 ordering PG&E to, for the time being, return to 17

the systemwide regression approach in use prior to 1993, but using 18

location-specific costs aggregated to the system level. In ordering a return to 19

the earlier methodology, the Commission stated: 20

…if PG&E is to get reliable [transmission and distribution (T&D)] marginal 21 cost estimates, it must improve the breadth, accuracy, and texture of its 22 [T&D planning] data. (Id., at p. 217). 23

11 Although the CPUC’s 1993 PG&E GRC Phase II decision (D.92-12-057) mentioned the

PWM as the only method for incorporating opportunity costs or time value of money with regard to the timing and size of T&D investments, it has recognized the Discounted Total Investment Method (DTIM) as another method that, likewise, recognizes the time value of money with respect to the timing and magnitude of investments. (See D.92-12-058.)

12 The Commission explicitly characterized PG&E’s marginal cost proposals in A.91-11-036 as “advancements.” (D.92-12-057, 47 CPUC 2d at p. 236.)

13 As the 1996 GRC was being litigated, the Commission was also involved in the early stages of electric industry restructuring, with Assembly Bill 1890 being signed in September 1996, during the pendency of that GRC proceeding. PG&E’s rates had also been frozen for several years. Rather than suspend the proceeding, the Commission chose to adopt new marginal costs “…or the limited purposes of payments to qualifying facilities…, evaluation of demand-side management (DSM) cost effectiveness, and price floors for discounted special contracts.” (D.97-03-017, 71 CPUC 2d 212, 217 (1997).) The Commission left the door open to further reconsideration once PG&E addressed and resolved the CPUC’s planning process concerns.

(PG&E-2)

1-8

In contrast, in its decision on Southern California Edison Company’s 1

(SCE) 1995 GRC Phase II (D.96-04-050), the Commission found that SCE 2

should compute T&D marginal costs by planning area consistent with the 3

methodology adopted for PG&E in D.92-12-057.14 PG&E has since addressed 4

these Commission concerns about its distribution planning process, as 5

discussed below.15 6

For PG&E’s 1999 GRC Phase II, PG&E continued to show that 7

location-specific16 marginal distribution costs are very important to a number 8

of marginal cost applications, including the two applications mentioned in 9

D.97-03-017 (DSM cost-effectiveness evaluation and price floors for flexible 10

pricing schedules). Therefore, in Application (A.) 99-03-014, PG&E submitted a 11

“Report of Findings Relating to Commission Directive from D.97-03-017 (Report 12

of Findings)” in which it detailed the improvements made to PG&E’s distribution 13

planning process in response to the 1996 GRC decision. Given the belief that 14

its improved distribution planning process satisfied the Commission’s concerns, 15

PG&E again proposed distribution marginal cost methodology with similar 16

location-specific, timing-sensitive characteristics as the methodologies adopted 17

in D.92-12-057. Before hearings could be held in the 1999 the GRC, the CPUC 18

suspended that proceeding in December 2000, due to the California Energy 19

Crisis.17 20

Although PG&E fully addressed the Commission’s 1996 GRC Phase II 21

decision’s distribution planning concerns, those prior concerns were not 22

expressly discussed in the CPUC’s decisions on PG&E’s 2003, 2007, 2011, 23

2014 and 2017 GRC Phase II applications, all of which were settled. Because 24

14 See D.96-04-050, 65 CPUC 2d 362, 400-402 (1996). 15 PG&E’s testimony submitted in its 1999 GRC Phase II proceeding (A.99-03-014,

Exhibit (PG&E-2), Chapter 3B) and in its 2003 GRC Phase II proceeding (A.04-06-024, Exhibit (PG&E-2), Appendix C) addressed the issues the CPUC raised in D.97-03-017. (See also Chapter 5 of Exhibit (PG&E-2), for a description of PG&E’s distribution expansion planning process.)

16 That is, specific to either PG&E’s then-current 18 operating divisions or PG&E’s 240-plus DPAs. For the purposes of revenue allocation and rate design, PG&E proposes division-level MDCC as adopted in D.92-12-057. The more detailed DPA-specific MDCC developed in PG&E’s workpapers are not used directly in PG&E’s revenue allocation but may be useful for local integrated resource planning applications as well as flexible pricing schedules and price floor calculations for Schedule E-31.

17 The Commission subsequently dismissed A.99-03-014 in D.03-01-012.

(PG&E-2)

1-9

no parties raised issues in any of the proceedings after PG&E improved its 1

distribution planning process, PG&E believes those issues have been 2

effectively resolved. 3

Regarding the estimation of Marginal Customer Access Costs (MCAC), the 4

Revenue Cycle Services (RCS) decision (D.98-09-070) required PG&E to 5

develop a detailed study of the costs it can avoid when others perform RCS 6

activities on behalf of PG&E distribution customers. The RCS models 7

developed for these studies were updated for subsequent GRC Phase II 8

marginal cost exhibits and were used to estimate marginal costs for certain 9

customer services such as meter reading and billing. 10

D. PG&E’s Proposed Marginal Cost Approach 11

PG&E’s proposed marginal cost approach is based on the economic theory 12

of marginal cost and the methods the Commission adopted in the 1990s, as 13

described above, with refinements. PG&E assessed the usefulness of marginal 14

cost results in their relevant applications and introduced refinements to marginal 15

cost approaches mostly based on improved data availability. 16

The starting point for calculating marginal costs is to identify cost drivers, 17

that is, those fundamental aspects of customer electricity requirements that 18

directly cause PG&E to incur costs. The next step is to calculate marginal costs 19

for unit changes in each cost driver by dividing the change in total cost by the 20

total change in the cost driver. The final step is to attribute these marginal costs 21

to measurable aspects of customer requirements such as energy consumption, 22

peak demand, and customer type. This allows the rate components most 23

associated with these measurable customer requirements, specifically energy 24

charges, demand charges, and monthly basic service fees (also known as 25

customer charges or fixed cost charges) to be set based on the corresponding 26

marginal cost components. 27

PG&E’s marginal costs are based upon three main cost drivers: (1) time of 28

electricity usage; (2) amount of peak demand; and (3) number of customers. 29

As to the first driver—Time of Electricity Usage—the cost of procuring 30

electricity to meet changes in customer electricity usage varies hourly. PG&E 31

and other load-serving entities are required to procure dependable generation 32

resources with capacity adequate to meet 115 percent of forecast demand. 33

Marginal Generation Costs (MGC), including energy and capacity, are 34

(PG&E-2)

1-10

associated with the electricity usage cost driver and aggregated in TOU periods 1

which group together hours with similar loads and costs. 2

For the second driver—Demand—PG&E’s electricity delivery system 3

consists of a network of high-voltage (transmission) and low-voltage 4

(distribution) facilities which connect generation resources to customer facilities. 5

The delivery system is designed and constructed to meet the expected demand 6

placed on it, so demand (capacity) is the associated cost driver. Demand is a 7

localized cost driver, since portions of PG&E’s delivery system peak at different 8

times depending on the area and the mix of customers by area. 9

The third driver is Number of Customers, which affects the marginal costs of 10

customer access to the distribution system and various customer services. 11

Since the marginal costs of customer access and customer services vary by 12

type of customer, these marginal costs are disaggregated by customer class. 13

PG&E’s MCACs are calculated based on PG&E’s proposed return to the Real 14

Economic Carrying Charge (RECC) Method. PG&E, however, also calculates 15

these costs using the NCO Method to allow comparison to previous proceedings 16

and to PG&E’s RECC proposal here. 17

Finally, there are some additional cost drivers involved in estimation of 18

marginal RCS costs and these are discussed in Chapter 9. 19

Accordingly, in Exhibit (PG&E-2), marginal costs are estimated in 20

three areas: 21

1) Generation-related demand and energy requirements (marginal generation 22

capacity and energy costs); 23

2) T&D demand requirements (marginal T&D capacity costs); and 24

3) Customer connections and operations and maintenance requirements, as 25

well as RCS requirements. 26

Exhibit (PG&E-2) presents marginal costs for the components of PG&E’s 27

electric system as shown in Tables 1-1, 1-2, and 1-3 provided at the end of 28

this chapter. 29

The main ratemaking applications of marginal costs are revenue allocation 30

and rate design. Other uses include EE and DSM resource planning 31

cost-effectiveness evaluations and economic development flexible pricing floor 32

(PG&E-2)

1-11

prices.18 Accurately estimated marginal costs are essential to efficient 1

allocation of resources so that customers can respond by choosing the correct 2

level of consumption. In electric ratemaking, marginal cost-based rates send 3

price signals to customers about the cost of their decisions to consume 4

additional units.19 5

The following sub-sections of this chapter summarize PG&E’s proposed 6

marginal cost methodologies as they apply to each of the major electric 7

utility functions. 8

1. Marginal Generation Costs 9

As described above, California electric utilities have traditionally filed 10

MGC, consisting of separate components for energy and capacity. In 11

addition, Commission-adopted methodology required the use of a rolling 12

6-year average to calculate the Marginal Generation Capacity Costs 13

(MGCC), reflecting a balance between short- and long-term marginal costs 14

in rates. 15

18 Accurately estimated, location-specific marginal costs are essential inputs to the price

floors for flexible pricing options such as the Economic Development Rate (EDR). 19 As stated by Alfred E. Kahn, in his seminal work on public utility economics:

“The central policy prescription of microeconomics is the equation of price and marginal cost. If economic theory is to have any relevance to public utility pricing, that is the point at which the inquiry must begin.” The Economics of Regulation: Principles and Institutions, Volume I, 1970, John Wiley & Sons, Inc., New York, p. 65. Kahn explains: “If consumers are to make the choices that will yield them the greatest possible satisfaction from society’s limited aggregate productive capacity, the prices that they pay for various goods and services available to them must accurately reflect their respective opportunity costs; only then will buyers be judging, in deciding what to buy and what not, whether the satisfaction they get from the purchase of any particular product is worth the sacrifice of other goods and services that its production entails.” Ibid., p. 66. Further, “[I]f consumers are to decide intelligently whether to take somewhat more or somewhat less of any particular item, the price they have to pay for it…must reflect the cost of supplying somewhat more or somewhat less—in short, marginal opportunity cost.” (Id., p. 66.)

(PG&E-2)

1-12

The MGC20 is presented in Exhibit (PG&E-2), Chapter 2, “Marginal 1

Generation Costs,” are estimates of the changes in PG&E’s electric 2

procurement costs that would be caused by small changes in customers’ 3

energy usage and coincident peak demand. PG&E is relying on public data 4

sources such as the inputs and outputs from the 2019-2020 Integrated 5

Resources Plan (IRP) for its inputs and models that can be shared with and 6

executed by all parties that have signed non-disclosure agreements, 7

providing greater transparency for its MEC and MGCC proposals. 8

In this proceeding, PG&E is proposing to continue to use a 6-year 9

planning horizon (2021-2027) to calculate its MGCC for rates and cost of 10

service calculations, while using a single year at the beginning of that period 11

(2021) for its MECs. While the MEC model has been enhanced from its 12

2017 GRC Phase II version (adding four more years of data to fit the model, 13

adding two new explanatory variables, and accounting for the operation of 14

grid-scale energy storage), the MGCC modeling and analysis addresses two 15

new categories of capacity costs (local and flexible capacity) to the system 16

capacity costs considered in prior GRCs. In Exhibit (PG&E-2), Chapter 3, 17

PG&E utilizes appropriate generation cost drivers along with the MEC and 18

MGCC to determine the generation MCR and MCR per kWh. Separately, 19

PG&E also calculates MGCC and MEC to consider potential changes to its 20

TOU periods, but, in accordance with D.17-01-006, that analysis (presented 21

in Exhibit (PG&E-2), Chapter 11) uses 2025-2030 as the planning horizon 22

for MGCCs, and 2025 for MECs. 23

20 The MGCs proposed in this testimony are not applicable for determining Qualifying

Facility (QF) Short-Run Avoided Costs (SRAC) or as-delivered capacity prices. The QF/Combined Heat and Power Program Settlement Agreement (Settlement Agreement) approved in D.10-12-035 includes a calculation of SRAC which apply to QF contracts entered into pursuant to the Public Utilities Regulatory Policy Act and the pro forma contracts adopted by the Settlement Agreement. As-delivered capacity prices for these contracts have also been adopted by the Commission. (See also, D.07-09-040.) Thus, the MGCs recommended in this testimony are not applicable for determining SRAC or as-delivered capacity prices for QFs or any other facility governed by the Settlement Agreement.

(PG&E-2)

1-13

2. Marginal Transmission Capacity Costs 1

The marginal transmission capacity costs (MTCC) proposed here are 2

used only for CPUC-jurisdictional retail rate design purposes.21 MTCCs 3

include only those planned investments that can be avoided or deferred if 4

load growth fails to materialize as expected. PG&E describes these 5

investments in Exhibit (PG&E-2), Chapter 4, “Deferrable Transmission 6

Capacity Projects.” 7

PG&E’s estimate of MTCC is described in Exhibit (PG&E-2), Chapter 5, 8

“Marginal Transmission Capacity Costs and Cost Causation Study.” 9

PG&E’s proposed MTCC is computed using the DTIM to better reflect the 10

lumpiness of investments and the time value of money. In 1992, the 11

Commission adopted the Total Investment Method (TIM) for gas 12

transmission and the PWM for electric transmission.22 The TIM simply 13

computes an annualized average investment per unit of demand; it is 14

inaccurate when investments are “lumpy” (i.e., when costs are spread 15

unevenly over time with large year-to-year variations in investment sizes)23 16

because it fails to weight near-term investments more heavily than those 17

occurring further out in time. Not including the time value of money is a 18

deficiency of the TIM. 19

The DTIM proposed here for MTCC (and MDCC) corrects for the defect 20

in the TIM of not including the time value of money. That is, the DTIM is 21

simply the discounted version of the TIM that takes into account the time 22

value of money. 23

While both the DTIM and the PWM incorporate the time value of money 24

to account for the timing and size of investments—a key methodology 25

attribute when costs are lumpy—PG&E proposes using the DTIM, because 26

it incorporates the time value for the capital additions and capacity additions, 27

and thus this method alone recognizes that capacity also has value. 28

21 Transmission rates, which are regulated by the Federal Energy Regulatory

Commission, are determined on an embedded cost basis. 22 See D.92-12-058 which adopted gas T&D marginal costs, and D.92-12-057 which

adopted electric marginal costs. 23 For example, major new high-voltage electric transmission facilities are not built every

year, and the costs for such a project will be incurred primarily in a few early years, with little or no costs in subsequent years.

(PG&E-2)

1-14

3. Marginal Distribution Capacity Costs 1

Consistent with the Commission’s practice since the advent of marginal 2

cost-based ratemaking in 1981, PG&E’s proposed MDCC are based on load 3

growth-related distribution investments only. Investments that are 4

unaffected by changes in demand, such as replacement costs and access 5

costs specifically incurred to connect new customers, are excluded from the 6

MDCC calculation. Exhibit (PG&E-2), Chapter 6, “PG&E’s Distribution 7

Expansion Planning Process and Project Costs,” describes the methods 8

used to develop circuit and substation transformer bank load forecasts and 9

distribution expansion plans, both of which are used to develop distribution 10

MDCCs. 11

PG&E’s service territory is very diverse in both geography and customer 12

density. PG&E has observed a variation in the marginal cost of distribution 13

capacity among its 243 DPAs that comprise its electric system, which have 14

been aggregated into 19 operating divisions as PG&E has done since its 15

1993 GRC Phase II. Therefore, PG&E has proposed MDCCs that utilize a 16

location-specific, timing-sensitive marginal cost methodology in every GRC 17

Phase II filing since 1993. PG&E proposes the same in this filing. 18

PG&E’s MDCCs consist of four components: (1) primary distribution 19

costs for the distribution circuits; (2) primary distribution costs for the 20

sub-stations; (3) new business primary costs; and (4) secondary costs. 21

PG&E proposes to use the same DTIM to calculate MDCCs of large 22

projects as it uses to estimate MTCC because of the lumpy nature of these 23

investments.24 DTIM-based MDCCs vary by area to reflect the fact that 24

investments during the planning horizon are needed at different times and in 25

different sizes for different areas depending on the installed capacity and 26

load growth unique to each area. PG&E does not recommend using the 27

National Economic Research Associates, Inc. (NERA) regression method, 28

because it mixes ten years of historical data with five years of forecast data. 29

It is inappropriate to mix historical data with forecast data because the 30

forecast used by distribution planners to determine whether an 31

24 While distribution investments are generally less lumpy than transmission investments,

geographic disaggregation increases the effect of lumpiness.

(PG&E-2)

1-15

enhancement is needed is based on a probability of one-in-ten years 1

occurrence. Historical data, therefore, does not align appropriately with the 2

planning forecast data. Furthermore, NERA regression utilizes cumulative 3

incremental capital addition along with cumulative incremental demand 4

growth. Consequently, historical data influences the estimate of MDCC 5

(which is the slope of zero intercept regression) disproportionately heavily 6

when compared to the forecast data, making the MDCC estimate hardly 7

forward looking. PG&E strongly recommends avoiding the use of historical 8

data for these reasons. The DTIM conforms to the Commission’s guidance 9

in D.92-12-057 and Commission-adopted marginal cost principles and is 10

well-suited for computing area-specific marginal costs. Because PG&E now 11

forecasts five years of load growth-related capacity investments for primary, 12

new business primary, and secondary projects, PG&E is proposing fully 13

forward-looking estimates for these components of MDCC. 14

PG&E’s estimates of MDCCs are discussed in detail in 15

Exhibit (PG&E-2), Chapter 7, “Marginal Distribution Capacity Costs.” 16

PG&E has also presented primary substation-related marginal capacity 17

cost calculations in Chapter 7, in compliance with D.18-08-013. In of Exhibit 18

(PG&E-2), Chapter 8, PG&E applies the appropriate distribution capacity 19

cost drivers to determine the MCR and MCR per kWh. 20

4. Marginal Customer Access Costs 21

MCACs are the sum of Marginal Connection Equipment Costs and 22

ongoing Marginal RCS costs. In this filing, PG&E proposes MCAC 23

estimates based on the RECC Method for the marginal connection 24

equipment cost, which has been used by PG&E in recent filings including 25

the 2017 GRC Phase II and was first adopted in 1993 GRC (D.92-12-057). 26

Marginal Connection Equipment Costs represents the connection equipment 27

capital costs, including transformers, service drops and meters. 28

PG&E has continued to use the Marginal RCS cost model used in its 29

2017 GRC Phase II filing. PG&E’s estimate of MCACs are provided in 30

Exhibit (PG&E-2), Chapter 9. 31

(PG&E-2)

1-16

E. Summary of Improvements Made 1

PG&E has implemented the following improvements in its cost of service 2

analysis: 3

1. Generation Marginal Cost Model Improvements 4

PG&E’s generation MEC Model described in Chapter 2 has been 5

improved by considering two additional explanatory variables (temperatures 6

East and North of California, and stream flows in the Pacific Northwest) to 7

capture the impacts of the energy imbalance market (EIM). The EIM, which 8

is a “residual market” that operates after the Day-Ahead market run by the 9

California Independent System Operator Corporation (CAISO), was initiated 10

in 2014 and now includes a large portion of load and generating units in the 11

Western Electricity Coordinating Council (WECC). The expansion of the 12

EIM has increased the linkage between California and the rest of the 13

WECC, and the additions to the MEC model demonstrably improve the 14

model fit. 15

As specified the TOU Order Instituting Rulemaking (OIR) (R.15-12-012), 16

PG&E is aligning both MECs and MGCCs with IRP modeling, by using both 17

inputs and outputs from the IRP in its forecasts. In addition, given the large 18

amount of Lithium-Ion Energy Storage (ES) capacity shown in the 19

Reference System Plan of the IRP, PG&E for the first time accounts for the 20

operation of grid-scale ES on MECs. 21

PG&E’s MGCC model described in Chapter 2 has also been improved, 22

by considering ES as a potential marginal capacity resource in addition to 23

the traditional resources of Combustion Turbines and Combined Cycle Gas 24

Turbines. Additionally, PG&E considers local capacity (in addition to the 25

system capacity which was modeled in prior GRCs), and flexible capacity 26

(ramping) need. In Chapter 3, PG&E calculates average generation MCR 27

per kWh based on the updated MGC for NEM and Non-NEM customers with 28

each customer class. 29

2. Distribution Cost of Service Model Improvements 30

PG&E’s MDCC model provides MDCC estimates at primary and 31

secondary voltage levels. At primary voltage level, it provides separate 32

MDCC estimates for the circuits and the sub-stations. It also provides a 33

(PG&E-2)

1-17

separate estimate for new business primary connections. In this 2020 GRC 1

Phase II showing, PG&E has improved how it calculates the peak demand 2

growth that is used in the denominator of MDCC calculations. Specifically, 3

PG&E now uses a cumulative, ratcheted (or “rolling maximum”) peak 4

demand calculation to more accurately reflect the fact that additional 5

distribution system capacity is needed only when the peak demand has 6

grown to levels beyond the maximum peak demand recorded in prior years. 7

A separate estimate of primary substation MDCC done for the first time 8

also serves to meet a compliance requirement from D.18-08-013.25 9

PG&E’s MCAC model has been improved to accurately include the 10

costs of Transformer, Service Drop and Meter (TSM), as well as RCS costs. 11

Upon reviewing its 2017 GRC Phase II calculations, PG&E found that it had 12

mistakenly included Rule 15 costs in the TSM costs for some non-residential 13

customers and in other cases, underestimated Rule 16 costs non-residential 14

customers. PG&E has corrected this in its current calculations. Similarly, 15

during development of residential customer costs, PG&E discovered that its 16

previous GRC filings had mistakenly assumed that Residential customers 17

might choose a 50 percent discount option in favor of the standard 18

refundable option, when, in fact, Residential customers do not qualify for 19

the 50 percent discount. PG&E has also corrected these prior errors in 20

this filing. 21

For its marginal RCS costs, PG&E has included two important changes 22

to the methodology it used in its 2017 GRC Phase II. First, the RCS costs 23

results are now based on the average of data from three years 24

(2015 - 2017) instead of a single year as was done in the 2017 filing.26 25

Additionally, PG&E is showing a breakout of the RCS costs at the customer 26

class level to allow comparison between NEM and Non-NEM customer 27

sub-groups. 28

25 D.18-08-013, Ordering Paragraph 23, p. 184. 26 Averaging over a period of three years was done based on the Motion for Adoption of

Marginal Cost and Revenue Allocation (MCRA) Settlement Agreement, filed at the CPUC and served on all parties on October 26, 2017 in A.16-06-013, PG&E’s 2017 GRC Phase II. See esp., p. 9, item 3 of the MCRA Settlement under the heading “Marginal Cost Settlement.”

(PG&E-2)

1-18

PG&E has improved its Distribution PCAF and FLT methods, as 1

mentioned in the “Introduction” section. These methodologies now 2

determine peak demand hours based on both “delivered” flows of electricity 3

to customer service points and “received” flows of electricity from customer 4

service points back to the grid, caused by technology adoptions such as 5

NEM and Storage. Additionally, distribution PCAFs have been calculated 6

based on circuit-level coincident demand, rather than division-level 7

coincident demand, to better capture the effects of delivered and received 8

energy flows, as well as to better align with PG&E’s demand-related capital 9

investment process. 10

3. Identifying NEM Customers for Cost of Service Analysis 11

PG&E has used the following three identifying categories to create NEM 12

customer sub-groups within each of its customer classes: 13

1. Standard NEM: Customers who have on-site electric generation 14

(solar, wind, fuel cell, etc.). This category accounts for approximately 15

98 percent of NEM customers’ net load. 16

2. NEM Aggregation: Customers with multiple meters on the same 17

property, or on adjacent properties who use renewable generation 18

(e.g., solar panels) to serve the aggregated load behind all eligible 19

meters and receive the benefits of NEM. 20

3. Virtual NEM: Customers in buildings with multiple individual meters who 21

use the renewable generation in the building to receive bill credits to 22

offset each benefiting account bill.27 23

F. Summary of What Remains the Same as in PG&E’s 2017 GRC Phase II 24

Filing 25

Two important aspects of the MDCC and MCAC methodologies have not 26

been changed because PG&E determined that the previous approaches remain 27

the best choices. For estimating MDCC, PG&E continues to employ the DTIM 28

which uses five years of forecast capital addition and demand growth data. 29

27 “RESBCTG” accounts, which are offsite generation-only facilities benefiting RESBCT

participants, have been excluded from the virtual NEM category.

(PG&E-2)

1-19

For estimating MCAC, PG&E continues to recommend using the RECC 1

Method for estimating TSM costs, as had been recommended in its 2017 GRC 2

Phase II testimony in A.16-06-013. 3

G. Importance and Relevance of Marginal Cost Applications 4

Accurately estimated marginal costs are essential to efficient electricity 5

pricing because they indicate to a consumer the cost of the customer’s decision 6

to consume an additional unit. Applications of marginal costs include revenue 7

allocation, rate design and flexible pricing to deter uneconomic bypass. When 8

based on marginal cost, rates provide appropriate incentives upon which 9

customers can make investment decisions. Pricing that is not based on 10

marginal costs can encourage customers to make consumption and investment 11

decisions that are not economically efficient. 12

1. Marginal Cost-Based Ratemaking 13

Prices set at marginal costs result in efficient resource allocation 14

because presumably consumers respond by choosing the correct level of 15

consumption.28 In a regulated environment, marginal costs are used to 16

promote economic efficiency by simulating the pricing structure—and the 17

resulting resource allocations—of a competitive market. However, 18

regulatory revenue requirements generally are greater than the revenues 19

that would result from pricing output at marginal cost. To overcome the 20

revenue shortfall, the Commission has adopted a scaler applicable to the 21

MCRs as necessary to meet revenue requirements relying on the Equal 22

Percentage of Marginal Costs method. 23

2. Price Floors for Flexible Pricing29 24

Location-specific distribution costs, along with customer-specific loads, 25

set the benchmark to gauge the benefit of a utility’s customer retention and 26

load attraction efforts for rates such as PG&E’s EDRs (Schedule EDR). For 27

28 That is, they would increase consumption until the point where the marginal benefit of

consuming an additional unit just equals (but does not exceed) the marginal cost. 29 In D.92-11-052, the Commission defined uneconomic bypass as follows: “Bypass is

uneconomic when a customer leaves the Utility system even though its cost to bypass is more than the marginal cost of utility service. In that situation, the Utility could still meet the bypass rate and obtain a positive contribution to its fixed costs, which helps keep other rates down.”

(PG&E-2)

1-20

instance, if the flexible price required to persuade a customer not to bypass 1

PG&E’s electric distribution system exceeds the marginal cost of serving the 2

customer, the remaining customers are made better off by PG&E’s customer 3

retention effort. Similarly, if new loads can be attracted to an area at rates in 4

excess of the area-specific marginal cost, they improve PG&E’s asset 5

utilization and help defray the rates paid by other customers.30 Accordingly, 6

marginal costs are used to evaluate the proposed EDR discussed in 7

Exhibit (PG&E-3), Chapter 7. 8

H. Organization of the Cost of Service Exhibit 9

Exhibit (PG&E-2) is organized as follows: 10

Chapter 1 – Introduction to Cost of Service Analysis Proposals. 11

Chapter 2 – Marginal Generation Costs. Chapter 2 describes the 12

two components of MGC: (1) MEC, expressed in cents per kWh; and 13

(2) MGCC, expressed in dollars per kW-year and also converted to cents 14

per kWh, with PG&E’s proposed MGC presented as the sum of MEC and 15

MGCC. PG&E uses MGCs to allocate generation costs to its bundled 16

electric service customers. 17

Chapter 3 – Generation Energy and Capacity Cost of Service. Chapter 3 18

describes how the MGC revenue per kWh compares across PG&E 19

customer classes. PG&E breaks out this data separately into NEM and 20

Non-NEM customer sub-groups to allow for comparison. The break-outs 21

within the customer classes show that NEM customer sub-groups 22

demonstrate a significantly higher average generation MCR per kWh when 23

compared to the corresponding Non-NEM customer sub-group. 24

Chapter 4 – Deferrable Transmission Capacity Projects. Chapter 4 25

describes the process PG&E uses to identify which transmission capacity 26

projects, contained in PG&E’s 2019 Electric Transmission Grid Expansion 27

Plan, could be deferred if electric demand growth did not materialize as 28

projected over the study period of five years from 2019-2024. These 29

Deferrable Transmission Capacity Projects provide the basis for PG&E’s 30

marginal Transmission Capacity Costs. 31

30 PG&E currently has the ability to offer contracts with reduced rates for purposes of load

attraction or retention under tariff Schedules E-31 and/or EDR.

(PG&E-2)

1-21

Chapter 5 – Marginal Transmission Capacity Costs and Cost Causation 1

Study. Chapter 5 presents PG&E’s estimates of demand-related Marginal 2

Transmission Capacity Costs, using the deferrable project costs identified in 3

Chapter 4 and applying DTIM. Chapter 5 also includes the reporting of 4

Transmission Cost of Service study, which is provided in compliance with 5

the requirements of the CPUC’s 2017 GRC Phase II Final Decision. 6

Chapter 6 – Distribution Expansion Planning Process and Project Costs. 7

Chapter 6 describes PG&E’s planning process for developing the 8

area-specific load forecasts and distribution expansion plans that PG&E 9

uses to develop its area marginal costs. 10

Chapter 7 – Marginal Distribution Capacity Costs. Chapter 7 presents 11

PG&E’s estimation of demand-related marginal new business primary, 12

primary and secondary distribution capacity costs, and also includes a 13

separate marginal capacity cost estimate for primary substations. PG&E 14

continues to use the DTIM method31 for estimating MDCC. Because area 15

expansion plan costs are driven by changes in demand, these changes in 16

demand represent marginal distribution system costs (excluding customer 17

access-related costs). 18

Chapter 8 –Distribution Capacity Cost of Service. Chapter 8 presents 19

PG&E’s improved distribution PCAF and FLT methods and the final 20

calculation of distribution capacity cost of service. MCR per kWh metric has 21

been used to break down the estimated cost of service separately for NEM 22

and Non-NEM customers. 23

Chapter 9 – Marginal Customer Access Costs. Chapter 9 describes PG&E’s 24

methodology for calculating MCAC and presents PG&E’s results. MCAC 25

costs have two major components: (1) TSM;32 and (2) RCS costs. PG&E 26

continues to propose RECC Method to estimate TSM costs. RCS 27

methodology and results are presented as well. 28

Chapter 10 – Marginal Costs Loaders and Financial Factors. Chapter 10 29

presents PG&E’s methodologies for T&D marginal cost loaders as well as 30

the financial factors needed for estimating marginal costs. 31

31 DTIM method was first proposed in 1999 by PG&E and has been used in all GRC filings

since then and is proposed in this GRC filing. 32 More properly, TSM.

(PG&E-2)

1-22

Chapter 11 –Time-of-Use Period Assessment and Analysis. Chapter 11 1

presents the required review of the appropriate months to include in the 2

summer season and TOU periods for summer and winter seasons, based 3

on the MGCs developed in Chapter 2 and primary MDCC developed in 4

Chapter 7 using the methodology approved by the Commission in its TOU 5

Periods, OIR (R.15-12-012). 6

In addition to the above chapters, Exhibit (PG&E-4) includes Appendix J, 7

“Schedule E-CREDIT Update,” which presents PG&E’s proposed credits for 8

Electric Schedule E-CREDIT applicable to DA and CCA providers. 9

I. Conclusion 10

Economic theory holds that economic efficiency is maximized when prices 11

are set at accurate marginal costs. For over three decades, the CPUC has 12

endorsed this principle as a basis for electric revenue allocation and rate design 13

among other applications, adding increased levels of accuracy over time. 14

PG&E’s marginal cost proposals in this proceeding are consistent with 15

sound marginal costing principles. Specifically, PG&E’s proposals: 16

1) Are forward-looking; 17

2) Reflect cost-causation; 18

3) Are based on PG&E-specific investments; 19

4) Signal the timing and magnitude of future investments; and 20

5) Reflect least-cost planning. 21

These attributes are important indicators of the suitability of marginal costs, 22

to provide an efficient, equitable, and practical means of allocating revenue and 23

designing rates, among other applications. In addition, the data and methods 24

described in the following chapters of Exhibit (PG&E-2) support the use of 25

PG&E’s marginal cost estimates in the ratemaking functions outlined in the 26

revenue allocation and rate design exhibit, among other applications described 27

above. 28

The methods PG&E has proposed here either comport with methods 29

previously adopted by the Commission, or are refinements to improve upon such 30

methods, including those enabled by significant improvements of the input data. 31

Such improvements have been made to marginal cost models such as MDCC 32

and MCAC. PG&E’s proposal also includes adopting the RECC (Rental 33

(PG&E-2)

1-23

Method) to estimate MCAC. These methodological improvements are discussed 1

in the respective chapters. 2

For the reasons stated herein, and in the succeeding chapters in 3

Exhibit (PG&E-2), the Commission should adopt PG&E’s proposed marginal 4

cost methodologies and results. 5

TABLE 1-1 OVERVIEW OF PROPOSED MARGINAL GENERATION COSTS DATA SOURCE, UNITS AND CALCULATION METHODOLOGY

Line No. Function

PG&E’s Methodology Submitted in the 2017 GRC Phase II

Proposed Methodology

1 Marginal Energy Cost

Data Source Forecasted market heat rates based on a model that is calibrated to historical market heat rates, using a public market price forecast for gas and GHG. Historical market heat rates are based on the historical hourly prices in the CAISO Market.

Same; extended calibration period with two additional explanatory variables, forecasting inputs from 2019-2020 IRP, with adjustments for the operation of grid-scale energy storage

2 Demand Measure

kWh by TOU period Same

3 Units ¢ per kWh Same

4 Marginal Generation Capacity Cost

Data Source Renewable Portfolio Standard (RPS) Calculator V6.1;

Long-Term Procurement Plan (LTPP) CPUC CAISO 2024, Trajectory scenario

2019-2020 IRP

5 Demand Measure

Peak kW Peak kW for System and Local capacity; kWh by TOU period for Flexible Capacity

6 Units $ per kW-year $ per kW-year for System and Local Capacity; ¢ per kWh for Flexible Capacity

7 Costing Methodology

A 6-year levelized capacity cost net of gross margin.

Resource balance year = beyond 2022.

Using PG&E’s Avoided Cost Model with an existing combined cycle unit for 2017-2022.

Reflects the difference between the annualized capital cost of a unit and the net energy benefits the unit earns in the energy market.

A 6-year levelized capacity cost net of gross margin.

Resource balance year = 2021

Using PG&E’s Avoided Cost Model with an existing combined cycle unit and public StorageVET model with new ES for 2021-2030.

Reflects the difference between the annualized capital cost of a unit and the net energy benefits the unit earns in the energy market.

(PG&E-2)

1-24

TABLE 1-2 OVERVIEW OF PROPOSED MARGINAL TRANSMISSION

AND DISTRIBUTION CAPACITY COSTS: DATA SOURCE, UNITS AND CALCULATION METHODOLOGY

MTCC Line No. Status Data Source Methodology Cost Driver Unit

1 Last Adopted CAISO filed PG&E transmission investment plans

Regression method

Peak Demand $ per kW-year

2 Proposed CAISO filed PG&E transmission investment plans

DTIM Peak Demand $ per kW-year

3 MDCCs: Primary, Secondary & New Business Primary

4 Status Data Source Methodology Cost Driver Unit

5 Last Adopted Investment plans and projections based on accounting data

As well as 10 years recorded data

Regression method

Division specific coincident Peak Demand for Primary

FLT non-coincident peak demand for secondary and new business primary

$ per kW-year

6 Proposed Investment plans and projections data

DTIM Distribution circuit specific coincident Peak Demand for Primary

FLT non-coincident peak demand for secondary and new business primary

$ per kW-year

(PG&E-2)

1-25

TABLE 1-3 OVERVIEW OF PROPOSED MARGINAL CUSTOMER ACCESS COSTS

DATA SOURCE, UNITS AND CALCULATION METHODOLOGY

MCAC Connection Equipment Component Line No. Status Data Source Methodology Cost Driver Unit

1 Last Adopted

Typical jobs from COMRESS estimating tool

NCO Method Net changes in billed customers

$ per customer-year

2 Proposed Customer Construction Billing System, Main Line Extension and various others

RECC Method

Number of new customers

$ per customer year

3 MCAC RCS Component

4 Status Data Source Methodology Cost Driver Unit

5 Last Adopted

Meter services data from Field Automation System, and other data from various other sources

Embedded costs as proxy for marginal cost

Labor rate and total job time

$ per customer year

6 Proposed Cost data from SAP, cost driver data from various PG&E internal data sources

Activity based costing

Cost drivers are designed for each sub-service. See Chapter 9 for details

$ per customer year

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 1

ATTACHMENT A

MARGINAL COST TABLE

Atta

chm

ent A

to C

hapt

er 1

, Exh

ibit

2M

argi

nal C

ost T

able

Sce

nario

Scen

ario

IDM

argi

nal C

ost

Rate

Cla

ssDG

Indi

cato

rDe

liver

ed/R

ecei

ved

Unit

Sum

mer

Off-

Peak

Sum

mer

Pea

kSu

mm

er P

art-

Peak

Win

ter O

ff-Pe

akW

inte

r Pea

kW

inte

r Sup

er-O

ff-Pe

akAn

nual

1GE

NCAP

A1N

DM

CR/k

Wh

$0.0

0000

$0.1

2887

$0.0

1837

$0.0

0009

$0.0

1405

$0.0

0000

1GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

0000

$0.1

6666

$0.0

2959

$0.0

0012

$0.0

1711

$0.0

0000

1GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0000

$0.0

1544

$0.0

0033

$0.0

0000

$0.0

0017

$0.0

0000

1GE

NCAP

A10

ND

MCR

/kW

h$0

.000

00$0

.132

05$0

.019

25$0

.000

10$0

.014

07$0

.000

001

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

0000

$0.1

6306

$0.0

2958

$0.0

0012

$0.0

1590

$0.0

0000

1GE

NCAP

A10

YR

MCR

/kW

h$0

.000

00$0

.024

58$0

.001

17$0

.000

01$0

.006

05$0

.000

001

GENC

APA6

ND

MCR

/kW

h$0

.000

00$0

.161

23$0

.027

06$0

.000

15$0

.016

44$0

.000

001

GENC

APA6

YD

MCR

/kW

h$0

.000

00$0

.195

50$0

.032

55$0

.000

13$0

.018

96$0

.000

001

GENC

APA6

YR

MCR

/kW

h$0

.000

00$0

.012

41$0

.000

16$0

.000

00$0

.000

08$0

.000

001

GENC

APAG

AN

DM

CR/k

Wh

$0.0

0000

$0.1

3346

$0.0

2395

$0.0

0018

$0.0

1495

$0.0

0000

1GE

NCAP

AGA

YD

MCR

/kW

h$0

.000

00$0

.177

66$0

.034

48$0

.000

21$0

.019

21$0

.000

001

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

1831

$0.0

0038

$0.0

0002

$0.0

0046

$0.0

0000

1GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

0000

$0.1

4012

$0.0

2365

$0.0

0016

$0.0

1628

$0.0

0000

1GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.1

8728

$0.0

3466

$0.0

0015

$0.0

2063

$0.0

0000

1GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0000

$0.0

1665

$0.0

0016

$0.0

0000

$0.0

0008

$0.0

0000

1GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.1

3117

$0.0

2208

$0.0

0018

$0.0

1600

$0.0

0000

1GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

0000

$0.2

0121

$0.0

3884

$0.0

0017

$0.0

2185

$0.0

0000

1GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

1642

$0.0

0020

$0.0

0000

$0.0

0016

$0.0

0000

1GE

NCAP

E1N

DM

CR/k

Wh

$0.0

0000

$0.1

3524

$0.0

2489

$0.0

0014

$0.0

1421

$0.0

0000

1GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

0000

$0.1

6540

$0.0

3401

$0.0

0019

$0.0

1636

$0.0

0000

1GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0000

$0.0

1312

$0.0

0023

$0.0

0000

$0.0

0004

$0.0

0000

1GE

NCAP

E19P

ND

MCR

/kW

h$0

.000

00$0

.133

62$0

.021

34$0

.000

09$0

.013

34$0

.000

001

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

0000

$0.1

8141

$0.0

3474

$0.0

0015

$0.0

1543

$0.0

0000

1GE

NCAP

E19P

YR

MCR

/kW

h$0

.000

00$0

.034

87$0

.001

85$0

.000

01$0

.005

54$0

.000

001

GENC

APE1

9PV

ND

MCR

/kW

h$0

.000

00$0

.143

24$0

.023

37$0

.000

09$0

.015

55$0

.000

001

GENC

APE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.165

22$0

.031

78$0

.000

14$0

.015

94$0

.000

001

GENC

APE1

9PV

YR

MCR

/kW

h$0

.000

00$0

.034

83$0

.003

11$0

.000

22$0

.001

96$0

.000

001

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.1

3970

$0.0

2161

$0.0

0010

$0.0

1448

$0.0

0000

1GE

NCAP

E19S

YD

MCR

/kW

h$0

.000

00$0

.170

20$0

.031

05$0

.000

13$0

.017

60$0

.000

001

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

1519

$0.0

0062

$0.0

0001

$0.0

0108

$0.0

0000

1GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

0000

$0.1

3970

$0.0

2295

$0.0

0011

$0.0

1422

$0.0

0000

1GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.1

6498

$0.0

3276

$0.0

0014

$0.0

1627

$0.0

0000

1GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0000

$0.0

8626

$0.0

0620

$0.0

0003

$0.0

0544

$0.0

0000

1GE

NCAP

E19T

ND

MCR

/kW

h$0

.000

00$0

.128

24$0

.021

41$0

.000

06$0

.010

48$0

.000

001

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE1

9TV

ND

MCR

/kW

h$0

.000

00$0

.136

82$0

.021

58$0

.000

09$0

.017

56$0

.000

001

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.1

3522

$0.0

2243

$0.0

0010

$0.0

1293

$0.0

0000

1GE

NCAP

E20P

YD

MCR

/kW

h$0

.000

00$0

.148

85$0

.024

48$0

.000

10$0

.015

38$0

.000

001

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

1606

$0.0

0035

$0.0

0000

$0.0

0066

$0.0

0000

1GE

NCAP

E20S

ND

MCR

/kW

h$0

.000

00$0

.136

29$0

.021

03$0

.000

10$0

.013

87$0

.000

001

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.1

7417

$0.0

2966

$0.0

0012

$0.0

2055

$0.0

0000

1GE

NCAP

E20S

YR

MCR

/kW

h$0

.000

00$0

.108

04$0

.004

71$0

.000

01$0

.008

98$0

.000

001

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.1

3369

$0.0

2233

$0.0

0009

$0.0

1280

$0.0

0000

1GE

NCAP

E20T

YD

MCR

/kW

h$0

.000

00$0

.170

85$0

.026

74$0

.000

12$0

.021

23$0

.000

001

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0518

$0.0

0071

$0.0

0000

$0.0

0063

$0.0

0000

1GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

0000

$0.1

4267

$0.0

2645

$0.0

0012

$0.0

1604

$0.0

0000

1GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.1

7288

$0.0

4906

$0.0

0018

$0.0

1882

$0.0

0000

1GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

1305

$0.0

0020

$0.0

0000

$0.0

0005

$0.0

0000

1GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.1

3276

$0.0

2222

$0.0

0012

$0.0

1334

$0.0

0000

1GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

0000

$0.1

9749

$0.0

3185

$0.0

0014

$0.0

1101

$0.0

0000

1GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.1

1419

$0.0

2005

$0.0

0012

$0.0

1730

$0.0

0000

1GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STL

ND

MCR

/kW

h$0

.000

00$0

.288

07$0

.042

87$0

.000

18$0

.020

53$0

.000

00

Page

1 o

f 29

(PG&E-2)

1-AtchA-1

1GE

NCAP

STL

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

TC1

ND

MCR

/kW

h$0

.000

00$0

.154

07$0

.026

84$0

.000

11$0

.015

03$0

.000

001

GENC

APTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

TC1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGA1

ND

MCR

/kW

h$0

.025

49$0

.065

00$0

.043

78$0

.029

87$0

.052

74-$

0.00

247

1GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3012

$0.0

6844

$0.0

4962

$0.0

3496

$0.0

5696

-$0.

0016

51

GENE

NGA1

YR

MCR

/kW

h$0

.014

15$0

.051

99$0

.031

30$0

.009

95$0

.020

54-$

0.00

458

1GE

NENG

A10

ND

MCR

/kW

h$0

.025

53$0

.065

17$0

.043

98$0

.029

79$0

.052

51-$

0.00

234

1GE

NENG

A10

YD

MCR

/kW

h$0

.028

83$0

.068

34$0

.049

57$0

.034

18$0

.056

04-$

0.00

222

1GE

NENG

A10

YR

MCR

/kW

h$0

.014

06$0

.051

47$0

.030

65$0

.013

96$0

.031

04-$

0.00

513

1GE

NENG

A6N

DM

CR/k

Wh

$0.0

2809

$0.0

6722

$0.0

4796

$0.0

3277

$0.0

5535

-$0.

0028

81

GENE

NGA6

YD

MCR

/kW

h$0

.031

46$0

.069

69$0

.050

79$0

.035

77$0

.058

24-$

0.00

207

1GE

NENG

A6Y

RM

CR/k

Wh

$0.0

1297

$0.0

5116

$0.0

3051

$0.0

0893

$0.0

1947

-$0.

0051

41

GENE

NGAG

AN

DM

CR/k

Wh

$0.0

2669

$0.0

6667

$0.0

4662

$0.0

3019

$0.0

5039

-$0.

0033

01

GENE

NGAG

AY

DM

CR/k

Wh

$0.0

2997

$0.0

7004

$0.0

5178

$0.0

3511

$0.0

5633

-$0.

0032

31

GENE

NGAG

AY

RM

CR/k

Wh

$0.0

1574

$0.0

5305

$0.0

3299

$0.0

1286

$0.0

2413

-$0.

0037

31

GENE

NGAG

BLg

ND

MCR

/kW

h$0

.026

50$0

.066

55$0

.046

04$0

.030

00$0

.051

01-$

0.00

263

1GE

NENG

AGBL

gY

DM

CR/k

Wh

$0.0

3103

$0.0

6907

$0.0

5106

$0.0

3482

$0.0

5611

-$0.

0023

41

GENE

NGAG

BLg

YR

MCR

/kW

h$0

.013

78$0

.052

99$0

.031

24$0

.011

76$0

.022

35-$

0.00

420

1GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

2518

$0.0

6630

$0.0

4552

$0.0

2783

$0.0

4901

-$0.

0025

71

GENE

NGAG

BSm

YD

MCR

/kW

h$0

.032

27$0

.070

10$0

.052

93$0

.035

67$0

.057

24-$

0.00

215

1GE

NENG

AGBS

mY

RM

CR/k

Wh

$0.0

1440

$0.0

5280

$0.0

3204

$0.0

1198

$0.0

2267

-$0.

0040

01

GENE

NGE1

ND

MCR

/kW

h$0

.026

53$0

.066

55$0

.046

45$0

.032

42$0

.055

20-$

0.00

367

1GE

NENG

E1Y

DM

CR/k

Wh

$0.0

3210

$0.0

6939

$0.0

5178

$0.0

3785

$0.0

5869

-$0.

0034

41

GENE

NGE1

YR

MCR

/kW

h$0

.015

44$0

.051

00$0

.032

08$0

.011

16$0

.020

61-$

0.00

360

1GE

NENG

E19P

ND

MCR

/kW

h$0

.025

88$0

.062

68$0

.043

46$0

.030

15$0

.050

86-$

0.00

265

1GE

NENG

E19P

YD

MCR

/kW

h$0

.029

61$0

.067

20$0

.050

04$0

.034

20$0

.054

16-$

0.00

210

1GE

NENG

E19P

YR

MCR

/kW

h$0

.013

25$0

.049

74$0

.028

83$0

.018

75$0

.036

34-$

0.00

483

1GE

NENG

E19P

VN

DM

CR/k

Wh

$0.0

2757

$0.0

6611

$0.0

4604

$0.0

3209

$0.0

5422

-$0.

0029

41

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.029

90$0

.068

18$0

.051

00$0

.035

88$0

.055

72-$

0.00

241

1GE

NENG

E19P

VY

RM

CR/k

Wh

$0.0

1388

$0.0

5857

$0.0

3126

$0.0

1436

$0.0

3268

-$0.

0040

51

GENE

NGE1

9SN

DM

CR/k

Wh

$0.0

2670

$0.0

6561

$0.0

4505

$0.0

3092

$0.0

5289

-$0.

0024

61

GENE

NGE1

9SY

DM

CR/k

Wh

$0.0

2931

$0.0

6877

$0.0

4996

$0.0

3437

$0.0

5585

-$0.

0023

61

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

1314

$0.0

4994

$0.0

3002

$0.0

1268

$0.0

2604

-$0.

0055

41

GENE

NGE1

9SV

ND

MCR

/kW

h$0

.026

76$0

.065

87$0

.045

62$0

.031

39$0

.053

33-$

0.00

306

1GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

2926

$0.0

6858

$0.0

5053

$0.0

3487

$0.0

5610

-$0.

0026

41

GENE

NGE1

9SV

YR

MCR

/kW

h$0

.022

51$0

.060

05$0

.036

41$0

.027

43$0

.039

93-$

0.00

434

1GE

NENG

E19T

ND

MCR

/kW

h$0

.025

95$0

.061

46$0

.043

19$0

.029

98$0

.049

24-$

0.00

337

1GE

NENG

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE1

9TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

E19T

VN

DM

CR/k

Wh

$0.0

2676

$0.0

6438

$0.0

4490

$0.0

3269

$0.0

5480

-$0.

0028

91

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

2626

$0.0

6293

$0.0

4396

$0.0

3031

$0.0

5067

-$0.

0028

21

GENE

NGE2

0PY

DM

CR/k

Wh

$0.0

2732

$0.0

6387

$0.0

4502

$0.0

3102

$0.0

5200

-$0.

0025

11

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

1195

$0.0

5015

$0.0

2909

$0.0

1094

$0.0

2382

-$0.

0049

11

GENE

NGE2

0SN

DM

CR/k

Wh

$0.0

2673

$0.0

6528

$0.0

4492

$0.0

3072

$0.0

5237

-$0.

0024

21

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

2982

$0.0

6786

$0.0

4853

$0.0

3372

$0.0

5535

-$0.

0023

01

GENE

NGE2

0SY

RM

CR/k

Wh

$0.0

1331

$0.0

5778

$0.0

3132

$0.0

1769

$0.0

4117

-$0.

0055

51

GENE

NGE2

0TN

DM

CR/k

Wh

$0.0

2616

$0.0

6206

$0.0

4349

$0.0

3005

$0.0

4996

-$0.

0029

21

GENE

NGE2

0TY

DM

CR/k

Wh

$0.0

2802

$0.0

6384

$0.0

4497

$0.0

3122

$0.0

5229

-$0.

0014

01

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

0929

$0.0

5865

$0.0

2914

$0.0

1613

$0.0

3815

-$0.

0087

41

GENE

NGNo

nE1

ND

MCR

/kW

h$0

.027

02$0

.066

76$0

.047

13$0

.032

57$0

.055

78-$

0.00

362

1GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

3247

$0.0

7033

$0.0

5650

$0.0

3787

$0.0

5903

-$0.

0034

11

GENE

NGNo

nE1

YR

MCR

/kW

h$0

.015

89$0

.051

47$0

.032

61$0

.012

12$0

.021

43-$

0.00

360

1GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

2592

$0.0

6215

$0.0

4519

$0.0

3107

$0.0

5181

-$0.

0032

81

GENE

NGST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYP

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYS

ND

MCR

/kW

h$0

.029

52$0

.067

80$0

.047

91$0

.032

57$0

.054

46-$

0.00

136

1GE

NENG

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

STBY

TN

DM

CR/k

Wh

$0.0

2508

$0.0

5976

$0.0

4258

$0.0

3182

$0.0

5323

-$0.

0017

81

GENE

NGST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYT

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

LN

DM

CR/k

Wh

$0.0

3600

$0.0

7350

$0.0

5687

$0.0

4052

$0.0

6318

-$0.

0033

4

Page

2 o

f 29

(PG&E-2)

1-AtchA-2

1GE

NENG

STL

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

TC1

ND

MCR

/kW

h$0

.029

13$0

.066

50$0

.047

96$0

.033

85$0

.055

30-$

0.00

337

1GE

NENG

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A10

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A10

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A10

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A6N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA6

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A6Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

AN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

AY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BLg

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BLg

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BSm

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XNo

nE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XNo

nE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYS

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

LN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

3 o

f 29

(PG&E-2)

1-AtchA-3

1GE

NFLE

XST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XTC

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1RC

SA1

0N

X$/

CUST

-YR

$146

.136

981

RCS

A10

YX

$/CU

ST-Y

R$6

90.7

8185

1RC

SA1

1PN

X$/

CUST

-YR

$40.

7401

91

RCS

A11P

YX

$/CU

ST-Y

R$4

32.4

1524

1RC

SA1

3PN

X$/

CUST

-YR

$48.

7312

91

RCS

A13P

YX

$/CU

ST-Y

R$5

72.4

5656

1RC

SA6

1PN

X$/

CUST

-YR

$40.

7401

91

RCS

A61P

YX

$/CU

ST-Y

R$4

32.4

1524

1RC

SA6

3PN

X$/

CUST

-YR

$48.

7312

91

RCS

A63P

YX

$/CU

ST-Y

R$5

72.4

5656

1RC

SAG

AN

X$/

CUST

-YR

$81.

6850

11

RCS

AGA

YX

$/CU

ST-Y

R$5

37.1

1738

1RC

SAG

BLg

NX

$/CU

ST-Y

R$1

99.2

5382

1RC

SAG

BLg

YX

$/CU

ST-Y

R$8

77.7

7068

1RC

SAG

BSm

NX

$/CU

ST-Y

R$1

43.4

6167

1RC

SAG

BSm

YX

$/CU

ST-Y

R$8

23.7

8007

1RC

SE1

NX

$/CU

ST-Y

R$3

0.65

877

1RC

SE1

YX

$/CU

ST-Y

R$9

5.64

576

1RC

SE1

9PN

X$/

CUST

-YR

$1,3

31.6

3063

1RC

SE1

9PY

X$/

CUST

-YR

$1,4

87.5

9280

1RC

SE1

9SN

X$/

CUST

-YR

$1,3

31.6

3063

1RC

SE1

9SY

X$/

CUST

-YR

$1,4

87.5

9280

1RC

SE1

9VP

NX

$/CU

ST-Y

R$1

46.1

3698

1RC

SE1

9VP

YX

$/CU

ST-Y

R$6

90.7

8185

1RC

SE1

9VS

NX

$/CU

ST-Y

R$1

46.1

3698

1RC

SE1

9VS

YX

$/CU

ST-Y

R$6

90.7

8185

1RC

SE2

0PN

X$/

CUST

-YR

$1,3

31.6

3063

1RC

SE2

0PY

X$/

CUST

-YR

$1,4

87.5

9280

1RC

SE2

0SN

X$/

CUST

-YR

$1,3

31.6

3063

1RC

SE2

0SY

X$/

CUST

-YR

$1,4

87.5

9280

1RC

SNo

nE1

NX

$/CU

ST-Y

R$4

4.88

502

1RC

SNo

nE1

YX

$/CU

ST-Y

R$1

16.9

3407

1RC

SLS

1N

X$/

CUST

-YR

$13.

1900

01

RCS

LS2

NX

$/CU

ST-Y

R$7

.740

001

RCS

LS3

NX

$/CU

ST-Y

R$1

2.94

000

1RC

SO

L1N

X$/

CUST

-YR

$14.

6000

01

RCS

TC1

NX

$/CU

ST-Y

R$3

1.21

801

1M

CEC

A10P

NX

$/CU

ST-Y

R$4

,649

.992

561

MCE

CA1

0PY

X$/

CUST

-YR

$4,6

49.9

9256

1M

CEC

A10S

NX

$/CU

ST-Y

R$3

,746

.295

841

MCE

CA1

0SY

X$/

CUST

-YR

$3,7

46.2

9584

1M

CEC

A11P

NX

$/CU

ST-Y

R$3

93.9

2632

1M

CEC

A11P

YX

$/CU

ST-Y

R$3

93.9

2632

1M

CEC

A13P

NX

$/CU

ST-Y

R$1

,486

.507

841

MCE

CA1

3PY

X$/

CUST

-YR

$1,4

86.5

0784

1M

CEC

A61P

NX

$/CU

ST-Y

R$3

93.9

2632

1M

CEC

A61P

YX

$/CU

ST-Y

R$3

93.9

2632

1M

CEC

A63P

NX

$/CU

ST-Y

R$1

,486

.507

841

MCE

CA6

3PY

X$/

CUST

-YR

$1,4

86.5

0784

1M

CEC

AGA

NX

$/CU

ST-Y

R$7

56.4

6342

1M

CEC

AGA

YX

$/CU

ST-Y

R$7

56.4

6342

1M

CEC

AGBL

gN

X$/

CUST

-YR

$2,5

45.2

5837

1M

CEC

AGBL

gY

X$/

CUST

-YR

$2,5

45.2

5837

1M

CEC

AGBS

mN

X$/

CUST

-YR

$2,5

45.2

5837

1M

CEC

AGBS

mY

X$/

CUST

-YR

$2,5

45.2

5837

1M

CEC

E1N

X$/

CUST

-YR

$102

.155

281

MCE

CE1

YX

$/CU

ST-Y

R$1

02.1

5528

1M

CEC

E19P

NX

$/CU

ST-Y

R$5

,070

.673

671

MCE

CE1

9PY

X$/

CUST

-YR

$5,0

70.6

7367

1M

CEC

E19S

NX

$/CU

ST-Y

R$5

,603

.542

331

MCE

CE1

9SY

X$/

CUST

-YR

$5,6

03.5

4233

Page

4 o

f 29

(PG&E-2)

1-AtchA-4

1M

CEC

E19V

PN

X$/

CUST

-YR

$4,6

49.9

9256

1M

CEC

E19V

PY

X$/

CUST

-YR

$4,6

49.9

9256

1M

CEC

E19V

SN

X$/

CUST

-YR

$3,7

46.2

9584

1M

CEC

E19V

SY

X$/

CUST

-YR

$3,7

46.2

9584

1M

CEC

E20P

NX

$/CU

ST-Y

R$5

,070

.673

671

MCE

CE2

0PY

X$/

CUST

-YR

$5,0

70.6

7367

1M

CEC

E20S

NX

$/CU

ST-Y

R$5

,919

.986

821

MCE

CE2

0SY

X$/

CUST

-YR

$5,9

19.9

8682

1M

CEC

NonE

1N

X$/

CUST

-YR

$102

.155

281

MCE

CNo

nE1

YX

$/CU

ST-Y

R$1

02.1

5528

1M

CEC

LS1

NX

$/CU

ST-Y

R$3

1.40

362

1M

CEC

LS2

NX

$/CU

ST-Y

R$1

7.84

780

1M

CEC

LS3

NX

$/CU

ST-Y

R$5

2.57

811

1M

CEC

TC1

NX

$/CU

ST-Y

R$4

84.9

6168

2GE

NCAP

A1N

DM

CR/k

Wh

$0.0

1934

$0.0

9398

$0.0

0000

$0.0

0060

$0.0

1476

$0.0

0000

2GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

2771

$0.1

2558

$0.0

0000

$0.0

0082

$0.0

1863

$0.0

0000

2GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0023

$0.0

1451

$0.0

0000

$0.0

0000

$0.0

0014

$0.0

0000

2GE

NCAP

A10

ND

MCR

/kW

h$0

.019

48$0

.096

47$0

.000

00$0

.000

62$0

.014

71$0

.000

002

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

2591

$0.1

2320

$0.0

0000

$0.0

0080

$0.0

1713

$0.0

0000

2GE

NCAP

A10

YR

MCR

/kW

h$0

.001

17$0

.018

12$0

.000

00$0

.000

08$0

.005

69$0

.000

002

GENC

APA6

ND

MCR

/kW

h$0

.023

39$0

.115

71$0

.000

00$0

.000

84$0

.017

81$0

.000

002

GENC

APA6

YD

MCR

/kW

h$0

.028

43$0

.155

08$0

.000

00$0

.000

92$0

.020

94$0

.000

002

GENC

APA6

YR

MCR

/kW

h$0

.000

07$0

.012

24$0

.000

00$0

.000

00$0

.000

07$0

.000

002

GENC

APAG

AN

DM

CR/k

Wh

$0.0

1910

$0.0

9420

$0.0

0000

$0.0

0090

$0.0

1490

$0.0

0000

2GE

NCAP

AGA

YD

MCR

/kW

h$0

.027

05$0

.135

48$0

.000

00$0

.001

18$0

.020

59$0

.000

002

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0025

$0.0

1754

$0.0

0000

$0.0

0004

$0.0

0018

$0.0

0000

2GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

1987

$0.0

9977

$0.0

0000

$0.0

0083

$0.0

1648

$0.0

0000

2GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

2914

$0.1

4352

$0.0

0000

$0.0

0108

$0.0

2218

$0.0

0000

2GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0005

$0.0

1659

$0.0

0000

$0.0

0000

$0.0

0007

$0.0

0000

2GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

1760

$0.0

9247

$0.0

0000

$0.0

0085

$0.0

1559

$0.0

0000

2GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

3212

$0.1

5820

$0.0

0000

$0.0

0124

$0.0

2361

$0.0

0000

2GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0007

$0.0

1635

$0.0

0000

$0.0

0000

$0.0

0015

$0.0

0000

2GE

NCAP

E1N

DM

CR/k

Wh

$0.0

3011

$0.0

9735

$0.0

0000

$0.0

0082

$0.0

1543

$0.0

0000

2GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

4289

$0.1

2348

$0.0

0000

$0.0

0117

$0.0

1831

$0.0

0000

2GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0012

$0.0

1259

$0.0

0000

$0.0

0000

$0.0

0004

$0.0

0000

2GE

NCAP

E19P

ND

MCR

/kW

h$0

.018

30$0

.097

51$0

.000

00$0

.000

58$0

.014

11$0

.000

002

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

2820

$0.1

3848

$0.0

0000

$0.0

0087

$0.0

1675

$0.0

0000

2GE

NCAP

E19P

YR

MCR

/kW

h$0

.002

19$0

.025

45$0

.000

00$0

.000

10$0

.005

52$0

.000

002

GENC

APE1

9PV

ND

MCR

/kW

h$0

.020

12$0

.103

63$0

.000

00$0

.000

60$0

.016

80$0

.000

002

GENC

APE1

9PV

YD

MCR

/kW

h$0

.025

65$0

.125

91$0

.000

00$0

.000

85$0

.017

24$0

.000

002

GENC

APE1

9PV

YR

MCR

/kW

h$0

.003

16$0

.020

63$0

.000

00$0

.000

29$0

.000

89$0

.000

002

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

1953

$0.1

0209

$0.0

0000

$0.0

0063

$0.0

1518

$0.0

0000

2GE

NCAP

E19S

YD

MCR

/kW

h$0

.026

19$0

.128

53$0

.000

00$0

.000

86$0

.019

01$0

.000

002

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0066

$0.0

1134

$0.0

0000

$0.0

0003

$0.0

0089

$0.0

0000

2GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

2134

$0.1

0159

$0.0

0000

$0.0

0067

$0.0

1504

$0.0

0000

2GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

2893

$0.1

2312

$0.0

0000

$0.0

0091

$0.0

1757

$0.0

0000

2GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0766

$0.0

6063

$0.0

0000

$0.0

0012

$0.0

0528

$0.0

0000

2GE

NCAP

E19T

ND

MCR

/kW

h$0

.016

87$0

.091

55$0

.000

00$0

.000

44$0

.011

22$0

.000

002

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE1

9TV

ND

MCR

/kW

h$0

.018

13$0

.100

82$0

.000

00$0

.000

68$0

.019

04$0

.000

002

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

1911

$0.0

9860

$0.0

0000

$0.0

0058

$0.0

1365

$0.0

0000

2GE

NCAP

E20P

YD

MCR

/kW

h$0

.020

92$0

.110

49$0

.000

00$0

.000

66$0

.016

51$0

.000

002

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0022

$0.0

1562

$0.0

0000

$0.0

0001

$0.0

0061

$0.0

0000

2GE

NCAP

E20S

ND

MCR

/kW

h$0

.018

55$0

.099

12$0

.000

00$0

.000

59$0

.014

48$0

.000

002

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

2737

$0.1

3228

$0.0

0000

$0.0

0101

$0.0

2211

$0.0

0000

2GE

NCAP

E20S

YR

MCR

/kW

h$0

.007

95$0

.088

52$0

.000

00$0

.000

08$0

.009

18$0

.000

002

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

1829

$0.0

9770

$0.0

0000

$0.0

0056

$0.0

1358

$0.0

0000

2GE

NCAP

E20T

YD

MCR

/kW

h$0

.023

08$0

.129

34$0

.000

00$0

.001

01$0

.022

48$0

.000

002

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0077

$0.0

0217

$0.0

0000

$0.0

0001

$0.0

0066

$0.0

0000

2GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

2655

$0.1

0283

$0.0

0000

$0.0

0075

$0.0

1769

$0.0

0000

2GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

4616

$0.1

3049

$0.0

0000

$0.0

0118

$0.0

2136

$0.0

0000

2GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0006

$0.0

1300

$0.0

0000

$0.0

0000

$0.0

0004

$0.0

0000

2GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

1512

$0.0

9762

$0.0

0000

$0.0

0060

$0.0

1443

$0.0

0000

Page

5 o

f 29

(PG&E-2)

1-AtchA-5

2GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

2603

$0.1

4913

$0.0

0000

$0.0

0050

$0.0

1179

$0.0

0000

2GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

1612

$0.0

8029

$0.0

0000

$0.0

0057

$0.0

1921

$0.0

0000

2GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STL

ND

MCR

/kW

h$0

.031

33$0

.249

30$0

.000

00$0

.001

04$0

.025

66$0

.000

002

GENC

APST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APTC

1N

DM

CR/k

Wh

$0.0

2171

$0.1

0960

$0.0

0000

$0.0

0067

$0.0

1646

$0.0

0000

2GE

NCAP

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

A1N

DM

CR/k

Wh

$0.0

3242

$0.0

6246

$0.0

0000

$0.0

2741

$0.0

5067

$0.0

0000

2GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3766

$0.0

6562

$0.0

0000

$0.0

3448

$0.0

5555

$0.0

0000

2GE

NENG

A1Y

RM

CR/k

Wh

$0.0

1880

$0.0

5189

$0.0

0000

$0.0

0458

$0.0

2039

$0.0

0000

2GE

NENG

A10

ND

MCR

/kW

h$0

.032

34$0

.062

61$0

.000

00$0

.027

22$0

.050

37$0

.000

002

GENE

NGA1

0Y

DM

CR/k

Wh

$0.0

3619

$0.0

6571

$0.0

0000

$0.0

3313

$0.0

5441

$0.0

0000

2GE

NENG

A10

YR

MCR

/kW

h$0

.018

57$0

.050

80$0

.000

00$0

.007

92$0

.028

62$0

.000

002

GENE

NGA6

ND

MCR

/kW

h$0

.035

19$0

.063

95$0

.000

00$0

.030

87$0

.053

27$0

.000

002

GENE

NGA6

YD

MCR

/kW

h$0

.038

67$0

.066

81$0

.000

00$0

.035

65$0

.057

01$0

.000

002

GENE

NGA6

YR

MCR

/kW

h$0

.017

92$0

.051

14$0

.000

00$0

.003

63$0

.019

41$0

.000

002

GENE

NGAG

AN

DM

CR/k

Wh

$0.0

3324

$0.0

6349

$0.0

0000

$0.0

2681

$0.0

4706

$0.0

0000

2GE

NENG

AGA

YD

MCR

/kW

h$0

.037

41$0

.067

13$0

.000

00$0

.034

72$0

.054

14$0

.000

002

GENE

NGAG

AY

RM

CR/k

Wh

$0.0

2058

$0.0

5297

$0.0

0000

$0.0

0764

$0.0

2363

$0.0

0000

2GE

NENG

AGBL

gN

DM

CR/k

Wh

$0.0

3306

$0.0

6357

$0.0

0000

$0.0

2657

$0.0

4811

$0.0

0000

2GE

NENG

AGBL

gY

DM

CR/k

Wh

$0.0

3823

$0.0

6638

$0.0

0000

$0.0

3433

$0.0

5439

$0.0

0000

2GE

NENG

AGBL

gY

RM

CR/k

Wh

$0.0

1843

$0.0

5298

$0.0

0000

$0.0

0631

$0.0

2232

$0.0

0000

2GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

3159

$0.0

6320

$0.0

0000

$0.0

2316

$0.0

4556

$0.0

0000

2GE

NENG

AGBS

mY

DM

CR/k

Wh

$0.0

3978

$0.0

6750

$0.0

0000

$0.0

3592

$0.0

5569

$0.0

0000

2GE

NENG

AGBS

mY

RM

CR/k

Wh

$0.0

1914

$0.0

5279

$0.0

0000

$0.0

0664

$0.0

2265

$0.0

0000

2GE

NENG

E1N

DM

CR/k

Wh

$0.0

3614

$0.0

6381

$0.0

0000

$0.0

3127

$0.0

5329

$0.0

0000

2GE

NENG

E1Y

DM

CR/k

Wh

$0.0

4295

$0.0

6654

$0.0

0000

$0.0

3884

$0.0

5752

$0.0

0000

2GE

NENG

E1Y

RM

CR/k

Wh

$0.0

1955

$0.0

5091

$0.0

0000

$0.0

0563

$0.0

2060

$0.0

0000

2GE

NENG

E19P

ND

MCR

/kW

h$0

.031

83$0

.060

05$0

.000

00$0

.028

12$0

.048

80$0

.000

002

GENE

NGE1

9PY

DM

CR/k

Wh

$0.0

3685

$0.0

6446

$0.0

0000

$0.0

3347

$0.0

5253

$0.0

0000

2GE

NENG

E19P

YR

MCR

/kW

h$0

.017

42$0

.049

13$0

.000

00$0

.013

21$0

.033

56$0

.000

002

GENE

NGE1

9PV

ND

MCR

/kW

h$0

.033

92$0

.063

34$0

.000

00$0

.030

35$0

.052

32$0

.000

002

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.037

10$0

.065

43$0

.000

00$0

.035

16$0

.053

95$0

.000

002

GENE

NGE1

9PV

YR

MCR

/kW

h$0

.018

24$0

.056

06$0

.000

00$0

.006

58$0

.030

13$0

.000

002

GENE

NGE1

9SN

DM

CR/k

Wh

$0.0

3305

$0.0

6296

$0.0

0000

$0.0

2862

$0.0

5075

$0.0

0000

2GE

NENG

E19S

YD

MCR

/kW

h$0

.036

50$0

.066

13$0

.000

00$0

.033

27$0

.054

17$0

.000

002

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

1771

$0.0

4933

$0.0

0000

$0.0

0693

$0.0

2392

$0.0

0000

2GE

NENG

E19S

VN

DM

CR/k

Wh

$0.0

3367

$0.0

6310

$0.0

0000

$0.0

2936

$0.0

5116

$0.0

0000

2GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

3719

$0.0

6585

$0.0

0000

$0.0

3418

$0.0

5438

$0.0

0000

2GE

NENG

E19S

VY

RM

CR/k

Wh

$0.0

2718

$0.0

5743

$0.0

0000

$0.0

2190

$0.0

3652

$0.0

0000

2GE

NENG

E19T

ND

MCR

/kW

h$0

.031

39$0

.058

67$0

.000

00$0

.027

93$0

.047

00$0

.000

002

GENE

NGE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE1

9TV

ND

MCR

/kW

h$0

.032

84$0

.061

48$0

.000

00$0

.031

18$0

.053

08$0

.000

002

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

3231

$0.0

6026

$0.0

0000

$0.0

2834

$0.0

4854

$0.0

0000

2GE

NENG

E20P

YD

MCR

/kW

h$0

.033

42$0

.061

29$0

.000

00$0

.029

65$0

.050

20$0

.000

002

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

1663

$0.0

5010

$0.0

0000

$0.0

0535

$0.0

2283

$0.0

0000

2GE

NENG

E20S

ND

MCR

/kW

h$0

.032

85$0

.062

64$0

.000

00$0

.028

27$0

.050

19$0

.000

002

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

3683

$0.0

6529

$0.0

0000

$0.0

3303

$0.0

5365

$0.0

0000

2GE

NENG

E20S

YR

MCR

/kW

h$0

.019

05$0

.056

82$0

.000

00$0

.011

14$0

.039

72$0

.000

002

GENE

NGE2

0TN

DM

CR/k

Wh

$0.0

3196

$0.0

5944

$0.0

0000

$0.0

2824

$0.0

4790

$0.0

0000

2GE

NENG

E20T

YD

MCR

/kW

h$0

.033

85$0

.061

59$0

.000

00$0

.030

52$0

.050

74$0

.000

002

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

1396

$0.0

5734

$0.0

0000

$0.0

1105

$0.0

3554

$0.0

0000

2GE

NENG

NonE

1N

DM

CR/k

Wh

$0.0

3537

$0.0

6393

$0.0

0000

$0.0

3134

$0.0

5409

$0.0

0000

2GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

4306

$0.0

6776

$0.0

0000

$0.0

3895

$0.0

5811

$0.0

0000

2GE

NENG

NonE

1Y

RM

CR/k

Wh

$0.0

2026

$0.0

5147

$0.0

0000

$0.0

0679

$0.0

2142

$0.0

0000

2GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

3153

$0.0

5836

$0.0

0000

$0.0

2873

$0.0

4976

$0.0

0000

Page

6 o

f 29

(PG&E-2)

1-AtchA-6

2GE

NENG

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

SN

DM

CR/k

Wh

$0.0

3556

$0.0

6531

$0.0

0000

$0.0

3056

$0.0

5254

$0.0

0000

2GE

NENG

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

TN

DM

CR/k

Wh

$0.0

3085

$0.0

5666

$0.0

0000

$0.0

3044

$0.0

5210

$0.0

0000

2GE

NENG

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STL

ND

MCR

/kW

h$0

.043

05$0

.069

63$0

.000

00$0

.041

58$0

.063

63$0

.000

002

GENE

NGST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGTC

1N

DM

CR/k

Wh

$0.0

3554

$0.0

6334

$0.0

0000

$0.0

3247

$0.0

5335

$0.0

0000

2GE

NENG

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A6N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA6

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A6Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BLg

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BLg

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BSm

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XNo

nE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XNo

nE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

7 o

f 29

(PG&E-2)

1-AtchA-7

2GE

NFLE

XST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYS

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

A1N

DM

CR/k

Wh

$0.0

1781

$0.0

0246

$0.0

7047

$0.0

0084

$0.0

0404

$0.0

0000

3GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

2013

$0.0

0323

$0.1

0355

$0.0

0093

$0.0

0640

$0.0

0000

3GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0176

$0.0

0091

$0.0

0171

$0.0

0000

$0.0

0001

$0.0

0000

3GE

NCAP

A10

ND

MCR

/kW

h$0

.017

57$0

.002

41$0

.070

92$0

.000

81$0

.003

99$0

.000

003

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

1991

$0.0

0317

$0.0

8994

$0.0

0091

$0.0

0571

$0.0

0000

3GE

NCAP

A10

YR

MCR

/kW

h$0

.002

59$0

.000

66$0

.004

64$0

.000

27$0

.000

47$0

.000

003

GENC

APA6

ND

MCR

/kW

h$0

.016

98$0

.002

28$0

.081

33$0

.000

87$0

.005

22$0

.000

003

GENC

APA6

YD

MCR

/kW

h$0

.019

54$0

.002

14$0

.112

38$0

.000

93$0

.007

25$0

.000

003

GENC

APA6

YR

MCR

/kW

h$0

.001

56$0

.000

99$0

.001

17$0

.000

00$0

.000

00$0

.000

003

GENC

APAG

AN

DM

CR/k

Wh

$0.0

1528

$0.0

0239

$0.0

6799

$0.0

0070

$0.0

0452

$0.0

0000

3GE

NCAP

AGA

YD

MCR

/kW

h$0

.019

91$0

.003

09$0

.100

81$0

.000

87$0

.007

89$0

.000

003

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0206

$0.0

0111

$0.0

0262

$0.0

0000

$0.0

0004

$0.0

0000

3GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

1608

$0.0

0254

$0.0

7238

$0.0

0073

$0.0

0429

$0.0

0000

3GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

2020

$0.0

0299

$0.1

0849

$0.0

0098

$0.0

0705

$0.0

0000

3GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0212

$0.0

0120

$0.0

0243

$0.0

0000

$0.0

0001

$0.0

0000

3GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

1518

$0.0

0231

$0.0

6151

$0.0

0064

$0.0

0381

$0.0

0000

3GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

2143

$0.0

0300

$0.1

1804

$0.0

0098

$0.0

0765

$0.0

0000

3GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0210

$0.0

0126

$0.0

0226

$0.0

0001

$0.0

0001

$0.0

0000

3GE

NCAP

E1N

DM

CR/k

Wh

$0.0

2557

$0.0

0308

$0.0

9029

$0.0

0101

$0.0

0513

$0.0

0000

3GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

3153

$0.0

0320

$0.1

2172

$0.0

0116

$0.0

0779

$0.0

0000

3GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0098

$0.0

0071

$0.0

0083

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

E19P

ND

MCR

/kW

h$0

.014

34$0

.002

16$0

.068

56$0

.000

66$0

.003

79$0

.000

003

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

1897

$0.0

0331

$0.1

0152

$0.0

0080

$0.0

0593

$0.0

0000

3GE

NCAP

E19P

YR

MCR

/kW

h$0

.003

67$0

.001

08$0

.009

45$0

.000

28$0

.000

72$0

.000

003

GENC

APE1

9PV

ND

MCR

/kW

h$0

.015

34$0

.002

35$0

.074

18$0

.000

81$0

.004

71$0

.000

003

GENC

APE1

9PV

YD

MCR

/kW

h$0

.018

33$0

.003

15$0

.093

60$0

.000

88$0

.006

08$0

.000

003

GENC

APE1

9PV

YR

MCR

/kW

h$0

.001

70$0

.000

99$0

.009

14$0

.000

00$0

.000

87$0

.000

003

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

1622

$0.0

0243

$0.0

7369

$0.0

0076

$0.0

0412

$0.0

0000

3GE

NCAP

E19S

YD

MCR

/kW

h$0

.019

67$0

.003

09$0

.093

54$0

.001

00$0

.006

42$0

.000

003

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0123

$0.0

0051

$0.0

0254

$0.0

0002

$0.0

0010

$0.0

0000

3GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

1726

$0.0

0239

$0.0

7543

$0.0

0079

$0.0

0423

$0.0

0000

3GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

2134

$0.0

0323

$0.0

9436

$0.0

0095

$0.0

0600

$0.0

0000

3GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0719

$0.0

0104

$0.0

3021

$0.0

0009

$0.0

0062

$0.0

0000

3GE

NCAP

E19T

ND

MCR

/kW

h$0

.012

86$0

.002

27$0

.063

47$0

.000

54$0

.003

02$0

.000

003

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE1

9TV

ND

MCR

/kW

h$0

.014

54$0

.002

11$0

.072

66$0

.000

90$0

.005

15$0

.000

003

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

1472

$0.0

0223

$0.0

7108

$0.0

0066

$0.0

0379

$0.0

0000

3GE

NCAP

E20P

YD

MCR

/kW

h$0

.016

11$0

.002

36$0

.080

52$0

.000

79$0

.004

71$0

.000

003

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0243

$0.0

0138

$0.0

0321

$0.0

0006

$0.0

0011

$0.0

0000

3GE

NCAP

E20S

ND

MCR

/kW

h$0

.015

19$0

.002

49$0

.071

16$0

.000

74$0

.003

99$0

.000

003

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

2089

$0.0

0280

$0.1

0265

$0.0

0111

$0.0

0682

$0.0

0000

3GE

NCAP

E20S

YR

MCR

/kW

h$0

.018

05$0

.000

56$0

.031

74$0

.000

53$0

.001

07$0

.000

003

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

1404

$0.0

0217

$0.0

6980

$0.0

0061

$0.0

0366

$0.0

0000

3GE

NCAP

E20T

YD

MCR

/kW

h$0

.019

45$0

.002

64$0

.098

32$0

.000

84$0

.006

00$0

.000

003

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0056

$0.0

0000

$0.0

0188

$0.0

0023

$0.0

0003

$0.0

0000

3GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

2087

$0.0

0265

$0.0

8373

$0.0

0100

$0.0

0546

$0.0

0000

3GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

3032

$0.0

0943

$0.1

2525

$0.0

0134

$0.0

0904

$0.0

0000

3GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0111

$0.0

0080

$0.0

0085

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

0994

$0.0

0090

$0.0

5638

$0.0

0056

$0.0

0433

$0.0

0000

Page

8 o

f 29

(PG&E-2)

1-AtchA-8

3GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

1710

$0.0

0163

$0.1

0402

$0.0

0051

$0.0

0321

$0.0

0000

3GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

1296

$0.0

0205

$0.0

6884

$0.0

0095

$0.0

0675

$0.0

0000

3GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STL

ND

MCR

/kW

h$0

.014

44$0

.002

32$0

.123

84$0

.000

87$0

.009

68$0

.000

003

GENC

APST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APTC

1N

DM

CR/k

Wh

$0.0

1458

$0.0

0222

$0.0

7879

$0.0

0075

$0.0

0487

$0.0

0000

3GE

NCAP

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

A1N

DM

CR/k

Wh

$0.0

3590

$0.0

2925

$0.0

4003

$0.0

3624

$0.0

2854

$0.0

0000

3GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3861

$0.0

3347

$0.0

5248

$0.0

3985

$0.0

3879

$0.0

0000

3GE

NENG

A1Y

RM

CR/k

Wh

$0.0

2162

$0.0

1933

$0.0

0703

$0.0

1021

$0.0

0410

$0.0

0000

3GE

NENG

A10

ND

MCR

/kW

h$0

.035

57$0

.028

99$0

.039

88$0

.036

02$0

.028

26$0

.000

003

GENE

NGA1

0Y

DM

CR/k

Wh

$0.0

3788

$0.0

3188

$0.0

4804

$0.0

3971

$0.0

3651

$0.0

0000

3GE

NENG

A10

YR

MCR

/kW

h$0

.023

32$0

.017

52$0

.007

30$0

.019

00$0

.005

81$0

.000

003

GENE

NGA6

ND

MCR

/kW

h$0

.037

10$0

.028

03$0

.043

91$0

.038

32$0

.032

72$0

.000

003

GENE

NGA6

YD

MCR

/kW

h$0

.038

59$0

.031

55$0

.055

11$0

.040

24$0

.041

40$0

.000

003

GENE

NGA6

YR

MCR

/kW

h$0

.021

47$0

.019

25$0

.006

04$0

.009

52$0

.003

41$0

.000

003

GENE

NGAG

AN

DM

CR/k

Wh

$0.0

3572

$0.0

2749

$0.0

4104

$0.0

3562

$0.0

2687

$0.0

0000

3GE

NENG

AGA

YD

MCR

/kW

h$0

.038

29$0

.031

40$0

.051

61$0

.040

18$0

.041

73$0

.000

003

GENE

NGAG

AY

RM

CR/k

Wh

$0.0

2284

$0.0

2156

$0.0

0963

$0.0

1322

$0.0

0687

$0.0

0000

3GE

NENG

AGBL

gN

DM

CR/k

Wh

$0.0

3565

$0.0

2831

$0.0

4033

$0.0

3511

$0.0

2598

$0.0

0000

3GE

NENG

AGBL

gY

DM

CR/k

Wh

$0.0

3891

$0.0

3182

$0.0

5191

$0.0

3977

$0.0

3812

$0.0

0000

3GE

NENG

AGBL

gY

RM

CR/k

Wh

$0.0

2171

$0.0

2083

$0.0

0857

$0.0

1288

$0.0

0596

$0.0

0000

3GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

3438

$0.0

2738

$0.0

3704

$0.0

3230

$0.0

2160

$0.0

0000

3GE

NENG

AGBS

mY

DM

CR/k

Wh

$0.0

3963

$0.0

3282

$0.0

5543

$0.0

4065

$0.0

4153

$0.0

0000

3GE

NENG

AGBS

mY

RM

CR/k

Wh

$0.0

2199

$0.0

2174

$0.0

0893

$0.0

1284

$0.0

0640

$0.0

0000

3GE

NENG

E1N

DM

CR/k

Wh

$0.0

3941

$0.0

3109

$0.0

4947

$0.0

3948

$0.0

3430

$0.0

0000

3GE

NENG

E1Y

DM

CR/k

Wh

$0.0

4276

$0.0

3755

$0.0

6192

$0.0

4289

$0.0

4623

$0.0

0000

3GE

NENG

E1Y

RM

CR/k

Wh

$0.0

2045

$0.0

1908

$0.0

0758

$0.0

1032

$0.0

0506

$0.0

0000

3GE

NENG

E19P

ND

MCR

/kW

h$0

.034

26$0

.026

80$0

.039

12$0

.035

60$0

.028

28$0

.000

003

GENE

NGE1

9PY

DM

CR/k

Wh

$0.0

3695

$0.0

3056

$0.0

5066

$0.0

3850

$0.0

3746

$0.0

0000

3GE

NENG

E19P

YR

MCR

/kW

h$0

.022

90$0

.018

18$0

.008

09$0

.026

64$0

.010

79$0

.000

003

GENE

NGE1

9PV

ND

MCR

/kW

h$0

.036

45$0

.028

57$0

.041

86$0

.037

91$0

.030

76$0

.000

003

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.038

21$0

.032

27$0

.050

45$0

.041

02$0

.038

33$0

.000

003

GENE

NGE1

9PV

YR

MCR

/kW

h$0

.021

54$0

.015

87$0

.014

17$0

.017

75$0

.004

09$0

.000

003

GENE

NGE1

9SN

DM

CR/k

Wh

$0.0

3587

$0.0

2888

$0.0

4056

$0.0

3682

$0.0

2904

$0.0

0000

3GE

NENG

E19S

YD

MCR

/kW

h$0

.038

13$0

.031

15$0

.048

46$0

.039

85$0

.036

32$0

.000

003

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

2164

$0.0

1747

$0.0

0628

$0.0

1765

$0.0

0505

$0.0

0000

3GE

NENG

E19S

VN

DM

CR/k

Wh

$0.0

3651

$0.0

2845

$0.0

4240

$0.0

3754

$0.0

3012

$0.0

0000

3GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

3867

$0.0

3180

$0.0

5035

$0.0

4051

$0.0

3766

$0.0

0000

3GE

NENG

E19S

VY

RM

CR/k

Wh

$0.0

3217

$0.0

1962

$0.0

2214

$0.0

3271

$0.0

1345

$0.0

0000

3GE

NENG

E19T

ND

MCR

/kW

h$0

.033

81$0

.025

83$0

.039

09$0

.034

83$0

.026

77$0

.000

003

GENE

NGE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE1

9TV

ND

MCR

/kW

h$0

.036

20$0

.027

63$0

.041

36$0

.038

23$0

.031

61$0

.000

003

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

3460

$0.0

2690

$0.0

4020

$0.0

3573

$0.0

2839

$0.0

0000

3GE

NENG

E20P

YD

MCR

/kW

h$0

.035

42$0

.028

27$0

.042

52$0

.036

52$0

.030

81$0

.000

003

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

2031

$0.0

1969

$0.0

0825

$0.0

1485

$0.0

0454

$0.0

0000

3GE

NENG

E20S

ND

MCR

/kW

h$0

.035

55$0

.028

87$0

.039

63$0

.036

48$0

.028

48$0

.000

003

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

3862

$0.0

3178

$0.0

4967

$0.0

3923

$0.0

3561

$0.0

0000

3GE

NENG

E20S

YR

MCR

/kW

h$0

.024

68$0

.018

21$0

.014

50$0

.026

49$0

.009

89$0

.000

003

GENE

NGE2

0TN

DM

CR/k

Wh

$0.0

3415

$0.0

2652

$0.0

3976

$0.0

3524

$0.0

2809

$0.0

0000

3GE

NENG

E20T

YD

MCR

/kW

h$0

.036

18$0

.030

68$0

.043

86$0

.036

61$0

.032

91$0

.000

003

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

1957

$0.0

1689

$0.0

1085

$0.0

2754

$0.0

0903

$0.0

0000

3GE

NENG

NonE

1N

DM

CR/k

Wh

$0.0

3801

$0.0

2988

$0.0

4639

$0.0

3906

$0.0

3369

$0.0

0000

3GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

4232

$0.0

3972

$0.0

6208

$0.0

4283

$0.0

4657

$0.0

0000

3GE

NENG

NonE

1Y

RM

CR/k

Wh

$0.0

2175

$0.0

2023

$0.0

0824

$0.0

1144

$0.0

0638

$0.0

0000

3GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

3335

$0.0

2236

$0.0

3810

$0.0

3580

$0.0

2941

$0.0

0000

Page

9 o

f 29

(PG&E-2)

1-AtchA-9

3GE

NENG

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

SN

DM

CR/k

Wh

$0.0

3709

$0.0

2855

$0.0

4288

$0.0

3868

$0.0

2992

$0.0

0000

3GE

NENG

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

TN

DM

CR/k

Wh

$0.0

3372

$0.0

2704

$0.0

3914

$0.0

3732

$0.0

3286

$0.0

0000

3GE

NENG

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STL

ND

MCR

/kW

h$0

.040

45$0

.028

09$0

.061

17$0

.043

21$0

.053

99$0

.000

003

GENE

NGST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGTC

1N

DM

CR/k

Wh

$0.0

3719

$0.0

2758

$0.0

4450

$0.0

3918

$0.0

3355

$0.0

0000

3GE

NENG

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A6N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA6

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A6Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BLg

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BLg

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BSm

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XNo

nE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XNo

nE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

10

of 2

9

(PG&E-2)

1-AtchA-10

3GE

NFLE

XST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYS

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1PR

IDIS

TA1

ND

MCR

/kW

h$0

.007

64$0

.067

14$0

.035

98$0

.001

18$0

.003

75$0

.001

631

PRID

IST

A1Y

DM

CR/k

Wh

$0.0

0459

$0.0

7616

$0.0

3097

$0.0

0103

$0.0

0413

$0.0

0101

1PR

IDIS

TA1

YR

MCR

/kW

h$0

.005

35$0

.038

58$0

.017

18$0

.000

85$0

.000

95$0

.000

491

PRID

IST

A10

ND

MCR

/kW

h$0

.008

62$0

.063

09$0

.037

33$0

.001

14$0

.002

95$0

.001

701

PRID

IST

A10

YD

MCR

/kW

h$0

.005

82$0

.067

83$0

.032

07$0

.001

51$0

.003

75$0

.002

761

PRID

IST

A10

YR

MCR

/kW

h$0

.005

20$0

.027

99$0

.015

48$0

.000

62$0

.000

87$0

.000

481

PRID

IST

A6N

DM

CR/k

Wh

$0.0

0499

$0.0

6534

$0.0

2618

$0.0

0094

$0.0

0421

$0.0

0127

1PR

IDIS

TA6

YD

MCR

/kW

h$0

.006

11$0

.080

01$0

.027

76$0

.001

66$0

.003

33$0

.001

591

PRID

IST

A6Y

RM

CR/k

Wh

$0.0

0350

$0.0

3669

$0.0

1683

$0.0

0058

$0.0

0066

-$0.

0001

21

PRID

IST

AGA

ND

MCR

/kW

h$0

.008

41$0

.075

58$0

.026

66$0

.001

12$0

.006

82$0

.001

401

PRID

IST

AGA

YD

MCR

/kW

h$0

.004

77$0

.083

07$0

.023

21$0

.002

26$0

.002

30$0

.000

741

PRID

IST

AGA

YR

MCR

/kW

h$0

.004

15$0

.040

69$0

.013

27$0

.000

20$0

.000

36$0

.000

271

PRID

IST

AGBL

gN

DM

CR/k

Wh

$0.0

2148

$0.0

4837

$0.0

2714

$0.0

0379

$0.0

0388

$0.0

0576

1PR

IDIS

TAG

BLg

YD

MCR

/kW

h$0

.015

08$0

.062

39$0

.025

15$0

.002

33$0

.009

20$0

.004

811

PRID

IST

AGBL

gY

RM

CR/k

Wh

$0.0

0488

$0.0

1973

$0.0

0617

-$0.

0005

9$0

.001

02$0

.001

141

PRID

IST

AGBS

mN

DM

CR/k

Wh

$0.0

2052

$0.0

7058

$0.0

3024

$0.0

0278

$0.0

0354

$0.0

0474

1PR

IDIS

TAG

BSm

YD

MCR

/kW

h$0

.012

51$0

.085

58$0

.031

34$0

.004

45$0

.006

12$0

.000

291

PRID

IST

AGBS

mY

RM

CR/k

Wh

$0.0

0646

$0.0

1914

$0.0

0666

-$0.

0006

6$0

.000

31$0

.000

231

PRID

IST

E1N

DM

CR/k

Wh

$0.0

0291

$0.1

0302

$0.0

3017

$0.0

0049

$0.0

0376

$0.0

0038

1PR

IDIS

TE1

YD

MCR

/kW

h$0

.001

23$0

.120

43$0

.023

60$0

.000

23$0

.002

24$0

.000

211

PRID

IST

E1Y

RM

CR/k

Wh

$0.0

0244

$0.0

3244

$0.0

1188

$0.0

0032

$0.0

0054

$0.0

0014

1PR

IDIS

TE1

9PN

DM

CR/k

Wh

$0.0

0951

$0.0

4713

$0.0

3003

$0.0

0244

$0.0

0302

$0.0

0232

1PR

IDIS

TE1

9PY

DM

CR/k

Wh

$0.0

0781

$0.0

6746

$0.0

3257

$0.0

0130

$0.0

0307

$0.0

0107

1PR

IDIS

TE1

9PY

RM

CR/k

Wh

$0.0

0405

$0.0

2276

$0.0

0994

$0.0

0012

$0.0

0018

$0.0

0049

1PR

IDIS

TE1

9SN

DM

CR/k

Wh

$0.0

1101

$0.0

4775

$0.0

3633

$0.0

0142

$0.0

0238

$0.0

0309

1PR

IDIS

TE1

9SY

DM

CR/k

Wh

$0.0

0465

$0.0

6839

$0.0

3771

$0.0

0103

$0.0

0209

$0.0

0170

1PR

IDIS

TE1

9SY

RM

CR/k

Wh

$0.0

0462

$0.0

2317

$0.0

1296

$0.0

0022

$0.0

0073

$0.0

0030

1PR

IDIS

TE2

0PN

DM

CR/k

Wh

$0.0

0998

$0.0

4405

$0.0

2850

$0.0

0248

$0.0

0303

$0.0

0273

1PR

IDIS

TE2

0PY

DM

CR/k

Wh

$0.0

0939

$0.0

5517

$0.0

3047

$0.0

0108

$0.0

0300

$0.0

0327

1PR

IDIS

TE2

0PY

RM

CR/k

Wh

$0.0

0132

$0.0

4487

$0.0

1105

$0.0

0001

$0.0

0001

$0.0

0006

1PR

IDIS

TE2

0SN

DM

CR/k

Wh

$0.0

1358

$0.0

3586

$0.0

3305

$0.0

0247

$0.0

0289

$0.0

0593

1PR

IDIS

TE2

0SY

DM

CR/k

Wh

$0.0

0690

$0.0

6826

$0.0

3332

$0.0

0052

$0.0

0195

$0.0

0072

1PR

IDIS

TE2

0SY

RM

CR/k

Wh

$0.0

0188

$0.0

0519

$0.0

0461

$0.0

0002

$0.0

0013

$0.0

0002

1PR

IDIS

TE1

9PV

ND

MCR

/kW

h$0

.009

09$0

.039

41$0

.023

42$0

.002

91$0

.004

46$0

.002

861

PRID

IST

E19P

VY

DM

CR/k

Wh

$0.0

0617

$0.0

6714

$0.0

3626

$0.0

0070

$0.0

0551

$0.0

0127

1PR

IDIS

TE1

9PV

YR

MCR

/kW

h$0

.003

69$0

.029

55$0

.012

67$0

.000

77$0

.000

93$0

.000

161

PRID

IST

E19S

VN

DM

CR/k

Wh

$0.0

0619

$0.0

5674

$0.0

2867

$0.0

0093

$0.0

0416

$0.0

0148

1PR

IDIS

TE1

9SV

YD

MCR

/kW

h$0

.005

42$0

.064

31$0

.027

09$0

.000

79$0

.003

20$0

.001

361

PRID

IST

E19S

VY

RM

CR/k

Wh

$0.0

0354

$0.0

1062

$0.0

1046

$0.0

0023

$0.0

0053

$0.0

0059

1PR

IDIS

TNo

nE1

ND

MCR

/kW

h$0

.001

98$0

.097

60$0

.025

96$0

.000

64$0

.004

88$0

.000

441

PRID

IST

NonE

1Y

DM

CR/k

Wh

$0.0

0977

$0.1

2194

$0.0

2300

$0.0

0027

$0.0

0287

$0.0

0023

1PR

IDIS

TNo

nE1

YR

MCR

/kW

h$0

.002

30$0

.031

11$0

.011

93$0

.000

31$0

.000

57$0

.000

141

PRID

IST

STL

ND

MCR

/kW

h$0

.001

04$0

.042

10$0

.011

19$0

.000

48$0

.003

47$0

.000

981

PRID

IST

STL

YD

MCR

/kW

h$0

.000

07$0

.040

50$0

.009

01$0

.000

79$0

.010

84$0

.027

861

PRID

IST

STL

YR

MCR

/kW

h$0

.011

61$0

.003

18$0

.010

68$0

.007

17$0

.003

41$0

.007

581

PRID

IST

TC1

ND

MCR

/kW

h$0

.002

75$0

.054

95$0

.020

34$0

.000

38$0

.002

58$0

.000

621

SECD

IST

A1N

DM

CR/k

Wh

$0.0

0081

1SE

CDIS

TA1

YD

MCR

/kW

h$0

.000

921

SECD

IST

A1Y

RM

CR/k

Wh

-$0.

0006

51

SECD

IST

A10

ND

MCR

/kW

h$0

.000

681

SECD

IST

A10

YD

MCR

/kW

h$0

.000

671

SECD

IST

A10

YR

MCR

/kW

h-$

0.00

069

Page

11

of 2

9

(PG&E-2)

1-AtchA-11

1SE

CDIS

TA6

ND

MCR

/kW

h$0

.001

071

SECD

IST

A6Y

DM

CR/k

Wh

$0.0

0086

1SE

CDIS

TA6

YR

MCR

/kW

h-$

0.00

111

1SE

CDIS

TAG

AN

DM

CR/k

Wh

$0.0

0196

1SE

CDIS

TAG

AY

DM

CR/k

Wh

$0.0

0298

1SE

CDIS

TAG

AY

RM

CR/k

Wh

-$0.

0004

91

SECD

IST

AGBL

gN

DM

CR/k

Wh

$0.0

0124

1SE

CDIS

TAG

BLg

YD

MCR

/kW

h$0

.000

861

SECD

IST

AGBL

gY

RM

CR/k

Wh

-$0.

0005

91

SECD

IST

AGBS

mN

DM

CR/k

Wh

$0.0

0235

1SE

CDIS

TAG

BSm

YD

MCR

/kW

h$0

.001

121

SECD

IST

AGBS

mY

RM

CR/k

Wh

-$0.

0006

71

SECD

IST

E1N

DM

CR/k

Wh

$0.0

0092

1SE

CDIS

TE1

YD

MCR

/kW

h$0

.001

141

SECD

IST

E1Y

RM

CR/k

Wh

-$0.

0001

51

SECD

IST

E19P

ND

MCR

/kW

h$0

.000

541

SECD

IST

E19P

YD

MCR

/kW

h$0

.000

741

SECD

IST

E19P

YR

MCR

/kW

h-$

0.00

046

1SE

CDIS

TE1

9PV

ND

MCR

/kW

h$0

.000

551

SECD

IST

E19P

VY

DM

CR/k

Wh

$0.0

0051

1SE

CDIS

TE1

9PV

YR

MCR

/kW

h-$

0.00

099

1SE

CDIS

TE1

9SN

DM

CR/k

Wh

$0.0

0055

1SE

CDIS

TE1

9SY

DM

CR/k

Wh

$0.0

0059

1SE

CDIS

TE1

9SY

RM

CR/k

Wh

-$0.

0014

71

SECD

IST

E19S

VN

DM

CR/k

Wh

$0.0

0045

1SE

CDIS

TE1

9SV

YD

MCR

/kW

h$0

.000

591

SECD

IST

E19S

VY

RM

CR/k

Wh

-$0.

0008

91

SECD

IST

E20P

ND

MCR

/kW

h$0

.000

571

SECD

IST

E20P

YD

MCR

/kW

h$0

.000

481

SECD

IST

E20P

YR

MCR

/kW

h-$

0.00

056

1SE

CDIS

TE2

0SN

DM

CR/k

Wh

$0.0

0055

1SE

CDIS

TE2

0SY

DM

CR/k

Wh

$0.0

0075

1SE

CDIS

TE2

0SY

RM

CR/k

Wh

-$0.

0006

91

SECD

IST

NonE

1N

DM

CR/k

Wh

$0.0

0130

1SE

CDIS

TNo

nE1

YD

MCR

/kW

h$0

.001

391

SECD

IST

NonE

1Y

RM

CR/k

Wh

-$0.

0000

91

SECD

IST

STBY

PN

DM

CR/k

Wh

$0.0

0341

1SE

CDIS

TST

BYP

YD

MCR

/kW

h$0

.000

001

SECD

IST

STBY

PY

RM

CR/k

Wh

$0.0

0000

1SE

CDIS

TST

BYS

ND

MCR

/kW

h$0

.003

781

SECD

IST

STBY

SY

DM

CR/k

Wh

$0.0

0000

1SE

CDIS

TST

BYS

YR

MCR

/kW

h$0

.000

001

SECD

IST

STL

ND

MCR

/kW

h$0

.000

471

SECD

IST

STL

YD

MCR

/kW

h$0

.000

501

SECD

IST

STL

YR

MCR

/kW

h$0

.000

001

SECD

IST

TC1

ND

MCR

/kW

h$0

.000

291

SECD

IST

TC1

YD

MCR

/kW

h$0

.000

001

SECD

IST

TC1

YR

MCR

/kW

h$0

.000

001

NBDI

STA1

ND

MCR

/kW

h$0

.013

011

NBDI

STA1

YD

MCR

/kW

h$0

.015

071

NBDI

STA1

YR

MCR

/kW

h-$

0.01

032

1NB

DIST

A10

ND

MCR

/kW

h$0

.010

921

NBDI

STA1

0Y

DM

CR/k

Wh

$0.0

1078

1NB

DIST

A10

YR

MCR

/kW

h-$

0.01

140

1NB

DIST

A6N

DM

CR/k

Wh

$0.0

1695

1NB

DIST

A6Y

DM

CR/k

Wh

$0.0

1409

1NB

DIST

A6Y

RM

CR/k

Wh

-$0.

0179

01

NBDI

STAG

AN

DM

CR/k

Wh

$0.0

3010

1NB

DIST

AGA

YD

MCR

/kW

h$0

.045

111

NBDI

STAG

AY

RM

CR/k

Wh

-$0.

0081

01

NBDI

STAG

BLg

ND

MCR

/kW

h$0

.017

811

NBDI

STAG

BLg

YD

MCR

/kW

h$0

.012

811

NBDI

STAG

BLg

YR

MCR

/kW

h-$

0.00

845

1NB

DIST

AGBS

mN

DM

CR/k

Wh

$0.0

3505

1NB

DIST

AGBS

mY

DM

CR/k

Wh

$0.0

1697

1NB

DIST

AGBS

mY

RM

CR/k

Wh

-$0.

0097

1

Page

12

of 2

9

(PG&E-2)

1-AtchA-12

1NB

DIST

E1N

DM

CR/k

Wh

$0.0

1458

1NB

DIST

E1Y

DM

CR/k

Wh

$0.0

1786

1NB

DIST

E1Y

RM

CR/k

Wh

-$0.

0026

81

NBDI

STE1

9PN

DM

CR/k

Wh

$0.0

0816

1NB

DIST

E19P

YD

MCR

/kW

h$0

.011

551

NBDI

STE1

9PY

RM

CR/k

Wh

-$0.

0069

61

NBDI

STE1

9PV

ND

MCR

/kW

h$0

.008

441

NBDI

STE1

9PV

YD

MCR

/kW

h$0

.007

731

NBDI

STE1

9PV

YR

MCR

/kW

h-$

0.01

414

1NB

DIST

E19S

ND

MCR

/kW

h$0

.008

741

NBDI

STE1

9SY

DM

CR/k

Wh

$0.0

0922

1NB

DIST

E19S

YR

MCR

/kW

h-$

0.02

180

1NB

DIST

E19S

VN

DM

CR/k

Wh

$0.0

0719

1NB

DIST

E19S

VY

DM

CR/k

Wh

$0.0

0951

1NB

DIST

E19S

VY

RM

CR/k

Wh

-$0.

0142

91

NBDI

STE2

0PN

DM

CR/k

Wh

$0.0

0878

1NB

DIST

E20P

YD

MCR

/kW

h$0

.007

121

NBDI

STE2

0PY

RM

CR/k

Wh

-$0.

0088

31

NBDI

STE2

0SN

DM

CR/k

Wh

$0.0

0874

1NB

DIST

E20S

YD

MCR

/kW

h$0

.012

221

NBDI

STE2

0SY

RM

CR/k

Wh

-$0.

0094

71

NBDI

STNo

nE1

ND

MCR

/kW

h$0

.021

701

NBDI

STNo

nE1

YD

MCR

/kW

h$0

.023

231

NBDI

STNo

nE1

YR

MCR

/kW

h-$

0.00

145

1NB

DIST

STBY

PN

DM

CR/k

Wh

$0.0

4731

1NB

DIST

STBY

PY

DM

CR/k

Wh

$0.0

0000

1NB

DIST

STBY

PY

RM

CR/k

Wh

$0.0

0000

1NB

DIST

STBY

SN

DM

CR/k

Wh

$0.0

6254

1NB

DIST

STBY

SY

DM

CR/k

Wh

$0.0

0000

1NB

DIST

STBY

SY

RM

CR/k

Wh

$0.0

0000

1NB

DIST

STL

ND

MCR

/kW

h$0

.007

641

NBDI

STST

LY

DM

CR/k

Wh

$0.0

0717

1NB

DIST

STL

YR

MCR

/kW

h$0

.000

001

NBDI

STTC

1N

DM

CR/k

Wh

$0.0

0457

1NB

DIST

TC1

YD

MCR

/kW

h$0

.000

001

NBDI

STTC

1Y

RM

CR/k

Wh

$0.0

0000

2PR

IDIS

TA1

ND

MCR

/kW

h$0

.019

97$0

.071

30$0

.000

00$0

.001

47$0

.003

93$0

.000

002

PRID

IST

A1Y

DM

CR/k

Wh

$0.0

1661

$0.0

8435

$0.0

0000

$0.0

0132

$0.0

0446

$0.0

0000

2PR

IDIS

TA1

YR

MCR

/kW

h$0

.010

41$0

.048

21$0

.000

00$0

.000

72$0

.000

80$0

.000

002

PRID

IST

A10

ND

MCR

/kW

h$0

.020

30$0

.066

26$0

.000

00$0

.001

39$0

.002

98$0

.000

002

PRID

IST

A10

YD

MCR

/kW

h$0

.016

94$0

.074

56$0

.000

00$0

.001

81$0

.004

01$0

.000

002

PRID

IST

A10

YR

MCR

/kW

h$0

.009

06$0

.033

48$0

.000

00$0

.000

59$0

.000

65$0

.000

002

PRID

IST

A6N

DM

CR/k

Wh

$0.0

1429

$0.0

7307

$0.0

0000

$0.0

0124

$0.0

0460

$0.0

0000

2PR

IDIS

TA6

YD

MCR

/kW

h$0

.015

79$0

.090

58$0

.000

00$0

.001

79$0

.003

56$0

.000

002

PRID

IST

A6Y

RM

CR/k

Wh

$0.0

0941

$0.0

4421

$0.0

0000

$0.0

0033

$0.0

0050

$0.0

0000

2PR

IDIS

TAG

AN

DM

CR/k

Wh

$0.0

1646

$0.0

8698

$0.0

0000

$0.0

0123

$0.0

1005

$0.0

0000

2PR

IDIS

TAG

AY

DM

CR/k

Wh

$0.0

1388

$0.0

9470

$0.0

0000

$0.0

0211

$0.0

0267

$0.0

0000

2PR

IDIS

TAG

AY

RM

CR/k

Wh

$0.0

0869

$0.0

5719

$0.0

0000

$0.0

0023

$0.0

0049

$0.0

0000

2PR

IDIS

TAG

BLg

ND

MCR

/kW

h$0

.024

44$0

.053

28$0

.000

00$0

.004

03$0

.004

13$0

.000

002

PRID

IST

AGBL

gY

DM

CR/k

Wh

$0.0

2080

$0.0

6895

$0.0

0000

$0.0

0285

$0.0

1127

$0.0

0000

2PR

IDIS

TAG

BLg

YR

MCR

/kW

h$0

.005

85$0

.029

08$0

.000

00$0

.000

08$0

.001

36$0

.000

002

PRID

IST

AGBS

mN

DM

CR/k

Wh

$0.0

2552

$0.0

7984

$0.0

0000

$0.0

0313

$0.0

0402

$0.0

0000

2PR

IDIS

TAG

BSm

YD

MCR

/kW

h$0

.022

11$0

.094

40$0

.000

00$0

.004

26$0

.007

21$0

.000

002

PRID

IST

AGBS

mY

RM

CR/k

Wh

$0.0

0704

$0.0

2841

$0.0

0000

-$0.

0003

2$0

.000

60$0

.000

002

PRID

IST

E1N

DM

CR/k

Wh

$0.0

2082

$0.1

1910

$0.0

0000

$0.0

0082

$0.0

0405

$0.0

0000

2PR

IDIS

TE1

YD

MCR

/kW

h$0

.020

33$0

.142

88$0

.000

00$0

.000

47$0

.002

39$0

.000

002

PRID

IST

E1Y

RM

CR/k

Wh

$0.0

0622

$0.0

4135

$0.0

0000

$0.0

0027

$0.0

0050

$0.0

0000

2PR

IDIS

TE1

9PN

DM

CR/k

Wh

$0.0

1696

$0.0

4902

$0.0

0000

$0.0

0251

$0.0

0284

$0.0

0000

2PR

IDIS

TE1

9PY

DM

CR/k

Wh

$0.0

1771

$0.0

7377

$0.0

0000

$0.0

0149

$0.0

0297

$0.0

0000

2PR

IDIS

TE1

9PY

RM

CR/k

Wh

$0.0

0679

$0.0

2623

$0.0

0000

$0.0

0023

$0.0

0008

$0.0

0000

2PR

IDIS

TE1

9SN

DM

CR/k

Wh

$0.0

1981

$0.0

4787

$0.0

0000

$0.0

0175

$0.0

0204

$0.0

0000

2PR

IDIS

TE1

9SY

DM

CR/k

Wh

$0.0

1721

$0.0

7588

$0.0

0000

$0.0

0120

$0.0

0210

$0.0

0000

2PR

IDIS

TE1

9SY

RM

CR/k

Wh

$0.0

0764

$0.0

2821

$0.0

0000

$0.0

0026

$0.0

0105

$0.0

0000

2PR

IDIS

TE2

0PN

DM

CR/k

Wh

$0.0

1663

$0.0

4613

$0.0

0000

$0.0

0255

$0.0

0309

$0.0

0000

2PR

IDIS

TE2

0PY

DM

CR/k

Wh

$0.0

1739

$0.0

5923

$0.0

0000

$0.0

0146

$0.0

0304

$0.0

0000

2PR

IDIS

TE2

0PY

RM

CR/k

Wh

$0.0

0699

$0.0

5493

$0.0

0000

$0.0

0003

$0.0

0001

$0.0

0000

Page

13

of 2

9

(PG&E-2)

1-AtchA-13

2PR

IDIS

TE2

0SN

DM

CR/k

Wh

$0.0

1970

$0.0

3494

$0.0

0000

$0.0

0298

$0.0

0217

$0.0

0000

2PR

IDIS

TE2

0SY

DM

CR/k

Wh

$0.0

1741

$0.0

7408

$0.0

0000

$0.0

0068

$0.0

0192

$0.0

0000

2PR

IDIS

TE2

0SY

RM

CR/k

Wh

$0.0

0292

$0.0

0510

$0.0

0000

$0.0

0003

$0.0

0014

$0.0

0000

2PR

IDIS

TE1

9PV

ND

MCR

/kW

h$0

.014

59$0

.041

81$0

.000

00$0

.003

01$0

.004

75$0

.000

002

PRID

IST

E19P

VY

DM

CR/k

Wh

$0.0

1815

$0.0

6954

$0.0

0000

$0.0

0109

$0.0

0648

$0.0

0000

2PR

IDIS

TE1

9PV

YR

MCR

/kW

h$0

.007

85$0

.034

42$0

.000

00$0

.000

52$0

.001

25$0

.000

002

PRID

IST

E19S

VN

DM

CR/k

Wh

$0.0

1553

$0.0

6117

$0.0

0000

$0.0

0119

$0.0

0499

$0.0

0000

2PR

IDIS

TE1

9SV

YD

MCR

/kW

h$0

.015

64$0

.071

10$0

.000

00$0

.001

07$0

.003

43$0

.000

002

PRID

IST

E19S

VY

RM

CR/k

Wh

$0.0

0544

$0.0

0914

$0.0

0000

$0.0

0030

$0.0

0050

$0.0

0000

2PR

IDIS

TNo

nE1

ND

MCR

/kW

h$0

.014

88$0

.113

99$0

.000

00$0

.000

99$0

.005

20$0

.000

002

PRID

IST

NonE

1Y

DM

CR/k

Wh

$0.0

2377

$0.1

4527

$0.0

0000

$0.0

0057

$0.0

0304

$0.0

0000

2PR

IDIS

TNo

nE1

YR

MCR

/kW

h$0

.006

41$0

.038

71$0

.000

00$0

.000

27$0

.000

56$0

.000

002

PRID

IST

STL

ND

MCR

/kW

h$0

.006

17$0

.052

65$0

.000

00$0

.000

74$0

.003

75$0

.000

002

PRID

IST

STL

YD

MCR

/kW

h$0

.004

49$0

.051

89$0

.000

00$0

.001

92$0

.009

51$0

.000

002

PRID

IST

STL

YR

MCR

/kW

h$0

.011

50$0

.000

00$0

.000

00$0

.007

30$0

.000

00$0

.000

002

PRID

IST

TC1

ND

MCR

/kW

h$0

.010

24$0

.062

17$0

.000

00$0

.000

58$0

.002

78$0

.000

003

PRID

IST

A1N

DM

CR/k

Wh

$0.0

0796

$0.0

4387

$0.0

2149

$0.0

0025

$0.0

0045

$0.0

0000

3PR

IDIS

TA1

YD

MCR

/kW

h$0

.007

90$0

.051

85$0

.028

71$0

.000

22$0

.000

58$0

.000

003

PRID

IST

A1Y

RM

CR/k

Wh

$0.0

0514

$0.0

1385

$0.0

0548

$0.0

0001

$0.0

0004

$0.0

0000

3PR

IDIS

TA1

0N

DM

CR/k

Wh

$0.0

0677

$0.0

4582

$0.0

1968

$0.0

0014

$0.0

0025

$0.0

0000

3PR

IDIS

TA1

0Y

DM

CR/k

Wh

$0.0

0795

$0.0

4900

$0.0

2564

$0.0

0018

$0.0

0035

$0.0

0000

3PR

IDIS

TA1

0Y

RM

CR/k

Wh

$0.0

0420

$0.0

1239

$0.0

0426

$0.0

0000

$0.0

0003

$0.0

0000

3PR

IDIS

TA6

ND

MCR

/kW

h$0

.006

78$0

.038

27$0

.023

01$0

.000

44$0

.000

80$0

.000

003

PRID

IST

A6Y

DM

CR/k

Wh

$0.0

0963

$0.0

5127

$0.0

3066

$0.0

0027

$0.0

0207

$0.0

0000

3PR

IDIS

TA6

YR

MCR

/kW

h$0

.003

73$0

.013

85$0

.004

30$0

.000

01$0

.000

36$0

.000

003

PRID

IST

AGA

ND

MCR

/kW

h$0

.012

65$0

.024

44$0

.038

09$0

.001

08$0

.000

45$0

.000

003

PRID

IST

AGA

YD

MCR

/kW

h$0

.012

13$0

.025

65$0

.040

06$0

.000

19$0

.005

88$0

.000

003

PRID

IST

AGA

YR

MCR

/kW

h$0

.008

11$0

.007

56$0

.004

90$0

.000

00$0

.000

02$0

.000

003

PRID

IST

AGBL

gN

DM

CR/k

Wh

$0.0

1403

$0.0

2567

$0.0

3824

$0.0

0356

$0.0

0197

$0.0

0000

3PR

IDIS

TAG

BLg

YD

MCR

/kW

h$0

.011

07$0

.029

45$0

.048

96$0

.000

20$0

.000

58$0

.000

003

PRID

IST

AGBL

gY

RM

CR/k

Wh

$0.0

0375

$0.0

0488

$0.0

0755

-$0.

0000

4$0

.000

19$0

.000

003

PRID

IST

AGBS

mN

DM

CR/k

Wh

$0.0

1956

$0.0

2277

$0.0

4062

$0.0

0261

$0.0

0118

$0.0

0000

3PR

IDIS

TAG

BSm

YD

MCR

/kW

h$0

.017

00$0

.021

43$0

.055

04$0

.000

78$0

.000

04$0

.000

003

PRID

IST

AGBS

mY

RM

CR/k

Wh

$0.0

0442

$0.0

0482

$0.0

0771

-$0.

0000

2$0

.000

05$0

.000

003

PRID

IST

E1N

DM

CR/k

Wh

$0.0

1752

$0.0

5000

$0.0

4069

$0.0

0031

$0.0

0088

$0.0

0000

3PR

IDIS

TE1

YD

MCR

/kW

h$0

.019

43$0

.075

13$0

.062

34$0

.000

11$0

.000

46$0

.000

003

PRID

IST

E1Y

RM

CR/k

Wh

$0.0

0383

$0.0

0743

$0.0

0240

$0.0

0001

$0.0

0001

$0.0

0000

3PR

IDIS

TE1

9PN

DM

CR/k

Wh

$0.0

0454

$0.0

4233

$0.0

2104

$0.0

0163

$0.0

0087

$0.0

0000

3PR

IDIS

TE1

9PY

DM

CR/k

Wh

$0.0

0556

$0.0

5569

$0.0

3056

$0.0

0025

$0.0

0071

$0.0

0000

3PR

IDIS

TE1

9PY

RM

CR/k

Wh

$0.0

0224

$0.0

1066

$0.0

0661

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE1

9SN

DM

CR/k

Wh

$0.0

0409

$0.0

5085

$0.0

1856

$0.0

0007

$0.0

0010

$0.0

0000

3PR

IDIS

TE1

9SY

DM

CR/k

Wh

$0.0

0521

$0.0

6048

$0.0

2574

$0.0

0002

$0.0

0009

$0.0

0000

3PR

IDIS

TE1

9SY

RM

CR/k

Wh

$0.0

0272

$0.0

1107

$0.0

0514

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE2

0PN

DM

CR/k

Wh

$0.0

0541

$0.0

4071

$0.0

2207

$0.0

0188

$0.0

0053

$0.0

0000

3PR

IDIS

TE2

0PY

DM

CR/k

Wh

$0.0

0684

$0.0

4263

$0.0

2617

$0.0

0011

$0.0

0010

$0.0

0000

3PR

IDIS

TE2

0PY

RM

CR/k

Wh

$0.0

0508

$0.0

1081

$0.0

0568

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE2

0SN

DM

CR/k

Wh

$0.0

0366

$0.0

5154

$0.0

2085

$0.0

0008

$0.0

0019

$0.0

0000

3PR

IDIS

TE2

0SY

DM

CR/k

Wh

$0.0

0459

$0.0

4536

$0.0

3036

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE2

0SY

RM

CR/k

Wh

$0.0

0015

$0.0

0448

$0.0

0171

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE1

9PV

ND

MCR

/kW

h$0

.006

62$0

.029

62$0

.021

03$0

.002

37$0

.001

37$0

.000

003

PRID

IST

E19P

VY

DM

CR/k

Wh

$0.0

0636

$0.0

4909

$0.0

3286

$0.0

0004

$0.0

0168

$0.0

0000

3PR

IDIS

TE1

9PV

YR

MCR

/kW

h$0

.004

04$0

.011

16$0

.003

43$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E19S

VN

DM

CR/k

Wh

$0.0

0603

$0.0

4145

$0.0

2145

$0.0

0018

$0.0

0036

$0.0

0000

3PR

IDIS

TE1

9SV

YD

MCR

/kW

h$0

.007

26$0

.046

53$0

.024

61$0

.000

23$0

.000

65$0

.000

003

PRID

IST

E19S

VY

RM

CR/k

Wh

$0.0

0110

$0.0

1127

$0.0

0598

$0.0

0002

$0.0

0007

$0.0

0000

3PR

IDIS

TNo

nE1

ND

MCR

/kW

h$0

.013

98$0

.037

04$0

.031

17$0

.000

49$0

.001

41$0

.000

003

PRID

IST

NonE

1Y

DM

CR/k

Wh

$0.0

2357

$0.0

7711

$0.0

5741

$0.0

0017

$0.0

0069

$0.0

0000

3PR

IDIS

TNo

nE1

YR

MCR

/kW

h$0

.004

53$0

.007

41$0

.002

09$0

.000

01$0

.000

01$0

.000

003

PRID

IST

STL

ND

MCR

/kW

h$0

.002

68$0

.028

22$0

.022

05$0

.000

23$0

.001

18$0

.000

003

PRID

IST

STL

YD

MCR

/kW

h$0

.001

01$0

.030

46$0

.047

52$0

.000

00$0

.000

00$0

.000

003

PRID

IST

STL

YR

MCR

/kW

h$0

.001

60$0

.029

61$0

.028

47$0

.000

00$0

.000

00$0

.000

003

PRID

IST

TC1

ND

MCR

/kW

h$0

.004

55$0

.032

96$0

.018

28$0

.000

11$0

.000

47$0

.000

00

Scen

ario

IDM

argi

nal C

ost

Rate

Cla

ssDG

Indi

cato

rDe

liver

ed/R

ecei

ved

Unit

Sum

mer

Off-

Peak

Sum

mer

Pea

kSu

mm

er P

art-

Peak

Win

ter O

ff-Pe

akW

inte

r Pea

kW

inte

r Sup

er-O

ff-Pe

akAn

nual

1GE

NCAP

A1N

DM

CR/k

Wh

$0.0

0000

$0.1

3785

$0.0

2369

$0.0

0012

$0.0

1732

$0.0

0000

Page

14

of 2

9

(PG&E-2)

1-AtchA-14

1GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

0000

$0.1

8342

$0.0

3830

$0.0

0016

$0.0

2111

$0.0

0000

1GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0833

$0.0

0030

$0.0

0000

$0.0

0020

$0.0

0000

1GE

NCAP

A10

ND

MCR

/kW

h$0

.000

00$0

.141

53$0

.024

84$0

.000

13$0

.017

33$0

.000

001

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

0000

$0.1

7824

$0.0

3828

$0.0

0016

$0.0

1960

$0.0

0000

1GE

NCAP

A10

YR

MCR

/kW

h$0

.000

00$0

.021

87$0

.001

41$0

.000

02$0

.007

47$0

.000

001

GENC

APA6

ND

MCR

/kW

h$0

.000

00$0

.180

47$0

.035

04$0

.000

19$0

.020

31$0

.000

001

GENC

APA6

YD

MCR

/kW

h$0

.000

00$0

.219

40$0

.042

18$0

.000

17$0

.023

42$0

.000

001

GENC

APA6

YR

MCR

/kW

h$0

.000

00$0

.005

84$0

.000

11$0

.000

00$0

.000

09$0

.000

001

GENC

APAG

AN

DM

CR/k

Wh

$0.0

0000

$0.1

4529

$0.0

3101

$0.0

0023

$0.0

1840

$0.0

0000

1GE

NCAP

AGA

YD

MCR

/kW

h$0

.000

00$0

.197

69$0

.044

71$0

.000

26$0

.023

69$0

.000

001

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0988

$0.0

0031

$0.0

0002

$0.0

0051

$0.0

0000

1GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

0000

$0.1

5168

$0.0

3059

$0.0

0020

$0.0

2005

$0.0

0000

1GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.2

0805

$0.0

4490

$0.0

0019

$0.0

2550

$0.0

0000

1GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0000

$0.0

0835

$0.0

0008

$0.0

0000

$0.0

0007

$0.0

0000

1GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.1

4150

$0.0

2856

$0.0

0023

$0.0

1967

$0.0

0000

1GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

0000

$0.2

2542

$0.0

5036

$0.0

0022

$0.0

2702

$0.0

0000

1GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0805

$0.0

0010

$0.0

0000

$0.0

0017

$0.0

0000

1GE

NCAP

E1N

DM

CR/k

Wh

$0.0

0000

$0.1

4584

$0.0

3223

$0.0

0018

$0.0

1752

$0.0

0000

1GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

0000

$0.1

8194

$0.0

4398

$0.0

0024

$0.0

2019

$0.0

0000

1GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0631

$0.0

0015

$0.0

0000

$0.0

0004

$0.0

0000

1GE

NCAP

E19P

ND

MCR

/kW

h$0

.000

00$0

.145

46$0

.027

60$0

.000

12$0

.016

45$0

.000

001

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

0000

$0.2

0169

$0.0

4500

$0.0

0020

$0.0

1901

$0.0

0000

1GE

NCAP

E19P

YR

MCR

/kW

h$0

.000

00$0

.033

14$0

.002

26$0

.000

01$0

.006

87$0

.000

001

GENC

APE1

9PV

ND

MCR

/kW

h$0

.000

00$0

.156

16$0

.030

22$0

.000

12$0

.019

20$0

.000

001

GENC

APE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.182

04$0

.041

19$0

.000

18$0

.019

66$0

.000

001

GENC

APE1

9PV

YR

MCR

/kW

h$0

.000

00$0

.033

40$0

.004

04$0

.000

27$0

.001

94$0

.000

001

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.1

5099

$0.0

2793

$0.0

0013

$0.0

1785

$0.0

0000

1GE

NCAP

E19S

YD

MCR

/kW

h$0

.000

00$0

.186

77$0

.040

20$0

.000

17$0

.021

72$0

.000

001

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

1215

$0.0

0076

$0.0

0001

$0.0

0133

$0.0

0000

1GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

0000

$0.1

5188

$0.0

2968

$0.0

0014

$0.0

1753

$0.0

0000

1GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.1

8200

$0.0

4247

$0.0

0018

$0.0

2008

$0.0

0000

1GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0000

$0.0

9004

$0.0

0797

$0.0

0003

$0.0

0662

$0.0

0000

1GE

NCAP

E19T

ND

MCR

/kW

h$0

.000

00$0

.141

18$0

.027

79$0

.000

08$0

.012

93$0

.000

001

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE1

9TV

ND

MCR

/kW

h$0

.000

00$0

.150

02$0

.027

92$0

.000

12$0

.021

69$0

.000

001

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.1

4743

$0.0

2903

$0.0

0013

$0.0

1593

$0.0

0000

1GE

NCAP

E20P

YD

MCR

/kW

h$0

.000

00$0

.163

33$0

.031

67$0

.000

13$0

.018

99$0

.000

001

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0876

$0.0

0031

$0.0

0000

$0.0

0080

$0.0

0000

1GE

NCAP

E20S

ND

MCR

/kW

h$0

.000

00$0

.147

00$0

.027

16$0

.000

12$0

.017

09$0

.000

001

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.1

9220

$0.0

3840

$0.0

0016

$0.0

2540

$0.0

0000

1GE

NCAP

E20S

YR

MCR

/kW

h$0

.000

00$0

.119

55$0

.006

12$0

.000

02$0

.011

05$0

.000

001

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.1

4598

$0.0

2890

$0.0

0012

$0.0

1579

$0.0

0000

1GE

NCAP

E20T

YD

MCR

/kW

h$0

.000

00$0

.188

64$0

.034

61$0

.000

16$0

.026

28$0

.000

001

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0445

$0.0

0094

$0.0

0000

$0.0

0077

$0.0

0000

1GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

0000

$0.1

5573

$0.0

3426

$0.0

0015

$0.0

1980

$0.0

0000

1GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.1

9019

$0.0

6406

$0.0

0023

$0.0

2324

$0.0

0000

1GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0579

$0.0

0010

$0.0

0000

$0.0

0004

$0.0

0000

1GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.1

5160

$0.0

2890

$0.0

0015

$0.0

1646

$0.0

0000

1GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

0000

$0.2

1996

$0.0

4129

$0.0

0017

$0.0

1336

$0.0

0000

1GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.1

2435

$0.0

2586

$0.0

0015

$0.0

2125

$0.0

0000

1GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STL

ND

MCR

/kW

h$0

.000

00$0

.355

48$0

.055

68$0

.000

24$0

.025

58$0

.000

001

GENC

APST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NCAP

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APTC

1N

DM

CR/k

Wh

$0.0

0000

$0.1

7224

$0.0

3477

$0.0

0014

$0.0

1859

$0.0

0000

1GE

NCAP

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENC

APTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

A1N

DM

CR/k

Wh

$0.0

2777

$0.0

6063

$0.0

4252

$0.0

3078

$0.0

5078

-$0.

0001

5

Page

15

of 2

9

(PG&E-2)

1-AtchA-15

1GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3202

$0.0

6451

$0.0

4900

$0.0

3568

$0.0

5529

$0.0

0078

1GE

NENG

A1Y

RM

CR/k

Wh

$0.0

1689

$0.0

4660

$0.0

2902

$0.0

1126

$0.0

1647

-$0.

0025

21

GENE

NGA1

0N

DM

CR/k

Wh

$0.0

2789

$0.0

6083

$0.0

4274

$0.0

3072

$0.0

5052

-$0.

0000

11

GENE

NGA1

0Y

DM

CR/k

Wh

$0.0

3093

$0.0

6429

$0.0

4890

$0.0

3490

$0.0

5427

$0.0

0022

1GE

NENG

A10

YR

MCR

/kW

h$0

.016

64$0

.046

29$0

.028

43$0

.015

15$0

.027

79-$

0.00

309

1GE

NENG

A6N

DM

CR/k

Wh

$0.0

3006

$0.0

6336

$0.0

4724

$0.0

3358

$0.0

5348

-$0.

0005

61

GENE

NGA6

YD

MCR

/kW

h$0

.033

17$0

.066

18$0

.050

41$0

.036

52$0

.056

64$0

.000

371

GENE

NGA6

YR

MCR

/kW

h$0

.015

68$0

.045

72$0

.028

20$0

.010

16$0

.015

37-$

0.00

310

1GE

NENG

AGA

ND

MCR

/kW

h$0

.028

84$0

.062

37$0

.045

81$0

.031

17$0

.047

96-$

0.00

098

1GE

NENG

AGA

YD

MCR

/kW

h$0

.031

80$0

.066

18$0

.051

51$0

.035

94$0

.054

37-$

0.00

073

1GE

NENG

AGA

YR

MCR

/kW

h$0

.018

62$0

.047

80$0

.030

73$0

.014

31$0

.020

32-$

0.00

157

1GE

NENG

AGBL

gN

DM

CR/k

Wh

$0.0

2875

$0.0

6229

$0.0

4512

$0.0

3100

$0.0

4873

-$0.

0002

71

GENE

NGAG

BLg

YD

MCR

/kW

h$0

.032

74$0

.065

40$0

.050

66$0

.035

55$0

.054

30$0

.000

101

GENE

NGAG

BLg

YR

MCR

/kW

h$0

.016

78$0

.047

37$0

.028

93$0

.013

20$0

.018

54-$

0.00

207

1GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

2764

$0.0

6201

$0.0

4457

$0.0

2904

$0.0

4643

-$0.

0001

91

GENE

NGAG

BSm

YD

MCR

/kW

h$0

.033

82$0

.066

61$0

.052

74$0

.036

40$0

.055

48$0

.000

361

GENE

NGAG

BSm

YR

MCR

/kW

h$0

.017

43$0

.047

34$0

.029

74$0

.013

41$0

.018

82-$

0.00

185

1GE

NENG

E1N

DM

CR/k

Wh

$0.0

2853

$0.0

6211

$0.0

4552

$0.0

3322

$0.0

5334

-$0.

0014

01

GENE

NGE1

YD

MCR

/kW

h$0

.033

53$0

.065

39$0

.051

44$0

.038

41$0

.057

04-$

0.00

102

1GE

NENG

E1Y

RM

CR/k

Wh

$0.0

1831

$0.0

4586

$0.0

2985

$0.0

1248

$0.0

1660

-$0.

0014

81

GENE

NGE1

9PN

DM

CR/k

Wh

$0.0

2791

$0.0

5869

$0.0

4252

$0.0

3096

$0.0

4902

-$0.

0004

21

GENE

NGE1

9PY

DM

CR/k

Wh

$0.0

3130

$0.0

6364

$0.0

4977

$0.0

3484

$0.0

5252

$0.0

0025

1GE

NENG

E19P

YR

MCR

/kW

h$0

.015

95$0

.044

73$0

.026

71$0

.019

78$0

.033

63-$

0.00

280

1GE

NENG

E19P

VN

DM

CR/k

Wh

$0.0

2963

$0.0

6193

$0.0

4508

$0.0

3291

$0.0

5238

-$0.

0006

11

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.031

77$0

.064

28$0

.050

58$0

.036

45$0

.053

93$0

.000

051

GENE

NGE1

9PV

YR

MCR

/kW

h$0

.016

87$0

.051

81$0

.029

37$0

.015

46$0

.028

50-$

0.00

197

1GE

NENG

E19S

ND

MCR

/kW

h$0

.028

94$0

.061

35$0

.043

96$0

.031

81$0

.050

93-$

0.00

011

1GE

NENG

E19S

YD

MCR

/kW

h$0

.031

29$0

.064

74$0

.049

37$0

.035

06$0

.054

06$0

.000

071

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

1579

$0.0

4499

$0.0

2777

$0.0

1395

$0.0

2245

-$0.

0035

11

GENE

NGE1

9SV

ND

MCR

/kW

h$0

.028

89$0

.061

65$0

.044

62$0

.032

23$0

.051

38-$

0.00

074

1GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

3122

$0.0

6457

$0.0

5000

$0.0

3552

$0.0

5431

-$0.

0002

21

GENE

NGE1

9SV

YR

MCR

/kW

h$0

.024

31$0

.055

59$0

.034

70$0

.028

24$0

.037

17-$

0.00

224

1GE

NENG

E19T

ND

MCR

/kW

h$0

.027

92$0

.057

60$0

.042

37$0

.030

72$0

.047

44-$

0.00

120

1GE

NENG

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE1

9TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

E19T

VN

DM

CR/k

Wh

$0.0

2881

$0.0

6039

$0.0

4390

$0.0

3348

$0.0

5302

-$0.

0005

51

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

2822

$0.0

5896

$0.0

4308

$0.0

3111

$0.0

4881

-$0.

0006

01

GENE

NGE2

0PY

DM

CR/k

Wh

$0.0

2920

$0.0

6002

$0.0

4423

$0.0

3180

$0.0

5024

-$0.

0002

61

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

1489

$0.0

4485

$0.0

2690

$0.0

1213

$0.0

2017

-$0.

0029

01

GENE

NGE2

0SN

DM

CR/k

Wh

$0.0

2896

$0.0

6102

$0.0

4382

$0.0

3163

$0.0

5039

-$0.

0000

61

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

3168

$0.0

6400

$0.0

4781

$0.0

3446

$0.0

5354

$0.0

0009

1GE

NENG

E20S

YR

MCR

/kW

h$0

.016

19$0

.053

31$0

.029

14$0

.018

89$0

.038

61-$

0.00

349

1GE

NENG

E20T

ND

MCR

/kW

h$0

.028

03$0

.058

16$0

.042

67$0

.030

82$0

.048

15-$

0.00

074

1GE

NENG

E20T

YD

MCR

/kW

h$0

.029

78$0

.060

30$0

.044

18$0

.031

97$0

.050

58$0

.000

861

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

1254

$0.0

4821

$0.0

2724

$0.0

1727

$0.0

3608

-$0.

0069

51

GENE

NGNo

nE1

ND

MCR

/kW

h$0

.029

08$0

.062

48$0

.046

30$0

.033

41$0

.053

99-$

0.00

131

1GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

3392

$0.0

6602

$0.0

5665

$0.0

3845

$0.0

5745

-$0.

0009

41

GENE

NGNo

nE1

YR

MCR

/kW

h$0

.018

68$0

.046

31$0

.030

37$0

.013

43$0

.017

35-$

0.00

151

1GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

2779

$0.0

5887

$0.0

4457

$0.0

3177

$0.0

5004

-$0.

0010

51

GENE

NGST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYP

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYS

ND

MCR

/kW

h$0

.031

49$0

.064

02$0

.047

13$0

.033

30$0

.052

74$0

.001

001

GENE

NGST

BYS

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYT

ND

MCR

/kW

h$0

.026

96$0

.056

29$0

.041

72$0

.032

53$0

.051

67$0

.000

431

GENE

NGST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

BYT

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGST

LN

DM

CR/k

Wh

$0.0

3658

$0.0

7211

$0.0

5720

$0.0

4082

$0.0

6179

-$0.

0009

91

GENE

NGST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENE

NGTC

1N

DM

CR/k

Wh

$0.0

3086

$0.0

6261

$0.0

4726

$0.0

3455

$0.0

5348

-$0.

0010

61

GENE

NGTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NENG

TC1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

16

of 2

9

(PG&E-2)

1-AtchA-16

1GE

NFLE

XA1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA1

0N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA1

0Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA1

0Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA6

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

A6Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XA6

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGA

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGA

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGA

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBL

gN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BLg

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBL

gY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BSm

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

AGBS

mY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XAG

BSm

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19P

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9PV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9SV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19S

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E19T

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20S

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20S

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20S

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

E20T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

NonE

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XNo

nE1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

NonE

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYP

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYP

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYS

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYT

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1GE

NFLE

XST

BYT

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STL

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STL

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

TC1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

GENF

LEX

TC1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

001

RCS

A10

NX

$/CU

ST-Y

R$1

47.7

5761

Page

17

of 2

9

(PG&E-2)

1-AtchA-17

1RC

SA1

0Y

X$/

CUST

-YR

$698

.442

501

RCS

A11P

NX

$/CU

ST-Y

R$4

1.19

199

1RC

SA1

1PY

X$/

CUST

-YR

$437

.210

651

RCS

A13P

NX

$/CU

ST-Y

R$4

9.27

171

1RC

SA1

3PY

X$/

CUST

-YR

$578

.805

001

RCS

A61P

NX

$/CU

ST-Y

R$4

1.19

199

1RC

SA6

1PY

X$/

CUST

-YR

$437

.210

651

RCS

A63P

NX

$/CU

ST-Y

R$4

9.27

171

1RC

SA6

3PY

X$/

CUST

-YR

$578

.805

001

RCS

AGA

NX

$/CU

ST-Y

R$8

2.59

089

1RC

SAG

AY

X$/

CUST

-YR

$543

.073

921

RCS

AGBL

gN

X$/

CUST

-YR

$201

.463

511

RCS

AGBL

gY

X$/

CUST

-YR

$887

.505

001

RCS

AGBS

mN

X$/

CUST

-YR

$145

.052

631

RCS

AGBS

mY

X$/

CUST

-YR

$832

.915

641

RCS

E1N

X$/

CUST

-YR

$30.

9987

71

RCS

E1Y

X$/

CUST

-YR

$96.

7064

61

RCS

E19P

NX

$/CU

ST-Y

R$1

,346

.398

181

RCS

E19P

YX

$/CU

ST-Y

R$1

,504

.089

941

RCS

E19S

NX

$/CU

ST-Y

R$1

,346

.398

181

RCS

E19S

YX

$/CU

ST-Y

R$1

,504

.089

941

RCS

E19V

PN

X$/

CUST

-YR

$147

.757

611

RCS

E19V

PY

X$/

CUST

-YR

$698

.442

501

RCS

E19V

SN

X$/

CUST

-YR

$147

.757

611

RCS

E19V

SY

X$/

CUST

-YR

$698

.442

501

RCS

E20P

NX

$/CU

ST-Y

R$1

,346

.398

181

RCS

E20P

YX

$/CU

ST-Y

R$1

,504

.089

941

RCS

E20S

NX

$/CU

ST-Y

R$1

,346

.398

181

RCS

E20S

YX

$/CU

ST-Y

R$1

,504

.089

941

RCS

NonE

1N

X$/

CUST

-YR

$45.

3827

81

RCS

NonE

1Y

X$/

CUST

-YR

$118

.230

851

RCS

LS1

NX

$/CU

ST-Y

R$1

3.33

627

1RC

SLS

2N

X$/

CUST

-YR

$7.8

2584

1RC

SLS

3N

X$/

CUST

-YR

$13.

0835

01

RCS

OL1

NX

$/CU

ST-Y

R$1

4.76

191

1RC

STC

1N

X$/

CUST

-YR

$31.

5642

11

MCE

CA1

0PN

X$/

CUST

-YR

$4,1

24.1

6737

1M

CEC

A10P

YX

$/CU

ST-Y

R$4

,124

.167

371

MCE

CA1

0SN

X$/

CUST

-YR

$3,5

32.1

1691

1M

CEC

A10S

YX

$/CU

ST-Y

R$3

,532

.116

911

MCE

CA1

1PN

X$/

CUST

-YR

$369

.430

131

MCE

CA1

1PY

X$/

CUST

-YR

$369

.430

131

MCE

CA1

3PN

X$/

CUST

-YR

$1,3

93.5

7946

1M

CEC

A13P

YX

$/CU

ST-Y

R$1

,393

.579

461

MCE

CA6

1PN

X$/

CUST

-YR

$369

.430

131

MCE

CA6

1PY

X$/

CUST

-YR

$369

.430

131

MCE

CA6

3PN

X$/

CUST

-YR

$1,3

93.5

7946

1M

CEC

A63P

YX

$/CU

ST-Y

R$1

,393

.579

461

MCE

CAG

AN

X$/

CUST

-YR

$743

.001

291

MCE

CAG

AY

X$/

CUST

-YR

$743

.001

291

MCE

CAG

BLg

NX

$/CU

ST-Y

R$2

,476

.625

601

MCE

CAG

BLg

YX

$/CU

ST-Y

R$2

,476

.625

601

MCE

CAG

BSm

NX

$/CU

ST-Y

R$2

,476

.625

601

MCE

CAG

BSm

YX

$/CU

ST-Y

R$2

,476

.625

601

MCE

CE1

NX

$/CU

ST-Y

R$1

03.1

0785

1M

CEC

E1Y

X$/

CUST

-YR

$103

.107

851

MCE

CE1

9PN

X$/

CUST

-YR

$4,2

92.7

5541

1M

CEC

E19P

YX

$/CU

ST-Y

R$4

,292

.755

411

MCE

CE1

9SN

X$/

CUST

-YR

$5,3

10.0

3825

1M

CEC

E19S

YX

$/CU

ST-Y

R$5

,310

.038

251

MCE

CE1

9VP

NX

$/CU

ST-Y

R$4

,124

.167

371

MCE

CE1

9VP

YX

$/CU

ST-Y

R$4

,124

.167

371

MCE

CE1

9VS

NX

$/CU

ST-Y

R$3

,532

.116

911

MCE

CE1

9VS

YX

$/CU

ST-Y

R$3

,532

.116

911

MCE

CE2

0PN

X$/

CUST

-YR

$4,2

92.7

5541

1M

CEC

E20P

YX

$/CU

ST-Y

R$4

,292

.755

41

Page

18

of 2

9

(PG&E-2)

1-AtchA-18

1M

CEC

E20S

NX

$/CU

ST-Y

R$5

,751

.049

091

MCE

CE2

0SY

X$/

CUST

-YR

$5,7

51.0

4909

1M

CEC

NonE

1N

X$/

CUST

-YR

$103

.107

851

MCE

CNo

nE1

YX

$/CU

ST-Y

R$1

03.1

0785

1M

CEC

LS1

NX

$/CU

ST-Y

R$3

1.40

362

1M

CEC

LS2

NX

$/CU

ST-Y

R$1

7.84

780

1M

CEC

LS3

NX

$/CU

ST-Y

R$5

2.57

811

1M

CEC

TC1

NX

$/CU

ST-Y

R$4

43.2

4651

2GE

NCAP

A1N

DM

CR/k

Wh

$0.0

2485

$0.0

8754

$0.0

0000

$0.0

0078

$0.0

1807

$0.0

0000

2GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

3563

$0.1

2159

$0.0

0000

$0.0

0106

$0.0

2284

$0.0

0000

2GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0026

$0.0

0709

$0.0

0000

$0.0

0000

$0.0

0016

$0.0

0000

2GE

NCAP

A10

ND

MCR

/kW

h$0

.025

03$0

.090

16$0

.000

00$0

.000

79$0

.017

99$0

.000

002

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

3331

$0.1

1845

$0.0

0000

$0.0

0102

$0.0

2097

$0.0

0000

2GE

NCAP

A10

YR

MCR

/kW

h$0

.001

47$0

.013

31$0

.000

00$0

.000

11$0

.007

00$0

.000

002

GENC

APA6

ND

MCR

/kW

h$0

.030

08$0

.113

72$0

.000

00$0

.001

08$0

.021

83$0

.000

002

GENC

APA6

YD

MCR

/kW

h$0

.036

55$0

.155

69$0

.000

00$0

.001

19$0

.025

69$0

.000

002

GENC

APA6

YR

MCR

/kW

h$0

.000

07$0

.005

61$0

.000

00$0

.000

00$0

.000

07$0

.000

002

GENC

APAG

AN

DM

CR/k

Wh

$0.0

2459

$0.0

8776

$0.0

0000

$0.0

0116

$0.0

1809

$0.0

0000

2GE

NCAP

AGA

YD

MCR

/kW

h$0

.034

82$0

.132

55$0

.000

00$0

.001

53$0

.025

12$0

.000

002

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0027

$0.0

0883

$0.0

0000

$0.0

0005

$0.0

0014

$0.0

0000

2GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

2556

$0.0

9286

$0.0

0000

$0.0

0108

$0.0

2008

$0.0

0000

2GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

3747

$0.1

4154

$0.0

0000

$0.0

0138

$0.0

2721

$0.0

0000

2GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0004

$0.0

0827

$0.0

0000

$0.0

0000

$0.0

0006

$0.0

0000

2GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

2266

$0.0

8501

$0.0

0000

$0.0

0110

$0.0

1888

$0.0

0000

2GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

4130

$0.1

5839

$0.0

0000

$0.0

0159

$0.0

2897

$0.0

0000

2GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0004

$0.0

0796

$0.0

0000

$0.0

0000

$0.0

0016

$0.0

0000

2GE

NCAP

E1N

DM

CR/k

Wh

$0.0

3874

$0.0

9059

$0.0

0000

$0.0

0106

$0.0

1889

$0.0

0000

2GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

5518

$0.1

1836

$0.0

0000

$0.0

0151

$0.0

2241

$0.0

0000

2GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0011

$0.0

0560

$0.0

0000

$0.0

0000

$0.0

0003

$0.0

0000

2GE

NCAP

E19P

ND

MCR

/kW

h$0

.023

53$0

.092

96$0

.000

00$0

.000

75$0

.017

28$0

.000

002

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

3627

$0.1

3524

$0.0

0000

$0.0

0113

$0.0

2045

$0.0

0000

2GE

NCAP

E19P

YR

MCR

/kW

h$0

.002

78$0

.020

74$0

.000

00$0

.000

12$0

.006

84$0

.000

002

GENC

APE1

9PV

ND

MCR

/kW

h$0

.025

87$0

.098

61$0

.000

00$0

.000

77$0

.020

62$0

.000

002

GENC

APE1

9PV

YD

MCR

/kW

h$0

.032

99$0

.122

65$0

.000

00$0

.001

10$0

.021

09$0

.000

002

GENC

APE1

9PV

YR

MCR

/kW

h$0

.004

06$0

.014

15$0

.000

00$0

.000

36$0

.000

52$0

.000

002

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

2510

$0.0

9650

$0.0

0000

$0.0

0080

$0.0

1859

$0.0

0000

2GE

NCAP

E19S

YD

MCR

/kW

h$0

.033

67$0

.124

14$0

.000

00$0

.001

11$0

.023

28$0

.000

002

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0085

$0.0

0697

$0.0

0000

$0.0

0004

$0.0

0108

$0.0

0000

2GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

2744

$0.0

9643

$0.0

0000

$0.0

0086

$0.0

1841

$0.0

0000

2GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

3722

$0.1

1929

$0.0

0000

$0.0

0118

$0.0

2153

$0.0

0000

2GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0984

$0.0

5406

$0.0

0000

$0.0

0015

$0.0

0634

$0.0

0000

2GE

NCAP

E19T

ND

MCR

/kW

h$0

.021

70$0

.087

88$0

.000

00$0

.000

57$0

.013

74$0

.000

002

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE1

9TV

ND

MCR

/kW

h$0

.023

31$0

.097

60$0

.000

00$0

.000

86$0

.023

40$0

.000

002

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

2456

$0.0

9403

$0.0

0000

$0.0

0075

$0.0

1670

$0.0

0000

2GE

NCAP

E20P

YD

MCR

/kW

h$0

.026

89$0

.106

92$0

.000

00$0

.000

85$0

.020

25$0

.000

002

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0024

$0.0

0816

$0.0

0000

$0.0

0002

$0.0

0073

$0.0

0000

2GE

NCAP

E20S

ND

MCR

/kW

h$0

.023

84$0

.093

31$0

.000

00$0

.000

75$0

.017

72$0

.000

002

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

3516

$0.1

2978

$0.0

0000

$0.0

0129

$0.0

2717

$0.0

0000

2GE

NCAP

E20S

YR

MCR

/kW

h$0

.010

17$0

.092

96$0

.000

00$0

.000

11$0

.011

26$0

.000

002

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

2351

$0.0

9336

$0.0

0000

$0.0

0072

$0.0

1663

$0.0

0000

2GE

NCAP

E20T

YD

MCR

/kW

h$0

.029

67$0

.127

34$0

.000

00$0

.001

29$0

.027

66$0

.000

002

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0100

$0.0

0040

$0.0

0000

$0.0

0001

$0.0

0080

$0.0

0000

2GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

3417

$0.0

9751

$0.0

0000

$0.0

0096

$0.0

2170

$0.0

0000

2GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

5951

$0.1

2527

$0.0

0000

$0.0

0152

$0.0

2621

$0.0

0000

2GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0003

$0.0

0572

$0.0

0000

$0.0

0000

$0.0

0004

$0.0

0000

2GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

1951

$0.0

9943

$0.0

0000

$0.0

0077

$0.0

1768

$0.0

0000

2GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

3339

$0.1

4945

$0.0

0000

$0.0

0062

$0.0

1426

$0.0

0000

2GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

2076

$0.0

7507

$0.0

0000

$0.0

0073

$0.0

2346

$0.0

0000

Page

19

of 2

9

(PG&E-2)

1-AtchA-19

2GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STL

ND

MCR

/kW

h$0

.040

32$0

.288

61$0

.000

00$0

.001

34$0

.031

78$0

.000

002

GENC

APST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NCAP

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APTC

1N

DM

CR/k

Wh

$0.0

2792

$0.1

0752

$0.0

0000

$0.0

0086

$0.0

2022

$0.0

0000

2GE

NCAP

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENC

APTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

A1N

DM

CR/k

Wh

$0.0

3377

$0.0

5710

$0.0

0000

$0.0

2838

$0.0

4857

$0.0

0000

2GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3892

$0.0

6036

$0.0

0000

$0.0

3518

$0.0

5379

$0.0

0000

2GE

NENG

A1Y

RM

CR/k

Wh

$0.0

2019

$0.0

4648

$0.0

0000

$0.0

0617

$0.0

1631

$0.0

0000

2GE

NENG

A10

ND

MCR

/kW

h$0

.033

78$0

.057

26$0

.000

00$0

.028

22$0

.048

24$0

.000

002

GENE

NGA1

0Y

DM

CR/k

Wh

$0.0

3763

$0.0

6038

$0.0

0000

$0.0

3385

$0.0

5254

$0.0

0000

2GE

NENG

A10

YR

MCR

/kW

h$0

.019

90$0

.045

47$0

.000

00$0

.009

38$0

.025

22$0

.000

002

GENE

NGA6

ND

MCR

/kW

h$0

.036

51$0

.058

69$0

.000

00$0

.031

70$0

.051

25$0

.000

002

GENE

NGA6

YD

MCR

/kW

h$0

.039

86$0

.061

80$0

.000

00$0

.036

35$0

.055

34$0

.000

002

GENE

NGA6

YR

MCR

/kW

h$0

.019

22$0

.045

70$0

.000

00$0

.005

17$0

.015

30$0

.000

002

GENE

NGAG

AN

DM

CR/k

Wh

$0.0

3472

$0.0

5791

$0.0

0000

$0.0

2787

$0.0

4439

$0.0

0000

2GE

NENG

AGA

YD

MCR

/kW

h$0

.038

76$0

.061

68$0

.000

00$0

.035

49$0

.052

01$0

.000

002

GENE

NGAG

AY

RM

CR/k

Wh

$0.0

2203

$0.0

4770

$0.0

0000

$0.0

0931

$0.0

1980

$0.0

0000

2GE

NENG

AGBL

gN

DM

CR/k

Wh

$0.0

3457

$0.0

5810

$0.0

0000

$0.0

2766

$0.0

4562

$0.0

0000

2GE

NENG

AGBL

gY

DM

CR/k

Wh

$0.0

3944

$0.0

6132

$0.0

0000

$0.0

3503

$0.0

5243

$0.0

0000

2GE

NENG

AGBL

gY

RM

CR/k

Wh

$0.0

2002

$0.0

4736

$0.0

0000

$0.0

0798

$0.0

1851

$0.0

0000

2GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

3330

$0.0

5770

$0.0

0000

$0.0

2450

$0.0

4271

$0.0

0000

2GE

NENG

AGBS

mY

DM

CR/k

Wh

$0.0

4091

$0.0

6252

$0.0

0000

$0.0

3658

$0.0

5378

$0.0

0000

2GE

NENG

AGBS

mY

RM

CR/k

Wh

$0.0

2074

$0.0

4733

$0.0

0000

$0.0

0832

$0.0

1880

$0.0

0000

2GE

NENG

E1N

DM

CR/k

Wh

$0.0

3720

$0.0

5820

$0.0

0000

$0.0

3205

$0.0

5130

$0.0

0000

2GE

NENG

E1Y

DM

CR/k

Wh

$0.0

4376

$0.0

6110

$0.0

0000

$0.0

3928

$0.0

5578

$0.0

0000

2GE

NENG

E1Y

RM

CR/k

Wh

$0.0

2117

$0.0

4576

$0.0

0000

$0.0

0725

$0.0

1659

$0.0

0000

2GE

NENG

E19P

ND

MCR

/kW

h$0

.033

17$0

.055

00$0

.000

00$0

.028

97$0

.046

83$0

.000

002

GENE

NGE1

9PY

DM

CR/k

Wh

$0.0

3808

$0.0

5937

$0.0

0000

$0.0

3409

$0.0

5079

$0.0

0000

2GE

NENG

E19P

YR

MCR

/kW

h$0

.018

91$0

.043

96$0

.000

00$0

.014

47$0

.030

67$0

.000

002

GENE

NGE1

9PV

ND

MCR

/kW

h$0

.035

27$0

.058

01$0

.000

00$0

.031

20$0

.050

36$0

.000

002

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.038

44$0

.060

24$0

.000

00$0

.035

72$0

.052

05$0

.000

002

GENE

NGE1

9PV

YR

MCR

/kW

h$0

.020

28$0

.048

42$0

.000

00$0

.008

09$0

.025

78$0

.000

002

GENE

NGE1

9SN

DM

CR/k

Wh

$0.0

3450

$0.0

5764

$0.0

0000

$0.0

2956

$0.0

4866

$0.0

0000

2GE

NENG

E19S

YD

MCR

/kW

h$0

.037

88$0

.060

76$0

.000

00$0

.033

97$0

.052

27$0

.000

002

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

1908

$0.0

4427

$0.0

0000

$0.0

0844

$0.0

2021

$0.0

0000

2GE

NENG

E19S

VN

DM

CR/k

Wh

$0.0

3502

$0.0

5775

$0.0

0000

$0.0

3024

$0.0

4907

$0.0

0000

2GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

3853

$0.0

6047

$0.0

0000

$0.0

3481

$0.0

5248

$0.0

0000

2GE

NENG

E19S

VY

RM

CR/k

Wh

$0.0

2810

$0.0

5227

$0.0

0000

$0.0

2293

$0.0

3357

$0.0

0000

2GE

NENG

E19T

ND

MCR

/kW

h$0

.032

76$0

.053

70$0

.000

00$0

.028

71$0

.045

07$0

.000

002

GENE

NGE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE1

9TV

ND

MCR

/kW

h$0

.034

19$0

.056

41$0

.000

00$0

.031

99$0

.051

18$0

.000

002

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

3362

$0.0

5520

$0.0

0000

$0.0

2917

$0.0

4655

$0.0

0000

2GE

NENG

E20P

YD

MCR

/kW

h$0

.034

69$0

.056

28$0

.000

00$0

.030

45$0

.048

32$0

.000

002

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

1818

$0.0

4478

$0.0

0000

$0.0

0683

$0.0

1912

$0.0

0000

2GE

NENG

E20S

ND

MCR

/kW

h$0

.034

32$0

.057

33$0

.000

00$0

.029

24$0

.048

07$0

.000

002

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

3806

$0.0

6017

$0.0

0000

$0.0

3376

$0.0

5170

$0.0

0000

2GE

NENG

E20S

YR

MCR

/kW

h$0

.020

58$0

.051

94$0

.000

00$0

.012

57$0

.037

07$0

.000

002

GENE

NGE2

0TN

DM

CR/k

Wh

$0.0

3322

$0.0

5443

$0.0

0000

$0.0

2904

$0.0

4596

$0.0

0000

2GE

NENG

E20T

YD

MCR

/kW

h$0

.035

05$0

.056

90$0

.000

00$0

.031

26$0

.048

88$0

.000

002

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

1612

$0.0

4653

$0.0

0000

$0.0

1229

$0.0

3340

$0.0

0000

2GE

NENG

NonE

1N

DM

CR/k

Wh

$0.0

3662

$0.0

5843

$0.0

0000

$0.0

3216

$0.0

5218

$0.0

0000

2GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

4412

$0.0

6184

$0.0

0000

$0.0

3940

$0.0

5645

$0.0

0000

2GE

NENG

NonE

1Y

RM

CR/k

Wh

$0.0

2174

$0.0

4630

$0.0

0000

$0.0

0837

$0.0

1734

$0.0

0000

2GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

3289

$0.0

5387

$0.0

0000

$0.0

2949

$0.0

4787

$0.0

0000

2GE

NENG

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

SN

DM

CR/k

Wh

$0.0

3692

$0.0

6039

$0.0

0000

$0.0

3135

$0.0

5072

$0.0

0000

2GE

NENG

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

TN

DM

CR/k

Wh

$0.0

3211

$0.0

5226

$0.0

0000

$0.0

3116

$0.0

5048

$0.0

0000

Page

20

of 2

9

(PG&E-2)

1-AtchA-20

2GE

NENG

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STL

ND

MCR

/kW

h$0

.043

58$0

.065

78$0

.000

00$0

.041

78$0

.062

23$0

.000

002

GENE

NGST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NENG

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGTC

1N

DM

CR/k

Wh

$0.0

3672

$0.0

5812

$0.0

0000

$0.0

3318

$0.0

5140

$0.0

0000

2GE

NENG

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENE

NGTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A10

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A6N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XA6

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

A6Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BLg

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BLg

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XAG

BSm

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19P

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9SV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

E19T

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XNo

nE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XNo

nE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYS

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

21

of 2

9

(PG&E-2)

1-AtchA-21

2GE

NFLE

XST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

002

GENF

LEX

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

2GE

NFLE

XTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

A1N

DM

CR/k

Wh

$0.0

1926

$0.0

0085

$0.0

8066

$0.0

0103

$0.0

0494

$0.0

0000

3GE

NCAP

A1Y

DM

CR/k

Wh

$0.0

2237

$0.0

0107

$0.1

2011

$0.0

0114

$0.0

0784

$0.0

0000

3GE

NCAP

A1Y

RM

CR/k

Wh

$0.0

0092

$0.0

0035

$0.0

0133

$0.0

0001

$0.0

0002

$0.0

0000

3GE

NCAP

A10

ND

MCR

/kW

h$0

.019

05$0

.000

84$0

.081

23$0

.001

00$0

.004

88$0

.000

003

GENC

APA1

0Y

DM

CR/k

Wh

$0.0

2200

$0.0

0108

$0.1

0390

$0.0

0111

$0.0

0699

$0.0

0000

3GE

NCAP

A10

YR

MCR

/kW

h$0

.002

47$0

.000

26$0

.005

22$0

.000

34$0

.000

58$0

.000

003

GENC

APA6

ND

MCR

/kW

h$0

.019

25$0

.000

79$0

.095

71$0

.001

07$0

.006

40$0

.000

003

GENC

APA6

YD

MCR

/kW

h$0

.022

13$0

.000

59$0

.131

53$0

.001

15$0

.008

87$0

.000

003

GENC

APA6

YR

MCR

/kW

h$0

.000

67$0

.000

40$0

.000

80$0

.000

00$0

.000

01$0

.000

003

GENC

APAG

AN

DM

CR/k

Wh

$0.0

1687

$0.0

0088

$0.0

7828

$0.0

0086

$0.0

0551

$0.0

0000

3GE

NCAP

AGA

YD

MCR

/kW

h$0

.022

40$0

.001

07$0

.117

06$0

.001

07$0

.009

64$0

.000

003

GENC

APAG

AY

RM

CR/k

Wh

$0.0

0110

$0.0

0040

$0.0

0201

$0.0

0000

$0.0

0005

$0.0

0000

3GE

NCAP

AGBL

gN

DM

CR/k

Wh

$0.0

1765

$0.0

0094

$0.0

8315

$0.0

0090

$0.0

0524

$0.0

0000

3GE

NCAP

AGBL

gY

DM

CR/k

Wh

$0.0

2271

$0.0

0101

$0.1

2647

$0.0

0120

$0.0

0862

$0.0

0000

3GE

NCAP

AGBL

gY

RM

CR/k

Wh

$0.0

0097

$0.0

0048

$0.0

0158

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

AGBS

mN

DM

CR/k

Wh

$0.0

1661

$0.0

0086

$0.0

7042

$0.0

0079

$0.0

0463

$0.0

0000

3GE

NCAP

AGBS

mY

DM

CR/k

Wh

$0.0

2428

$0.0

0098

$0.1

3797

$0.0

0121

$0.0

0936

$0.0

0000

3GE

NCAP

AGBS

mY

RM

CR/k

Wh

$0.0

0097

$0.0

0048

$0.0

0148

$0.0

0001

$0.0

0001

$0.0

0000

3GE

NCAP

E1N

DM

CR/k

Wh

$0.0

2786

$0.0

0109

$0.1

0351

$0.0

0124

$0.0

0628

$0.0

0000

3GE

NCAP

E1Y

DM

CR/k

Wh

$0.0

3487

$0.0

0104

$0.1

4042

$0.0

0143

$0.0

0953

$0.0

0000

3GE

NCAP

E1Y

RM

CR/k

Wh

$0.0

0045

$0.0

0026

$0.0

0060

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

E19P

ND

MCR

/kW

h$0

.015

82$0

.000

76$0

.079

27$0

.000

81$0

.004

64$0

.000

003

GENC

APE1

9PY

DM

CR/k

Wh

$0.0

2132

$0.0

0115

$0.1

1832

$0.0

0098

$0.0

0725

$0.0

0000

3GE

NCAP

E19P

YR

MCR

/kW

h$0

.003

52$0

.000

44$0

.010

75$0

.000

34$0

.000

88$0

.000

003

GENC

APE1

9PV

ND

MCR

/kW

h$0

.016

98$0

.000

83$0

.085

91$0

.001

00$0

.005

77$0

.000

003

GENC

APE1

9PV

YD

MCR

/kW

h$0

.020

41$0

.001

06$0

.108

75$0

.001

08$0

.007

44$0

.000

003

GENC

APE1

9PV

YR

MCR

/kW

h$0

.001

52$0

.000

58$0

.009

85$0

.000

01$0

.000

98$0

.000

003

GENC

APE1

9SN

DM

CR/k

Wh

$0.0

1776

$0.0

0086

$0.0

8483

$0.0

0094

$0.0

0504

$0.0

0000

3GE

NCAP

E19S

YD

MCR

/kW

h$0

.021

84$0

.001

07$0

.108

28$0

.001

23$0

.007

86$0

.000

003

GENC

APE1

9SY

RM

CR/k

Wh

$0.0

0087

$0.0

0019

$0.0

0282

$0.0

0003

$0.0

0013

$0.0

0000

3GE

NCAP

E19S

VN

DM

CR/k

Wh

$0.0

1901

$0.0

0084

$0.0

8708

$0.0

0097

$0.0

0518

$0.0

0000

3GE

NCAP

E19S

VY

DM

CR/k

Wh

$0.0

2383

$0.0

0111

$0.1

0961

$0.0

0117

$0.0

0735

$0.0

0000

3GE

NCAP

E19S

VY

RM

CR/k

Wh

$0.0

0757

$0.0

0038

$0.0

3402

$0.0

0011

$0.0

0076

$0.0

0000

3GE

NCAP

E19T

ND

MCR

/kW

h$0

.014

44$0

.000

79$0

.073

81$0

.000

67$0

.003

69$0

.000

003

GENC

APE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE1

9TV

ND

MCR

/kW

h$0

.016

20$0

.000

71$0

.084

34$0

.001

11$0

.006

31$0

.000

003

GENC

APE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APE2

0PN

DM

CR/k

Wh

$0.0

1626

$0.0

0079

$0.0

8216

$0.0

0081

$0.0

0463

$0.0

0000

3GE

NCAP

E20P

YD

MCR

/kW

h$0

.017

88$0

.000

80$0

.093

38$0

.000

97$0

.005

77$0

.000

003

GENC

APE2

0PY

RM

CR/k

Wh

$0.0

0126

$0.0

0056

$0.0

0237

$0.0

0007

$0.0

0013

$0.0

0000

3GE

NCAP

E20S

ND

MCR

/kW

h$0

.016

60$0

.000

88$0

.081

90$0

.000

91$0

.004

88$0

.000

003

GENC

APE2

0SY

DM

CR/k

Wh

$0.0

2332

$0.0

0092

$0.1

1929

$0.0

0136

$0.0

0836

$0.0

0000

3GE

NCAP

E20S

YR

MCR

/kW

h$0

.020

13$0

.000

19$0

.036

44$0

.000

65$0

.001

31$0

.000

003

GENC

APE2

0TN

DM

CR/k

Wh

$0.0

1554

$0.0

0077

$0.0

8076

$0.0

0075

$0.0

0448

$0.0

0000

3GE

NCAP

E20T

YD

MCR

/kW

h$0

.021

79$0

.000

77$0

.114

00$0

.001

04$0

.007

35$0

.000

003

GENC

APE2

0TY

RM

CR/k

Wh

$0.0

0058

$0.0

0000

$0.0

0167

$0.0

0028

$0.0

0003

$0.0

0000

3GE

NCAP

NonE

1N

DM

CR/k

Wh

$0.0

2305

$0.0

0091

$0.0

9687

$0.0

0124

$0.0

0670

$0.0

0000

3GE

NCAP

NonE

1Y

DM

CR/k

Wh

$0.0

3387

$0.0

0337

$0.1

4515

$0.0

0165

$0.0

1108

$0.0

0000

3GE

NCAP

NonE

1Y

RM

CR/k

Wh

$0.0

0047

$0.0

0029

$0.0

0054

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

PN

DM

CR/k

Wh

$0.0

1146

$0.0

0030

$0.0

6727

$0.0

0069

$0.0

0530

$0.0

0000

3GE

NCAP

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

SN

DM

CR/k

Wh

$0.0

1898

$0.0

0040

$0.1

2066

$0.0

0062

$0.0

0395

$0.0

0000

3GE

NCAP

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

TN

DM

CR/k

Wh

$0.0

1428

$0.0

0056

$0.0

7924

$0.0

0116

$0.0

0828

$0.0

0000

Page

22

of 2

9

(PG&E-2)

1-AtchA-22

3GE

NCAP

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STL

ND

MCR

/kW

h$0

.017

97$0

.000

86$0

.155

16$0

.001

07$0

.011

95$0

.000

003

GENC

APST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NCAP

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APTC

1N

DM

CR/k

Wh

$0.0

1654

$0.0

0079

$0.0

9278

$0.0

0092

$0.0

0597

$0.0

0000

3GE

NCAP

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENC

APTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

A1N

DM

CR/k

Wh

$0.0

3672

$0.0

2715

$0.0

4064

$0.0

3671

$0.0

2900

$0.0

0000

3GE

NENG

A1Y

DM

CR/k

Wh

$0.0

3943

$0.0

3089

$0.0

5207

$0.0

4015

$0.0

3888

$0.0

0000

3GE

NENG

A1Y

RM

CR/k

Wh

$0.0

2275

$0.0

1822

$0.0

1030

$0.0

1174

$0.0

0538

$0.0

0000

3GE

NENG

A10

ND

MCR

/kW

h$0

.036

50$0

.026

91$0

.040

52$0

.036

49$0

.028

72$0

.000

003

GENE

NGA1

0Y

DM

CR/k

Wh

$0.0

3875

$0.0

2945

$0.0

4812

$0.0

3996

$0.0

3663

$0.0

0000

3GE

NENG

A10

YR

MCR

/kW

h$0

.024

43$0

.016

52$0

.010

45$0

.020

16$0

.007

05$0

.000

003

GENE

NGA6

ND

MCR

/kW

h$0

.038

00$0

.025

98$0

.044

35$0

.038

70$0

.032

95$0

.000

003

GENE

NGA6

YD

MCR

/kW

h$0

.039

49$0

.029

25$0

.054

48$0

.040

58$0

.041

37$0

.000

003

GENE

NGA6

YR

MCR

/kW

h$0

.022

43$0

.017

99$0

.009

28$0

.010

99$0

.004

60$0

.000

003

GENE

NGAG

AN

DM

CR/k

Wh

$0.0

3679

$0.0

2532

$0.0

4144

$0.0

3617

$0.0

2712

$0.0

0000

3GE

NENG

AGA

YD

MCR

/kW

h$0

.039

22$0

.028

79$0

.051

21$0

.040

49$0

.041

55$0

.000

003

GENE

NGAG

AY

RM

CR/k

Wh

$0.0

2404

$0.0

2028

$0.0

1279

$0.0

1476

$0.0

0816

$0.0

0000

3GE

NENG

AGBL

gN

DM

CR/k

Wh

$0.0

3670

$0.0

2623

$0.0

4094

$0.0

3570

$0.0

2639

$0.0

0000

3GE

NENG

AGBL

gY

DM

CR/k

Wh

$0.0

3966

$0.0

2932

$0.0

5168

$0.0

4006

$0.0

3815

$0.0

0000

3GE

NENG

AGBL

gY

RM

CR/k

Wh

$0.0

2303

$0.0

1942

$0.0

1162

$0.0

1442

$0.0

0720

$0.0

0000

3GE

NENG

AGBS

mN

DM

CR/k

Wh

$0.0

3565

$0.0

2541

$0.0

3794

$0.0

3316

$0.0

2218

$0.0

0000

3GE

NENG

AGBS

mY

DM

CR/k

Wh

$0.0

4036

$0.0

3013

$0.0

5488

$0.0

4090

$0.0

4146

$0.0

0000

3GE

NENG

AGBS

mY

RM

CR/k

Wh

$0.0

2331

$0.0

2032

$0.0

1208

$0.0

1438

$0.0

0767

$0.0

0000

3GE

NENG

E1N

DM

CR/k

Wh

$0.0

3993

$0.0

2848

$0.0

4889

$0.0

3977

$0.0

3447

$0.0

0000

3GE

NENG

E1Y

DM

CR/k

Wh

$0.0

4311

$0.0

3451

$0.0

6024

$0.0

4300

$0.0

4590

$0.0

0000

3GE

NENG

E1Y

RM

CR/k

Wh

$0.0

2182

$0.0

1821

$0.0

1106

$0.0

1190

$0.0

0640

$0.0

0000

3GE

NENG

E19P

ND

MCR

/kW

h$0

.035

17$0

.024

82$0

.039

65$0

.035

99$0

.028

63$0

.000

003

GENE

NGE1

9PY

DM

CR/k

Wh

$0.0

3774

$0.0

2814

$0.0

5038

$0.0

3873

$0.0

3744

$0.0

0000

3GE

NENG

E19P

YR

MCR

/kW

h$0

.024

11$0

.016

83$0

.011

09$0

.027

39$0

.011

69$0

.000

003

GENE

NGE1

9PV

ND

MCR

/kW

h$0

.037

35$0

.026

47$0

.042

38$0

.038

29$0

.031

16$0

.000

003

GENE

NGE1

9PV

YD

MCR

/kW

h$0

.039

03$0

.029

71$0

.050

29$0

.041

11$0

.038

26$0

.000

003

GENE

NGE1

9PV

YR

MCR

/kW

h$0

.023

18$0

.014

48$0

.016

03$0

.019

31$0

.005

16$0

.000

003

GENE

NGE1

9SN

DM

CR/k

Wh

$0.0

3683

$0.0

2678

$0.0

4120

$0.0

3726

$0.0

2946

$0.0

0000

3GE

NENG

E19S

YD

MCR

/kW

h$0

.038

97$0

.028

74$0

.048

50$0

.040

06$0

.036

43$0

.000

003

GENE

NGE1

9SY

RM

CR/k

Wh

$0.0

2285

$0.0

1648

$0.0

0952

$0.0

1894

$0.0

0629

$0.0

0000

3GE

NENG

E19S

VN

DM

CR/k

Wh

$0.0

3738

$0.0

2629

$0.0

4276

$0.0

3794

$0.0

3045

$0.0

0000

3GE

NENG

E19S

VY

DM

CR/k

Wh

$0.0

3946

$0.0

2921

$0.0

5014

$0.0

4067

$0.0

3767

$0.0

0000

3GE

NENG

E19S

VY

RM

CR/k

Wh

$0.0

3277

$0.0

1834

$0.0

2389

$0.0

3332

$0.0

1439

$0.0

0000

3GE

NENG

E19T

ND

MCR

/kW

h$0

.034

74$0

.023

84$0

.039

54$0

.035

23$0

.026

98$0

.000

003

GENE

NGE1

9TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE1

9TV

ND

MCR

/kW

h$0

.037

11$0

.025

59$0

.041

91$0

.038

60$0

.031

99$0

.000

003

GENE

NGE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE1

9TV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGE2

0PN

DM

CR/k

Wh

$0.0

3547

$0.0

2485

$0.0

4060

$0.0

3612

$0.0

2872

$0.0

0000

3GE

NENG

E20P

YD

MCR

/kW

h$0

.036

26$0

.026

16$0

.042

84$0

.036

86$0

.031

13$0

.000

003

GENE

NGE2

0PY

RM

CR/k

Wh

$0.0

2149

$0.0

1803

$0.0

1077

$0.0

1613

$0.0

0541

$0.0

0000

3GE

NENG

E20S

ND

MCR

/kW

h$0

.036

52$0

.026

77$0

.040

36$0

.036

94$0

.028

92$0

.000

003

GENE

NGE2

0SY

DM

CR/k

Wh

$0.0

3935

$0.0

2936

$0.0

4960

$0.0

3952

$0.0

3573

$0.0

0000

3GE

NENG

E20S

YR

MCR

/kW

h$0

.026

11$0

.016

86$0

.016

82$0

.027

38$0

.010

92$0

.000

003

GENE

NGE2

0TN

DM

CR/k

Wh

$0.0

3501

$0.0

2451

$0.0

4015

$0.0

3562

$0.0

2840

$0.0

0000

3GE

NENG

E20T

YD

MCR

/kW

h$0

.036

93$0

.028

66$0

.044

29$0

.036

93$0

.033

13$0

.000

003

GENE

NGE2

0TY

RM

CR/k

Wh

$0.0

2190

$0.0

1458

$0.0

1210

$0.0

2838

$0.0

0985

$0.0

0000

3GE

NENG

NonE

1N

DM

CR/k

Wh

$0.0

3881

$0.0

2750

$0.0

4631

$0.0

3939

$0.0

3398

$0.0

0000

3GE

NENG

NonE

1Y

DM

CR/k

Wh

$0.0

4279

$0.0

3453

$0.0

6051

$0.0

4294

$0.0

4628

$0.0

0000

3GE

NENG

NonE

1Y

RM

CR/k

Wh

$0.0

2293

$0.0

1930

$0.0

1174

$0.0

1297

$0.0

0772

$0.0

0000

3GE

NENG

STBY

PN

DM

CR/k

Wh

$0.0

3432

$0.0

2073

$0.0

3886

$0.0

3619

$0.0

2962

$0.0

0000

3GE

NENG

STBY

PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

SN

DM

CR/k

Wh

$0.0

3799

$0.0

2664

$0.0

4334

$0.0

3895

$0.0

3051

$0.0

0000

3GE

NENG

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

TN

DM

CR/k

Wh

$0.0

3464

$0.0

2521

$0.0

3967

$0.0

3756

$0.0

3344

$0.0

0000

Page

23

of 2

9

(PG&E-2)

1-AtchA-23

3GE

NENG

STBY

TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STL

ND

MCR

/kW

h$0

.040

86$0

.025

89$0

.061

03$0

.043

30$0

.053

38$0

.000

003

GENE

NGST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NENG

STL

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGTC

1N

DM

CR/k

Wh

$0.0

3797

$0.0

2544

$0.0

4482

$0.0

3948

$0.0

3375

$0.0

0000

3GE

NENG

TC1

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENE

NGTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A10

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A6N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XA6

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

A6Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

AY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BLg

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBL

gY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BLg

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBS

mN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XAG

BSm

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

AGBS

mY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9PV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19P

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SV

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19S

VY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9SV

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

VN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE1

9TV

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

E19T

VY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0SY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XE2

0TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XNo

nE1

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

NonE

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XNo

nE1

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

PN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYP

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

PY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYS

ND

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

SY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

BYS

YR

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

TN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

Page

24

of 2

9

(PG&E-2)

1-AtchA-24

3GE

NFLE

XST

BYT

YD

MCR

/kW

h$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

00$0

.000

003

GENF

LEX

STBY

TY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LN

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LY

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XST

LY

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1N

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1Y

DM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

3GE

NFLE

XTC

1Y

RM

CR/k

Wh

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

$0.0

0000

1PR

IDIS

TA1

ND

MCR

/kW

h$0

.007

64$0

.067

14$0

.035

98$0

.001

18$0

.003

75$0

.001

631

PRID

IST

A1Y

DM

CR/k

Wh

$0.0

0459

$0.0

7616

$0.0

3097

$0.0

0103

$0.0

0413

$0.0

0101

1PR

IDIS

TA1

YR

MCR

/kW

h$0

.005

35$0

.038

58$0

.017

18$0

.000

85$0

.000

95$0

.000

491

PRID

IST

A10

ND

MCR

/kW

h$0

.008

62$0

.063

09$0

.037

33$0

.001

14$0

.002

95$0

.001

701

PRID

IST

A10

YD

MCR

/kW

h$0

.005

82$0

.067

83$0

.032

07$0

.001

51$0

.003

75$0

.002

761

PRID

IST

A10

YR

MCR

/kW

h$0

.005

20$0

.027

99$0

.015

48$0

.000

62$0

.000

87$0

.000

481

PRID

IST

A6N

DM

CR/k

Wh

$0.0

0499

$0.0

6534

$0.0

2618

$0.0

0094

$0.0

0421

$0.0

0127

1PR

IDIS

TA6

YD

MCR

/kW

h$0

.006

11$0

.080

01$0

.027

76$0

.001

66$0

.003

33$0

.001

591

PRID

IST

A6Y

RM

CR/k

Wh

$0.0

0350

$0.0

3669

$0.0

1683

$0.0

0058

$0.0

0066

-$0.

0001

21

PRID

IST

AGA

ND

MCR

/kW

h$0

.008

41$0

.075

58$0

.026

66$0

.001

12$0

.006

82$0

.001

401

PRID

IST

AGA

YD

MCR

/kW

h$0

.004

77$0

.083

07$0

.023

21$0

.002

26$0

.002

30$0

.000

741

PRID

IST

AGA

YR

MCR

/kW

h$0

.004

15$0

.040

69$0

.013

27$0

.000

20$0

.000

36$0

.000

271

PRID

IST

AGBL

gN

DM

CR/k

Wh

$0.0

2148

$0.0

4837

$0.0

2714

$0.0

0379

$0.0

0388

$0.0

0576

1PR

IDIS

TAG

BLg

YD

MCR

/kW

h$0

.015

08$0

.062

39$0

.025

15$0

.002

33$0

.009

20$0

.004

811

PRID

IST

AGBL

gY

RM

CR/k

Wh

$0.0

0488

$0.0

1973

$0.0

0617

-$0.

0005

9$0

.001

02$0

.001

141

PRID

IST

AGBS

mN

DM

CR/k

Wh

$0.0

2052

$0.0

7058

$0.0

3024

$0.0

0278

$0.0

0354

$0.0

0474

1PR

IDIS

TAG

BSm

YD

MCR

/kW

h$0

.012

51$0

.085

58$0

.031

34$0

.004

45$0

.006

12$0

.000

291

PRID

IST

AGBS

mY

RM

CR/k

Wh

$0.0

0646

$0.0

1914

$0.0

0666

-$0.

0006

6$0

.000

31$0

.000

231

PRID

IST

E1N

DM

CR/k

Wh

$0.0

0291

$0.1

0302

$0.0

3017

$0.0

0049

$0.0

0376

$0.0

0038

1PR

IDIS

TE1

YD

MCR

/kW

h$0

.001

23$0

.120

43$0

.023

60$0

.000

23$0

.002

24$0

.000

211

PRID

IST

E1Y

RM

CR/k

Wh

$0.0

0244

$0.0

3244

$0.0

1188

$0.0

0032

$0.0

0054

$0.0

0014

1PR

IDIS

TE1

9PN

DM

CR/k

Wh

$0.0

0951

$0.0

4713

$0.0

3003

$0.0

0244

$0.0

0302

$0.0

0232

1PR

IDIS

TE1

9PY

DM

CR/k

Wh

$0.0

0781

$0.0

6746

$0.0

3257

$0.0

0130

$0.0

0307

$0.0

0107

1PR

IDIS

TE1

9PY

RM

CR/k

Wh

$0.0

0405

$0.0

2276

$0.0

0994

$0.0

0012

$0.0

0018

$0.0

0049

1PR

IDIS

TE1

9SN

DM

CR/k

Wh

$0.0

1101

$0.0

4775

$0.0

3633

$0.0

0142

$0.0

0238

$0.0

0309

1PR

IDIS

TE1

9SY

DM

CR/k

Wh

$0.0

0465

$0.0

6839

$0.0

3771

$0.0

0103

$0.0

0209

$0.0

0170

1PR

IDIS

TE1

9SY

RM

CR/k

Wh

$0.0

0462

$0.0

2317

$0.0

1296

$0.0

0022

$0.0

0073

$0.0

0030

1PR

IDIS

TE2

0PN

DM

CR/k

Wh

$0.0

0998

$0.0

4405

$0.0

2850

$0.0

0248

$0.0

0303

$0.0

0273

1PR

IDIS

TE2

0PY

DM

CR/k

Wh

$0.0

0939

$0.0

5517

$0.0

3047

$0.0

0108

$0.0

0300

$0.0

0327

1PR

IDIS

TE2

0PY

RM

CR/k

Wh

$0.0

0132

$0.0

4487

$0.0

1105

$0.0

0001

$0.0

0001

$0.0

0006

1PR

IDIS

TE2

0SN

DM

CR/k

Wh

$0.0

1358

$0.0

3586

$0.0

3305

$0.0

0247

$0.0

0289

$0.0

0593

1PR

IDIS

TE2

0SY

DM

CR/k

Wh

$0.0

0690

$0.0

6826

$0.0

3332

$0.0

0052

$0.0

0195

$0.0

0072

1PR

IDIS

TE2

0SY

RM

CR/k

Wh

$0.0

0188

$0.0

0519

$0.0

0461

$0.0

0002

$0.0

0013

$0.0

0002

1PR

IDIS

TE1

9PV

ND

MCR

/kW

h$0

.009

09$0

.039

41$0

.023

42$0

.002

91$0

.004

46$0

.002

861

PRID

IST

E19P

VY

DM

CR/k

Wh

$0.0

0617

$0.0

6714

$0.0

3626

$0.0

0070

$0.0

0551

$0.0

0127

1PR

IDIS

TE1

9PV

YR

MCR

/kW

h$0

.003

69$0

.029

55$0

.012

67$0

.000

77$0

.000

93$0

.000

161

PRID

IST

E19S

VN

DM

CR/k

Wh

$0.0

0619

$0.0

5674

$0.0

2867

$0.0

0093

$0.0

0416

$0.0

0148

1PR

IDIS

TE1

9SV

YD

MCR

/kW

h$0

.005

42$0

.064

31$0

.027

09$0

.000

79$0

.003

20$0

.001

361

PRID

IST

E19S

VY

RM

CR/k

Wh

$0.0

0354

$0.0

1062

$0.0

1046

$0.0

0023

$0.0

0053

$0.0

0059

1PR

IDIS

TNo

nE1

ND

MCR

/kW

h$0

.001

98$0

.097

60$0

.025

96$0

.000

64$0

.004

88$0

.000

441

PRID

IST

NonE

1Y

DM

CR/k

Wh

$0.0

0977

$0.1

2194

$0.0

2300

$0.0

0027

$0.0

0287

$0.0

0023

1PR

IDIS

TNo

nE1

YR

MCR

/kW

h$0

.002

30$0

.031

11$0

.011

93$0

.000

31$0

.000

57$0

.000

141

PRID

IST

STL

ND

MCR

/kW

h$0

.001

04$0

.042

10$0

.011

19$0

.000

48$0

.003

47$0

.000

981

PRID

IST

STL

YD

MCR

/kW

h$0

.000

07$0

.040

50$0

.009

01$0

.000

79$0

.010

84$0

.027

861

PRID

IST

STL

YR

MCR

/kW

h$0

.011

61$0

.003

18$0

.010

68$0

.007

17$0

.003

41$0

.007

581

PRID

IST

TC1

ND

MCR

/kW

h$0

.002

75$0

.054

95$0

.020

34$0

.000

38$0

.002

58$0

.000

621

SECD

IST

A1N

DM

CR/k

Wh

$0.0

0081

1SE

CDIS

TA1

YD

MCR

/kW

h$0

.000

921

SECD

IST

A1Y

RM

CR/k

Wh

-$0.

0006

21

SECD

IST

A10

ND

MCR

/kW

h$0

.000

701

SECD

IST

A10

YD

MCR

/kW

h$0

.000

701

SECD

IST

A10

YR

MCR

/kW

h-$

0.00

090

1SE

CDIS

TA6

ND

MCR

/kW

h$0

.001

091

SECD

IST

A6Y

DM

CR/k

Wh

$0.0

0088

1SE

CDIS

TA6

YR

MCR

/kW

h-$

0.00

112

1SE

CDIS

TAG

AN

DM

CR/k

Wh

$0.0

0198

1SE

CDIS

TAG

AY

DM

CR/k

Wh

$0.0

0320

1SE

CDIS

TAG

AY

RM

CR/k

Wh

-$0.

0005

3

Page

25

of 2

9

(PG&E-2)

1-AtchA-25

1SE

CDIS

TAG

BLg

ND

MCR

/kW

h$0

.001

151

SECD

IST

AGBL

gY

DM

CR/k

Wh

$0.0

0086

1SE

CDIS

TAG

BLg

YR

MCR

/kW

h-$

0.00

051

1SE

CDIS

TAG

BSm

ND

MCR

/kW

h$0

.002

431

SECD

IST

AGBS

mY

DM

CR/k

Wh

$0.0

0114

1SE

CDIS

TAG

BSm

YR

MCR

/kW

h-$

0.00

071

1SE

CDIS

TE1

ND

MCR

/kW

h$0

.000

951

SECD

IST

E1Y

DM

CR/k

Wh

$0.0

0116

1SE

CDIS

TE1

YR

MCR

/kW

h-$

0.00

017

1SE

CDIS

TE1

9PN

DM

CR/k

Wh

$0.0

0055

1SE

CDIS

TE1

9PY

DM

CR/k

Wh

$0.0

0073

1SE

CDIS

TE1

9PY

RM

CR/k

Wh

-$0.

0006

31

SECD

IST

E19P

VN

DM

CR/k

Wh

$0.0

0054

1SE

CDIS

TE1

9PV

YD

MCR

/kW

h$0

.000

521

SECD

IST

E19P

VY

RM

CR/k

Wh

-$0.

0007

21

SECD

IST

E19S

ND

MCR

/kW

h$0

.000

531

SECD

IST

E19S

YD

MCR

/kW

h$0

.000

621

SECD

IST

E19S

YR

MCR

/kW

h-$

0.00

143

1SE

CDIS

TE1

9SV

ND

MCR

/kW

h$0

.000

461

SECD

IST

E19S

VY

DM

CR/k

Wh

$0.0

0063

1SE

CDIS

TE1

9SV

YR

MCR

/kW

h-$

0.00

097

1SE

CDIS

TE2

0PN

DM

CR/k

Wh

$0.0

0041

1SE

CDIS

TE2

0PY

DM

CR/k

Wh

$0.0

0062

1SE

CDIS

TE2

0PY

RM

CR/k

Wh

-$0.

0003

41

SECD

IST

E20S

ND

MCR

/kW

h$0

.000

461

SECD

IST

E20S

YD

MCR

/kW

h$0

.000

571

SECD

IST

E20S

YR

MCR

/kW

h-$

0.00

067

1SE

CDIS

TNo

nE1

ND

MCR

/kW

h$0

.001

431

SECD

IST

NonE

1Y

DM

CR/k

Wh

$0.0

0146

1SE

CDIS

TNo

nE1

YR

MCR

/kW

h-$

0.00

009

1SE

CDIS

TST

BYP

ND

MCR

/kW

h$0

.003

841

SECD

IST

STBY

PY

DM

CR/k

Wh

$0.0

0000

1SE

CDIS

TST

BYP

YR

MCR

/kW

h$0

.000

001

SECD

IST

STBY

SN

DM

CR/k

Wh

$0.0

0380

1SE

CDIS

TST

BYS

YD

MCR

/kW

h$0

.000

001

SECD

IST

STBY

SY

RM

CR/k

Wh

$0.0

0000

1SE

CDIS

TST

LN

DM

CR/k

Wh

$0.0

0047

1SE

CDIS

TST

LY

DM

CR/k

Wh

$0.0

0048

1SE

CDIS

TST

LY

RM

CR/k

Wh

$0.0

0000

1SE

CDIS

TTC

1N

DM

CR/k

Wh

$0.0

0030

1SE

CDIS

TTC

1Y

DM

CR/k

Wh

$0.0

0000

1SE

CDIS

TTC

1Y

RM

CR/k

Wh

$0.0

0000

1NB

DIST

A1N

DM

CR/k

Wh

$0.0

1314

1NB

DIST

A1Y

DM

CR/k

Wh

$0.0

1523

1NB

DIST

A1Y

RM

CR/k

Wh

-$0.

0098

21

NBDI

STA1

0N

DM

CR/k

Wh

$0.0

1114

1NB

DIST

A10

YD

MCR

/kW

h$0

.011

231

NBDI

STA1

0Y

RM

CR/k

Wh

-$0.

0147

31

NBDI

STA6

ND

MCR

/kW

h$0

.017

171

NBDI

STA6

YD

MCR

/kW

h$0

.014

421

NBDI

STA6

YR

MCR

/kW

h-$

0.01

807

1NB

DIST

AGA

ND

MCR

/kW

h$0

.030

381

NBDI

STAG

AY

DM

CR/k

Wh

$0.0

4923

1NB

DIST

AGA

YR

MCR

/kW

h-$

0.00

870

1NB

DIST

AGBL

gN

DM

CR/k

Wh

$0.0

1641

1NB

DIST

AGBL

gY

DM

CR/k

Wh

$0.0

1265

1NB

DIST

AGBL

gY

RM

CR/k

Wh

-$0.

0072

61

NBDI

STAG

BSm

ND

MCR

/kW

h$0

.036

271

NBDI

STAG

BSm

YD

MCR

/kW

h$0

.017

191

NBDI

STAG

BSm

YR

MCR

/kW

h-$

0.01

028

1NB

DIST

E1N

DM

CR/k

Wh

$0.0

1513

1NB

DIST

E1Y

DM

CR/k

Wh

$0.0

1816

1NB

DIST

E1Y

RM

CR/k

Wh

-$0.

0028

51

NBDI

STE1

9PN

DM

CR/k

Wh

$0.0

0831

1NB

DIST

E19P

YD

MCR

/kW

h$0

.011

091

NBDI

STE1

9PY

RM

CR/k

Wh

-$0.

0098

7

Page

26

of 2

9

(PG&E-2)

1-AtchA-26

1NB

DIST

E19P

VN

DM

CR/k

Wh

$0.0

0824

1NB

DIST

E19P

VY

DM

CR/k

Wh

$0.0

0782

1NB

DIST

E19P

VY

RM

CR/k

Wh

-$0.

0102

41

NBDI

STE1

9SN

DM

CR/k

Wh

$0.0

0849

1NB

DIST

E19S

YD

MCR

/kW

h$0

.009

661

NBDI

STE1

9SY

RM

CR/k

Wh

-$0.

0210

81

NBDI

STE1

9SV

ND

MCR

/kW

h$0

.007

321

NBDI

STE1

9SV

YD

MCR

/kW

h$0

.010

171

NBDI

STE1

9SV

YR

MCR

/kW

h-$

0.01

569

1NB

DIST

E20P

ND

MCR

/kW

h$0

.006

361

NBDI

STE2

0PY

DM

CR/k

Wh

$0.0

0916

1NB

DIST

E20P

YR

MCR

/kW

h-$

0.00

521

1NB

DIST

E20S

ND

MCR

/kW

h$0

.007

471

NBDI

STE2

0SY

DM

CR/k

Wh

$0.0

0862

1NB

DIST

E20S

YR

MCR

/kW

h-$

0.00

921

1NB

DIST

NonE

1N

DM

CR/k

Wh

$0.0

2377

1NB

DIST

NonE

1Y

DM

CR/k

Wh

$0.0

2441

1NB

DIST

NonE

1Y

RM

CR/k

Wh

-$0.

0015

01

NBDI

STST

BYP

ND

MCR

/kW

h$0

.052

441

NBDI

STST

BYP

YD

MCR

/kW

h$0

.000

001

NBDI

STST

BYP

YR

MCR

/kW

h$0

.000

001

NBDI

STST

BYS

ND

MCR

/kW

h$0

.062

721

NBDI

STST

BYS

YD

MCR

/kW

h$0

.000

001

NBDI

STST

BYS

YR

MCR

/kW

h$0

.000

001

NBDI

STST

LN

DM

CR/k

Wh

$0.0

0760

1NB

DIST

STL

YD

MCR

/kW

h$0

.006

911

NBDI

STST

LY

RM

CR/k

Wh

$0.0

0000

1NB

DIST

TC1

ND

MCR

/kW

h$0

.004

681

NBDI

STTC

1Y

DM

CR/k

Wh

$0.0

0000

1NB

DIST

TC1

YR

MCR

/kW

h$0

.000

002

PRID

IST

A1N

DM

CR/k

Wh

$0.0

1997

$0.0

7130

$0.0

0000

$0.0

0147

$0.0

0393

$0.0

0000

2PR

IDIS

TA1

YD

MCR

/kW

h$0

.016

61$0

.084

35$0

.000

00$0

.001

32$0

.004

46$0

.000

002

PRID

IST

A1Y

RM

CR/k

Wh

$0.0

1041

$0.0

4821

$0.0

0000

$0.0

0072

$0.0

0080

$0.0

0000

2PR

IDIS

TA1

0N

DM

CR/k

Wh

$0.0

2030

$0.0

6626

$0.0

0000

$0.0

0139

$0.0

0298

$0.0

0000

2PR

IDIS

TA1

0Y

DM

CR/k

Wh

$0.0

1694

$0.0

7456

$0.0

0000

$0.0

0181

$0.0

0401

$0.0

0000

2PR

IDIS

TA1

0Y

RM

CR/k

Wh

$0.0

0906

$0.0

3348

$0.0

0000

$0.0

0059

$0.0

0065

$0.0

0000

2PR

IDIS

TA6

ND

MCR

/kW

h$0

.014

29$0

.073

07$0

.000

00$0

.001

24$0

.004

60$0

.000

002

PRID

IST

A6Y

DM

CR/k

Wh

$0.0

1579

$0.0

9058

$0.0

0000

$0.0

0179

$0.0

0356

$0.0

0000

2PR

IDIS

TA6

YR

MCR

/kW

h$0

.009

41$0

.044

21$0

.000

00$0

.000

33$0

.000

50$0

.000

002

PRID

IST

AGA

ND

MCR

/kW

h$0

.016

46$0

.086

98$0

.000

00$0

.001

23$0

.010

05$0

.000

002

PRID

IST

AGA

YD

MCR

/kW

h$0

.013

88$0

.094

70$0

.000

00$0

.002

11$0

.002

67$0

.000

002

PRID

IST

AGA

YR

MCR

/kW

h$0

.008

69$0

.057

19$0

.000

00$0

.000

23$0

.000

49$0

.000

002

PRID

IST

AGBL

gN

DM

CR/k

Wh

$0.0

2444

$0.0

5328

$0.0

0000

$0.0

0403

$0.0

0413

$0.0

0000

2PR

IDIS

TAG

BLg

YD

MCR

/kW

h$0

.020

80$0

.068

95$0

.000

00$0

.002

85$0

.011

27$0

.000

002

PRID

IST

AGBL

gY

RM

CR/k

Wh

$0.0

0585

$0.0

2908

$0.0

0000

$0.0

0008

$0.0

0136

$0.0

0000

2PR

IDIS

TAG

BSm

ND

MCR

/kW

h$0

.025

52$0

.079

84$0

.000

00$0

.003

13$0

.004

02$0

.000

002

PRID

IST

AGBS

mY

DM

CR/k

Wh

$0.0

2211

$0.0

9440

$0.0

0000

$0.0

0426

$0.0

0721

$0.0

0000

2PR

IDIS

TAG

BSm

YR

MCR

/kW

h$0

.007

04$0

.028

41$0

.000

00-$

0.00

032

$0.0

0060

$0.0

0000

2PR

IDIS

TE1

ND

MCR

/kW

h$0

.020

82$0

.119

10$0

.000

00$0

.000

82$0

.004

05$0

.000

002

PRID

IST

E1Y

DM

CR/k

Wh

$0.0

2033

$0.1

4288

$0.0

0000

$0.0

0047

$0.0

0239

$0.0

0000

2PR

IDIS

TE1

YR

MCR

/kW

h$0

.006

22$0

.041

35$0

.000

00$0

.000

27$0

.000

50$0

.000

002

PRID

IST

E19P

ND

MCR

/kW

h$0

.016

96$0

.049

02$0

.000

00$0

.002

51$0

.002

84$0

.000

002

PRID

IST

E19P

YD

MCR

/kW

h$0

.017

71$0

.073

77$0

.000

00$0

.001

49$0

.002

97$0

.000

002

PRID

IST

E19P

YR

MCR

/kW

h$0

.006

79$0

.026

23$0

.000

00$0

.000

23$0

.000

08$0

.000

002

PRID

IST

E19S

ND

MCR

/kW

h$0

.019

81$0

.047

87$0

.000

00$0

.001

75$0

.002

04$0

.000

002

PRID

IST

E19S

YD

MCR

/kW

h$0

.017

21$0

.075

88$0

.000

00$0

.001

20$0

.002

10$0

.000

002

PRID

IST

E19S

YR

MCR

/kW

h$0

.007

64$0

.028

21$0

.000

00$0

.000

26$0

.001

05$0

.000

002

PRID

IST

E20P

ND

MCR

/kW

h$0

.016

63$0

.046

13$0

.000

00$0

.002

55$0

.003

09$0

.000

002

PRID

IST

E20P

YD

MCR

/kW

h$0

.017

39$0

.059

23$0

.000

00$0

.001

46$0

.003

04$0

.000

002

PRID

IST

E20P

YR

MCR

/kW

h$0

.006

99$0

.054

93$0

.000

00$0

.000

03$0

.000

01$0

.000

002

PRID

IST

E20S

ND

MCR

/kW

h$0

.019

70$0

.034

94$0

.000

00$0

.002

98$0

.002

17$0

.000

002

PRID

IST

E20S

YD

MCR

/kW

h$0

.017

41$0

.074

08$0

.000

00$0

.000

68$0

.001

92$0

.000

002

PRID

IST

E20S

YR

MCR

/kW

h$0

.002

92$0

.005

10$0

.000

00$0

.000

03$0

.000

14$0

.000

002

PRID

IST

E19P

VN

DM

CR/k

Wh

$0.0

1459

$0.0

4181

$0.0

0000

$0.0

0301

$0.0

0475

$0.0

0000

2PR

IDIS

TE1

9PV

YD

MCR

/kW

h$0

.018

15$0

.069

54$0

.000

00$0

.001

09$0

.006

48$0

.000

002

PRID

IST

E19P

VY

RM

CR/k

Wh

$0.0

0785

$0.0

3442

$0.0

0000

$0.0

0052

$0.0

0125

$0.0

0000

Page

27

of 2

9

(PG&E-2)

1-AtchA-27

2PR

IDIS

TE1

9SV

ND

MCR

/kW

h$0

.015

53$0

.061

17$0

.000

00$0

.001

19$0

.004

99$0

.000

002

PRID

IST

E19S

VY

DM

CR/k

Wh

$0.0

1564

$0.0

7110

$0.0

0000

$0.0

0107

$0.0

0343

$0.0

0000

2PR

IDIS

TE1

9SV

YR

MCR

/kW

h$0

.005

44$0

.009

14$0

.000

00$0

.000

30$0

.000

50$0

.000

002

PRID

IST

NonE

1N

DM

CR/k

Wh

$0.0

1488

$0.1

1399

$0.0

0000

$0.0

0099

$0.0

0520

$0.0

0000

2PR

IDIS

TNo

nE1

YD

MCR

/kW

h$0

.023

77$0

.145

27$0

.000

00$0

.000

57$0

.003

04$0

.000

002

PRID

IST

NonE

1Y

RM

CR/k

Wh

$0.0

0641

$0.0

3871

$0.0

0000

$0.0

0027

$0.0

0056

$0.0

0000

2PR

IDIS

TST

LN

DM

CR/k

Wh

$0.0

0617

$0.0

5265

$0.0

0000

$0.0

0074

$0.0

0375

$0.0

0000

2PR

IDIS

TST

LY

DM

CR/k

Wh

$0.0

0449

$0.0

5189

$0.0

0000

$0.0

0192

$0.0

0951

$0.0

0000

2PR

IDIS

TST

LY

RM

CR/k

Wh

$0.0

1150

$0.0

0000

$0.0

0000

$0.0

0730

$0.0

0000

$0.0

0000

2PR

IDIS

TTC

1N

DM

CR/k

Wh

$0.0

1024

$0.0

6217

$0.0

0000

$0.0

0058

$0.0

0278

$0.0

0000

3PR

IDIS

TA1

ND

MCR

/kW

h$0

.007

96$0

.043

87$0

.021

49$0

.000

25$0

.000

45$0

.000

003

PRID

IST

A1Y

DM

CR/k

Wh

$0.0

0790

$0.0

5185

$0.0

2871

$0.0

0022

$0.0

0058

$0.0

0000

3PR

IDIS

TA1

YR

MCR

/kW

h$0

.005

14$0

.013

85$0

.005

48$0

.000

01$0

.000

04$0

.000

003

PRID

IST

A10

ND

MCR

/kW

h$0

.006

77$0

.045

82$0

.019

68$0

.000

14$0

.000

25$0

.000

003

PRID

IST

A10

YD

MCR

/kW

h$0

.007

95$0

.049

00$0

.025

64$0

.000

18$0

.000

35$0

.000

003

PRID

IST

A10

YR

MCR

/kW

h$0

.004

20$0

.012

39$0

.004

26$0

.000

00$0

.000

03$0

.000

003

PRID

IST

A6N

DM

CR/k

Wh

$0.0

0678

$0.0

3827

$0.0

2301

$0.0

0044

$0.0

0080

$0.0

0000

3PR

IDIS

TA6

YD

MCR

/kW

h$0

.009

63$0

.051

27$0

.030

66$0

.000

27$0

.002

07$0

.000

003

PRID

IST

A6Y

RM

CR/k

Wh

$0.0

0373

$0.0

1385

$0.0

0430

$0.0

0001

$0.0

0036

$0.0

0000

3PR

IDIS

TAG

AN

DM

CR/k

Wh

$0.0

1265

$0.0

2444

$0.0

3809

$0.0

0108

$0.0

0045

$0.0

0000

3PR

IDIS

TAG

AY

DM

CR/k

Wh

$0.0

1213

$0.0

2565

$0.0

4006

$0.0

0019

$0.0

0588

$0.0

0000

3PR

IDIS

TAG

AY

RM

CR/k

Wh

$0.0

0811

$0.0

0756

$0.0

0490

$0.0

0000

$0.0

0002

$0.0

0000

3PR

IDIS

TAG

BLg

ND

MCR

/kW

h$0

.014

03$0

.025

67$0

.038

24$0

.003

56$0

.001

97$0

.000

003

PRID

IST

AGBL

gY

DM

CR/k

Wh

$0.0

1107

$0.0

2945

$0.0

4896

$0.0

0020

$0.0

0058

$0.0

0000

3PR

IDIS

TAG

BLg

YR

MCR

/kW

h$0

.003

75$0

.004

88$0

.007

55-$

0.00

004

$0.0

0019

$0.0

0000

3PR

IDIS

TAG

BSm

ND

MCR

/kW

h$0

.019

56$0

.022

77$0

.040

62$0

.002

61$0

.001

18$0

.000

003

PRID

IST

AGBS

mY

DM

CR/k

Wh

$0.0

1700

$0.0

2143

$0.0

5504

$0.0

0078

$0.0

0004

$0.0

0000

3PR

IDIS

TAG

BSm

YR

MCR

/kW

h$0

.004

42$0

.004

82$0

.007

71-$

0.00

002

$0.0

0005

$0.0

0000

3PR

IDIS

TE1

ND

MCR

/kW

h$0

.017

52$0

.050

00$0

.040

69$0

.000

31$0

.000

88$0

.000

003

PRID

IST

E1Y

DM

CR/k

Wh

$0.0

1943

$0.0

7513

$0.0

6234

$0.0

0011

$0.0

0046

$0.0

0000

3PR

IDIS

TE1

YR

MCR

/kW

h$0

.003

83$0

.007

43$0

.002

40$0

.000

01$0

.000

01$0

.000

003

PRID

IST

E19P

ND

MCR

/kW

h$0

.004

54$0

.042

33$0

.021

04$0

.001

63$0

.000

87$0

.000

003

PRID

IST

E19P

YD

MCR

/kW

h$0

.005

56$0

.055

69$0

.030

56$0

.000

25$0

.000

71$0

.000

003

PRID

IST

E19P

YR

MCR

/kW

h$0

.002

24$0

.010

66$0

.006

61$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E19S

ND

MCR

/kW

h$0

.004

09$0

.050

85$0

.018

56$0

.000

07$0

.000

10$0

.000

003

PRID

IST

E19S

YD

MCR

/kW

h$0

.005

21$0

.060

48$0

.025

74$0

.000

02$0

.000

09$0

.000

003

PRID

IST

E19S

YR

MCR

/kW

h$0

.002

72$0

.011

07$0

.005

14$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E20P

ND

MCR

/kW

h$0

.005

41$0

.040

71$0

.022

07$0

.001

88$0

.000

53$0

.000

003

PRID

IST

E20P

YD

MCR

/kW

h$0

.006

84$0

.042

63$0

.026

17$0

.000

11$0

.000

10$0

.000

003

PRID

IST

E20P

YR

MCR

/kW

h$0

.005

08$0

.010

81$0

.005

68$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E20S

ND

MCR

/kW

h$0

.003

66$0

.051

54$0

.020

85$0

.000

08$0

.000

19$0

.000

003

PRID

IST

E20S

YD

MCR

/kW

h$0

.004

59$0

.045

36$0

.030

36$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E20S

YR

MCR

/kW

h$0

.000

15$0

.004

48$0

.001

71$0

.000

00$0

.000

00$0

.000

003

PRID

IST

E19P

VN

DM

CR/k

Wh

$0.0

0662

$0.0

2962

$0.0

2103

$0.0

0237

$0.0

0137

$0.0

0000

3PR

IDIS

TE1

9PV

YD

MCR

/kW

h$0

.006

36$0

.049

09$0

.032

86$0

.000

04$0

.001

68$0

.000

003

PRID

IST

E19P

VY

RM

CR/k

Wh

$0.0

0404

$0.0

1116

$0.0

0343

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TE1

9SV

ND

MCR

/kW

h$0

.006

03$0

.041

45$0

.021

45$0

.000

18$0

.000

36$0

.000

003

PRID

IST

E19S

VY

DM

CR/k

Wh

$0.0

0726

$0.0

4653

$0.0

2461

$0.0

0023

$0.0

0065

$0.0

0000

3PR

IDIS

TE1

9SV

YR

MCR

/kW

h$0

.001

10$0

.011

27$0

.005

98$0

.000

02$0

.000

07$0

.000

003

PRID

IST

NonE

1N

DM

CR/k

Wh

$0.0

1398

$0.0

3704

$0.0

3117

$0.0

0049

$0.0

0141

$0.0

0000

3PR

IDIS

TNo

nE1

YD

MCR

/kW

h$0

.023

57$0

.077

11$0

.057

41$0

.000

17$0

.000

69$0

.000

003

PRID

IST

NonE

1Y

RM

CR/k

Wh

$0.0

0453

$0.0

0741

$0.0

0209

$0.0

0001

$0.0

0001

$0.0

0000

3PR

IDIS

TST

LN

DM

CR/k

Wh

$0.0

0268

$0.0

2822

$0.0

2205

$0.0

0023

$0.0

0118

$0.0

0000

3PR

IDIS

TST

LY

DM

CR/k

Wh

$0.0

0101

$0.0

3046

$0.0

4752

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TST

LY

RM

CR/k

Wh

$0.0

0160

$0.0

2961

$0.0

2847

$0.0

0000

$0.0

0000

$0.0

0000

3PR

IDIS

TTC

1N

DM

CR/k

Wh

$0.0

0455

$0.0

3296

$0.0

1828

$0.0

0011

$0.0

0047

$0.0

0000

Tabl

e Scen

ario

IDSc

enar

ioNa

me

Desc

riptio

n1

Base

Sce

nario

4 pm

-9 p

m S

umm

er a

nd W

inte

r Pea

ks, 2

-4, 9

-11

pm S

umm

er P

art-p

eak,

10

am -

2 pm

, Mar

ch-M

ay S

uper

off-

peak

: All

Sche

dule

s2

AG S

cena

rio5

pm-8

pm

Sum

mer

and

Win

ter P

eaks

, no

Part

-pea

k: A

ll Sc

hedu

les

3Le

gacy

Sce

nario

12 n

oon-

6 pm

Wee

kday

Sum

mer

Pea

k, 9

-12,

6-1

0 Su

mm

er P

art-p

eak,

Wee

kday

9 a

m-1

0 pm

Win

ter P

eak

Not

es

Page

28

of 2

9

(PG&E-2)

1-AtchA-28

Scen

ario

ID: U

niqu

e nu

mbe

r ide

ntify

ing

the

mar

gina

l cos

ts. S

cena

rioID

1 is

pre

limin

ary

mar

gina

l cos

t num

bers

for t

he G

RC.

Desc

riptio

n: B

rief d

escr

iptio

n of

the

mar

gina

l cos

tsPR

IMDI

STPr

imar

y Di

strib

utio

n Ca

pacit

yNB

DIST

New

Bus

ines

s Dist

ribut

ion

Capa

city

SECD

IST

Seco

ndar

y Di

strib

utio

n Ca

pacit

yRC

SRe

venu

e Cy

cle S

ervi

ceM

CEC

Mar

gina

l Con

nect

ion

Equi

pmen

t Cos

tGE

NCAP

Gene

ric G

ener

atio

n Ca

pacit

yGE

NFLE

XFl

exib

le G

ener

atio

n Ca

pacit

yGE

NENG

Gene

ratio

n En

ergy

Page

29

of 2

9

(PG&E-2)

1-AtchA-29

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 2

MARGINAL GENERATION COSTS

(PG&E-2)

2-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 2

MARGINAL GENERATION COSTS

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 2-1

1. Definition .................................................................................................... 2-2

2. Methodology .............................................................................................. 2-2

a. Marginal Energy Cost .......................................................................... 2-3

b. Marginal Generation Capacity Cost ..................................................... 2-3

1) System Capacity Cost ................................................................... 2-3

2) Local Capacity Cost ...................................................................... 2-5

3) Flexible (Flex) Capacity Cost ........................................................ 2-6

3. Proposed Methodology Compared to Last Filed Methodology ................... 2-7

4. PG&E’s MEC and MGCC Results .............................................................. 2-8

B. Costing Methodology ...................................................................................... 2-10

1. Marginal Energy Costs ............................................................................. 2-10

a. What Drives Electricity Prices? .......................................................... 2-11

b. Hourly Energy Price Forecast Methodology ...................................... 2-12

1) Hourly Marginal Energy Costs ..................................................... 2-13

2) Methodology ................................................................................ 2-18

3) Hourly Market Heat Rate Calibration and Forecasts ................... 2-20

c. Ramping Effects ................................................................................ 2-24

d. Intra-Day Spread ............................................................................... 2-27

e. Expanded Influence of Conditions in Rest of Western Electricity Coordinating Council ......................................................................... 2-29

1) Modeling Load Impacts on Imports Via Temperatures ................ 2-29

2) Modeling Impacts on Imports via River Flows ............................. 2-31

f. Calibration ......................................................................................... 2-32

(PG&E-2) PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 2 MARGINAL GENERATION COSTS

TABLE OF CONTENTS

(CONTINUED)

2-ii

g. Forecast ............................................................................................. 2-36

h. Computing MECs at the Transmission Voltage Level by TOU Period ................................................................................................ 2-38

i. Adjusting MECs for the Impact of Energy Storage ............................ 2-38

j. Computing MEC at the Distribution Voltage Levels Using Energy Line Loss Factors .............................................................................. 2-40

2. Marginal Generation Capacity Costs ....................................................... 2-41

a. Generic Capacity Price Forecast ....................................................... 2-41

1) Model and Data Transparency .................................................... 2-41

2) Principles ..................................................................................... 2-42

3) Energy Gross Margins ................................................................ 2-43

4) Short-Run Avoided Cost of Capacity for 2021-2030 ................... 2-51

5) Long-Run Avoided Cost of Capacity for 2023-2030 .................... 2-52

6) Computing Levelized MGCC for 2021-2026 and 2025-2030 and Adjusting by Voltage Level With Line Loss Factors .............. 2-55

b. Planning Reserve Margin .................................................................. 2-56

C. Conclusion ...................................................................................................... 2-56

(PG&E-2)

2-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 2 2

MARGINAL GENERATION COSTS 3

A. Introduction 4

The purpose of this chapter is to present Pacific Gas and Electric 5

Company’s (PG&E) estimates for the two components of its marginal generation 6

costs (MGC), revised as of late April 2020 for PG&E’s Update testimony. The 7

marginal costs associated with a change in customer electricity usage include 8

both an energy-related component and a capacity-related component. These 9

two components are: (1) Marginal Energy Costs (MEC) in cents per 10

kilowatt-hour (¢/kWh); and (2) Marginal Generation Capacity Costs (MGCC) in 11

dollars per kilowatt-year ($/kW-year). The MEC is calculated for six time-of-use 12

(TOU) rate periods and three voltage levels. MGCC is calculated for 13

three voltage levels.1 14

PG&E is also directed by Decision (D.) 17-01-006 to provide hourly load and 15

net load data, and compare estimated high and low marginal cost hours to the 16

shape of adjusted net load (ANL).2 The data, assumptions and explanations 17

required by this provision of D.17-01-006 are provided in this chapter where they 18

are introduced as part of the marginal cost analysis provided herein. 19

1 The MGCs proposed in this General Rate Case (GRC) testimony are not applicable for

determining Qualifying Facility short-run avoided costs or as-delivered capacity prices. However, PG&E’s MECs proposed in this testimony are a reasonable, publicly-sharable estimate of the value of renewable contracts that have traditionally used Time-of-Delivery (TOD) factors. As such, PG&E’s estimates of MECs are published as information-only TOD forecasts in accordance with California Public Utilities Commission (CPUC or Commission) D.19-02-007.

2 General Principle 3 (p. 7 of D.17-01-006) states: “As a secondary check on the marginal cost analysis, the IOUs should provide hourly load and net load data and explain any significant differences between estimated high and low marginal cost hours and the net load shapes (including adjusted net load data for PG&E). As part of its TOU period analysis, each investor-owned utility (IOU) should submit the latest data and assumptions, including those vetted in the Long-Term Procurement Planning (LTPP) and/or Integrated Resource Planning (IRP) or successor proceeding.”

(PG&E-2)

2-2

1. Definition 1

MEC reflects changes in generation costs associated with customers’ 2

energy usage. Specifically, MEC is the cost of procuring electricity to meet 3

one additional megawatt-hour (MWh) of load. 4

MGCC reflects changes in generation costs associated with customers’ 5

usage coincident with peak demand (i.e., capacity costs). The MGCC 6

actually consists of three components: generic, or system capacity (in which 7

the peak demand that defines the times when customer demand impacts 8

utility marginal costs refers to systemwide demand); local capacity (in which 9

the peak demand is measured at the local level); and flexible, or flex 10

capacity (in which the demand is currently measured based on the 11

three-hour positive ramp in system net load).3 12

The MCGs presented in this chapter are thus estimates of the change in 13

PG&E’s procurement costs that would be caused by small changes in 14

customer energy usage and peak demand or flex demand. They are not 15

intended to provide estimates of PG&E’s average or total electric 16

procurement costs. 17

In GRC Phase II proceedings, the CPUC addresses estimates of both 18

MECs and MGCCs for each investor-owned utility to determine the 19

generation-related rate components for that utility’s retail electric rates. 20

Using both MECs and MGCCs is an efficient approach for allocating PG&E’s 21

generation costs and providing the appropriate price signals to customers. 22

2. Methodology 23

For the 2020 GRC Phase II, as in the 2017 GRC Phase II, PG&E’s 24

proposed MECs and MGCCs are based on both publicly-available inputs 25

and a model available to other parties. By continuing to rely on public data 26

sources for its inputs and using models for avoided energy and capacity that 27

can be shared with other parties that have signed a Non-Disclosure 28

Agreement (NDA), PG&E is providing full transparency. 29

3 Net load is defined as measured, or gross load (customer needs less behind-the-meter

generation such as rooftop solar), less front-of-the-meter (FTM) solar and wind generation connected to the California Independent System Operator (CAISO) grid. Note that, as discussed below, the definition of flexible capacity is likely changing to one based on forecast error between the Day-Ahead (DA) and fifteen minute CAISO energy markets.

(PG&E-2)

2-3

a. Marginal Energy Cost 1

PG&E’s MEC is calculated based on hourly power price forecasts 2

for Northern California for the periods January 1, 2021 through 3

December 31, 2021, for setting rates in this proceeding, and January 1, 4

2025 through December 31, 2025, for evaluation of changes to TOU 5

periods in the future.4 The hourly power price forecasts are based on 6

the relationship between prices in the DA and Real-Time Markets (RTM) 7

of the CAISO, and load and generation in the CAISO, including the 8

impacts of conditions in the rest of the Western Electricity Coordinating 9

Council (WECC) and the operations of Energy Storage (ES) on prices of 10

energy and Ancillary Services (A/S). PG&E’s MEC is calculated for the 11

six TOU rate periods,5 and the three voltage levels,6 used by PG&E for 12

revenue allocation and rate design. 13

b. Marginal Generation Capacity Cost 14

1) System Capacity Cost 15

In previous GRC Phase II proceedings, the Commission 16

indicated a preference for calculating MGCC using a longer-term 17

4 In prior GRCs (e.g., 2017), PG&E considered MGCs in the TY of the proceeding

(i.e., 2017 for the 2017 GRC). However, with the hourly shape of costs changing as the generation mix changes going forward, costs as of the TY are no longer representative of costs that will be faced by customers once TOU periods and rates are set. In accordance with D.17-01-006, TOU periods must be evaluated using marginal costs forecasted as of at least three years after the new rates would go into effect. As discussed in Chapter 11, for this 2020 GRC, that forecast year is 2025. On the other hand, once TOU periods are set, PG&E sets both actual rates, and revenue allocation among PG&E’s bundled electric customer classes, based on the earliest year that rates determined in this GRC Phase II could be implemented, which is currently expected to be fall 2021.

5 PG&E’s current and proposed Summer season runs from June 1 through September 30, and its current and proposed Winter season runs from October 1 through May 31. The proposed peak period for most rates in both Summer and Winter is from 4 p.m. to 9 p.m., all days of the week; Agricultural rates have a 5 p.m. to 8 p.m. peak. For TOU rates that also have a partial-peak period in Summer, PG&E’s partial-peak period is from 2 p.m. to 4 p.m. and 9 p.m. to 11 p.m., all days of the week. PG&E’s Super Off-Peak is 9 a.m. to 2 p.m. all days of the week during March through May. All other hours not listed above are designated as off-peak. All hours are listed in Pacific Prevailing Time. PG&E is not proposing changes to its various existing TOU periods at this time, as explained in Chapter 11 of this exhibit.

6 The three voltage levels are transmission (60 kilovolt (kV) and above); primary distribution (between 4 kV and 60 kV); and, secondary distribution (below 4 kV).

(PG&E-2)

2-4

perspective and adopted a six-year planning horizon.7 Using that 1

precedent, PG&E has calculated the system capacity component of 2

its 2021 Test Year (TY) MGCC by levelizing six years of forecasted 3

annual Avoided Capacity Costs (ACC)8 from January 1, 2021 4

through December 31, 2026. Likewise, the 2025 MGCC is equal to 5

the levelized ACC from January 1, 2025 through December 31, 6

2030. PG&E’s annual ACCs are determined from the short-run 7

costs of capacity for years in which existing or forecasted capacity is 8

sufficient to meet the need,9 and from the long-run costs of capacity 9

for years in which existing or forecasted capacity is insufficient to 10

meet the need.10 11

In terms of which years between 2021 and 2026 should use 12

long-run versus short-run capacity costs, PG&E’s Update testimony 13

relies on modeling from the 2019-2020 Integrated Resource Plan 14

(IRP) Reference System Plan (RSP),11 and the IRP Procurement 15

Track initiated by the Assigned Commissioner Ruling (ACR) in 16

Rulemaking (R.) 16-02-007 on June 20, 2019, with D.19-11-016 17

being issued on November 7, 2020. Both the RSP and the ACR 18

Final Decision indicate a need for new procurement of capacity 19

starting in 2021 (which implies that long-run capacity costs should 20

be used starting in 2021). 21

The short-run cost of capacity is traditionally estimated by the 22

going-forward fixed costs of an existing marginal generation 23

resource net of its “energy gross margins”12 for that year. Because 24

7 Re Pacific Gas and Electric Company, D.89-12-057, 34 CPUC 2d 199, 317 (1989). 8 PG&E here uses ACC to refer to its calculation of annual ACCs. This is distinct from

the Avoided Cost Calculator model developed by Energy and Environmental Economics (E3) which is also often referred to as the ACC model.

9 The short-run cost of capacity is the additional payment required to keep online the most expensive existing resource required to meet the need.

10 The long-run cost of capacity is the additional payment required to incent building the least expensive new resource that can be built to meet the need.

11 The RSP was determined in D.20-03-028, published April 6, 2020. 12 EGM is the expected market revenue (revenues from sales of energy and A/S) net of

variable cost. The variable cost includes fuel and start-up (for thermal generators), charging energy (for ES), and Variable Operations and Maintenance (VOM).

(PG&E-2)

2-5

the generators most at “risk of retirement” are those with the highest 1

net costs, the marginal generation resource used for this calculation 2

is the one that yields the highest calculated cost of capacity 3

(excluding those that could be retired without running short of 4

capacity). 5

For years in which new capacity must be added to meet the 6

need, the long-run cost of capacity for each year is estimated using 7

the levelized Real Economic Carrying Charges (RECC) that 8

annualize the total fixed costs of a new marginal generation 9

resource net of Energy Gross Margins (EGM) over the assumed 10

asset life of the resource. Because the generators most likely to be 11

built are those with the lowest (net) cost, the marginal resource for 12

this calculation is the one that yields the lowest RECC. For reasons 13

explained below, PG&E assumes that FTM ES is the marginal 14

resource that will be procured and built to meet capacity needs 15

starting in 2021. 16

PG&E has calculated its MGCC for three voltage levels. 17

2) Local Capacity Cost 18

Based on the 2018 Resource Adequacy (RA) Report,13 there is 19

no premium for local capacity in PG&E’s service territory.14 PG&E 20

therefore considers that the local capacity cost is the same as the 21

overall average capacity cost for years 2021-2022. PG&E notes 22

that, per D.17-06-027, the definition of local areas is currently in the 23

process of transitioning from two areas (PG&E Bay Area and PG&E 24

Other) to seven areas (Greater Bay Area, Humboldt, North 25

Coast/North Bay, Sierra, Stockton, Fresno, and Kern). However, a 26

recent proposal by the Commission would restore the definition back 27

to the original definition until a central procurement entity is in place. 28

13 See Appendix A to Proposed Decision in R.19-11-009, issued February 21, 2020,

pp. 2-3. 14 The 85th percentile price over 2018-2022 in PG&E Service Territory (designated as

North of Path 26), is $4.00/kilowatt-month (kW-mo.) for both local RA and all RA; and $4.25/kW-mo. for CAISO system RA. (See Table 7 at p. 25 of 2018 RA Report.)

(PG&E-2)

2-6

The forecasts provided in the 2018 RA Report use the original 1

definition. 2

3) Flexible (Flex) Capacity Cost 3

The CAISO currently determines the flexibility requirement (Flex 4

requirement) based on the maximum three-hour ramp in net load, 5

plus the greater of the Most Severe Single Contingency and 6

3.5 percent of the expected (monthly) peak load.15 Because the 7

CAISO does not publish flexible capacity need forecasts beyond 8

two years, PG&E calculated these Flex requirements by day, for 9

each year from 2021 to 2030 and for each of the ten forecast 10

scenarios described in the MEC section. These were compared to 11

the total available Effective Flexible Capacity (EFC) using data from 12

the 2018 IRP proceeding, to establish days and hours in which EFC 13

was insufficient to meet the Flex requirement. Based on this 14

analysis, there are days on which EFC is less than the Flex 15

requirement in PG&E’s service territory (which PG&E refers to as 16

“Flex need days”) starting in 2021. 17

For its November 22, 2019 filing PG&E determined the Flex 18

Capacity Cost by considering the least-cost way to meet flexible 19

capacity constraints, which turned out to be curtailment of 20

contracted solar resources during the first hour of the maximum 21

ramp on Flex need days. The calculation yielded an hourly Flex 22

Capacity Cost which can be summed by TOU period similarly to the 23

MEC. 24

However, on December 20, 2019 the CAISO published its Third 25

Revised Straw Proposal for RA Enhancements, which proposes to 26

change the definition of Flex requirement to “a single flexible 27

capacity requirement equal to the historic forecasted net load error 28

between IFM and FMM plus a growth factor to account for additional 29

growth in uncertainty.” 16 Because the Flex capacity costs 30

15 See CAISO 2020 Final Flexible Capacity Need Assessment, p. 4. Available at

http://www.caiso.com/Documents/Final2020FlexibleCapacityNeedsAssessment.pdf. 16 Available at http://www.caiso.com/InitiativeDocuments/ThirdRevisedStrawProposal-

ResourceAdequacyEnhancements.pdf. See p. 72.

(PG&E-2)

2-7

calculated using the current definition were so low,17 while details of 1

the proposed Flex requirement calculations are still unclear,18 2

PG&E has set the Flex capacity costs to zero for this update. 3

3. Proposed Methodology Compared to Last Filed Methodology 4

Table 2-1 summarizes and compares the data sources and calculation 5

methodologies used by PG&E in both the previous 2017 GRC Phase II 6

proceeding and this proceeding. 7

17 The impact of Flex capacity costs on 2021 MECs was zero in summer months, and less

than 0.01 cents per kWh (i.e., less than $0.0001/kWh) in winter in HE15 (the highest-impact hour).

18 PG&E anticipates that Flex requirements are likely to be addressed in a Proposed Decision in R.10-11-009 setting the 2021 RA requirements, in May 2020.

(PG&E-2)

2-8

TABLE 2-1 OVERVIEW OF PROPOSED MARGINAL GENERATION COST DATA SOURCE, UNITS AND CALCULATION METHODOLOGY

Line No. Function

PG&E’s Methodology Submitted in the 2017 GRC Phase II

Proposed Methodology

1 MEC Data Source Forecasted market heat rates are based on a model that is calibrated to historical market heat rates, using a public market price forecast for gas and greenhouse gas (GHG). Historical market heat rates are based on the historical hourly prices in the CAISO Market.

Same as in 2017 GRC Phase II, except that the proposed forecasting model incorporates additional explanatory variables, is re-calibrated based on more recent data, and considers the impact of ES operations.

Demand Measure

kWh by TOU period kWh by TOU period

Units ¢ per kWh ¢ per kWh

2 MGCC Data Source RPS Calculator V6.1;

Long-Term Procurement Planning (LTPP) CPUC CAISO 2024, Trajectory scenario

2020 Integrated Resource Plan (RSP) – RESOLVE model inputs/outputs and Strategic Energy and Risk Valuation Model (SERVM) inputs

2018 Resource Adequacy Report

Demand Measure

Peak kW Peak kW

Units $ per kW-year $ per kW-year

Costing Methodology

A six-year levelized capacity cost net of gross margin.

Resource balance year (RBY) = beyond 2022.

PG&E’s public Avoided Cost Model with an existing combined cycle unit for 2017-2022.

Reflects the difference between the annualized capital cost of a unit and the net energy benefits the unit earns in the energy market.

A six-year levelized capacity cost net of gross margin.

RBY = 2021.

2021-2029: EPRI’s public StorageVET model with new 4-hr grid-scale battery.

2030: PG&E public Avoided Cost Model, with existing combined cycle unit.

4. PG&E’s MEC and MGCC Results 1

Table 2-2 summarizes PG&E’s proposed TY MEC costs averaged over 2

the year 2021, and Table 2-3 summarizes PG&E’s estimate of MGCC, 3

levelized over the period 2021-2026. These tables show MGCs used to 4

develop rates in this GRC. Tables 2-4 and 2-5 provide average MEC for the 5

year 2025 and MGCC levelized over 2025-2030; these tables show MGCs 6

used to establish TOU periods (in Chapter 11). 7

(PG&E-2)

2-9

TABLE 2-2 MARGINAL ENERGY COST BY TOU RATE PERIOD

AND VOLTAGE LEVEL AVERAGED OVER YEAR 2021 (¢/KWH)

Line No. TOU Rate Period

Voltage Level

Transmission Primary

Distribution Secondary Distribution

1 Summer Peak 5.841 5.951 6.246 2 Summer Partial-Peak 4.378 4.460 4.681 3 Summer Off-Peak 2.783 2.836 2.976 4 Winter Partial-Peak 5.013 5.107 5.360 5 Winter Off-Peak 2.922 2.976 3.124 6 Spring Super Off-Peak (0.041) (0.041) (0.043) 7 Bundled Load-Weighted Annual Average 3.633

TABLE 2-3 LEVELIZED MARGINAL GENERATION CAPACITY COST

BY VOLTAGE LEVEL FOR 2021-2026 ($/KW-YEAR)

Line No. Period

Voltage Level

Transmission Primary

Distribution Secondary Distribution

1 2021-2026 Levelized Cost $102.66 $105.68 $112.01

TABLE 2-4 MARGINAL ENERGY COST BY TOU RATE PERIOD

AND VOLTAGE LEVEL AVERAGED OVER YEAR 2025 (¢/KWH)

Line No. TOU Rate Period

Voltage Level

Transmission Primary

Distribution Secondary Distribution

1 Summer Peak 5.922 6.034 6.332 2 Summer Partial-Peak 4.407 4.490 4.712 3 Summer Off-Peak 2.966 3.021 3.171 4 Winter Partial-Peak 5.337 5.437 5.706 5 Winter Off-Peak 3.230 3.291 3.454 6 Spring Super Off-Peak (0.876) (0.983) (0.937) 7 Annual Average 3.870

(PG&E-2)

2-10

TABLE 2-5 LEVELIZED MARGINAL GENERATION CAPACITY COST

BY VOLTAGE LEVEL FOR 2025-2030 ($/KW-YEAR)

Line No. Period

Voltage Level

Transmission Primary

Distribution Secondary Distribution

1 2025-2030 Levelized Cost $59.45 $61.19 $64.86

The remainder of this chapter provides more detail on the context and 1

costing methodology of PG&E’s MGC proposal. 2

B. Costing Methodology 3

1. Marginal Energy Costs 4

PG&E meets its energy needs by using a combination of resources 5

including: its own generation assets; energy purchased under a mix of 6

long-term power purchase agreements and generation resource contracts; 7

and short-term power market purchases. Consistent with prior CPUC 8

decisions, PG&E procures such resources through competitive solicitations, 9

as appropriate. 10

However, to meet its residual needs, PG&E buys residual energy from 11

the DA and RTM operated by the CAISO. The CAISO economically 12

dispatches resources that are bid into the markets. Because PG&E meets 13

its residual needs by purchasing power in the CAISO DA market, a forecast 14

of DA market prices provides the most relevant estimate of PG&E’s future 15

MECs.19 Thus, PG&E calculates MEC by forecasting hourly prices for 16

2021-2030 (the “forecast period”) based on historical CAISO prices at the 17

PG&E Default Load Aggregation Point (DLAP)20 adjusted by the factors that 18

impact the prices such as cost of natural gas and VOM. 19

19 As explained below, PG&E actually considers a Weighted Average (WAVG) hourly

price consisting of 80-90 percent of the day-ahead price plus 10-20 percent of the RT price (where the latter is averaged over sixty, five-minute intervals).

20 Prices at the PG&E DLAP represent the costs of PG&E’s load. By contrast, prices at the North of Path 15 and ZP26 hubs represent the prices paid to generators in PG&E’s service territory. There are four DLAPs in the CAISO’s territory, corresponding to the service territories of PG&E, Southern California Edison Company (SCE), San Diego Gas & Electric Company (SDG&E) and Valley Electric Association.

(PG&E-2)

2-11

a. What Drives Electricity Prices? 1

The hourly power price forecasting methodology is based on the 2

observation that except during periods of over-supply or load-shedding, 3

thermal (natural gas-fired) generators are almost always on the margin 4

in California. Thus, the hourly power price is a function of the residual 5

load that must be met by gas-fired generation, after accounting for 6

production from baseload (nuclear) and renewable (wind, solar, 7

geothermal, biomass, biogas, and hydroelectric generation (hydro)) 8

generators. These non-gas-fired generators generally have limited 9

dispatchability and are essentially “must run,” thus are rarely on the 10

margin. PG&E calls the residual load “Adjusted Net Load”. The ANL 11

starts with the CAISO’s Net Load calculation, which backs out wind and 12

solar.21 Then PG&E also subtracts nuclear, hydro and the other 13

renewables (geothermal, biomass and biogas), because all of these 14

displace thermal generation just as wind and solar do—and because the 15

model that uses ANL fits the price data better than one that uses just 16

Net Load. In D.17-01-006, the TOU Periods Order Instituting 17

Rulemaking decision, the CPUC specifically approved PG&E’s use of 18

ANL.22 19

21 Net load is equal to measured gross load minus utility-scale wind and solar production.

This is the quantity tracked daily on the CAISO’s Renewables Watch web page, at http://www.caiso.com/TodaysOutlook/Pages/default.aspx#Renewables and made famous as the “duck chart,” which the CAISO explains at http://www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf. ANL is equal to net load minus nuclear production, geothermal and biomass/biogas production, and a smoothed function of daily hydro production. ANL thus represents the residual load that must be met by non-hydro dispatchable resources such as thermal generators, imports and ES.

22 See D.17-01-006, Appendix 1, p. 1, Item 3.

(PG&E-2)

2-12

PG&E’s ANL model is a heat-rate based model23 that forecasts a 1

market heat rate and multiplies by a forecasted gas cost (which includes 2

the commodity cost, transportation cost and a GHG adder). Note that 3

while the power and gas prices in the model correspond to the PG&E 4

service territory, loads and generation correspond to the entire CAISO 5

footprint.24 6

b. Hourly Energy Price Forecast Methodology 7

Estimating hourly MEC is an important step in determining accurate 8

TOU periods and seasons. The highest and lowest MGC hours are 9

used to identify candidate peak and off-peak periods, as well as the 10

most appropriate definition of the summer season when costs and peak 11

rates are the highest. MEC is affected by the significant changes to the 12

hourly shape of ANL brought about by the increased penetration of 13

variable renewable resources such as wind and solar. Already in 2015 14

and even more by 2018, the peak of ANL (and also the highest MEC 15

hours) had shifted to later in the day during all seasons and months of 16

the year. These ANL peaks are expected to shift even later by 2021 17

and 2025 based on forecasts of increased renewables penetration. For 18

example, the peak of ANL in Summer 2021 and 2025 occurs in Hour 19

Ending (HE) 21 (as shown in Figures 2-1 and 2-2), whereas in 2018 the 20

Summer ANL peak occurred in HE 20 (as shown in Figure 2-10). 21

23 Heat rate is equal to energy price divided by gas cost, with units of Million British

Thermal Units per megawatt-hour (MMBtu/MWh). A low heat rate is better, with the most efficient combined cycle natural gas generators having heat rates close to 6,000 MMBtu/MWh, while an inefficient single cycle gas generator that comes on only during heat waves might have a heat rate of 10,000 or more. A heat rate-based model of marginal energy prices was used in Energy and Environmental Economics, Inc. (E3) Avoided Cost Calculator, referenced in the CPUC’s 2007 R.07-01-041 and updated regularly since, and a heat rate-based model of marginal energy prices was also used in PG&E’s 2015 Rate Design Window (RDW) Application (A.14-11-104) and 2017 GRC Phase II (A.16-06-013). E3 explains its methodology at https://ethree.com/documents/CPUC%20DR/AvoidedCostDocumentation.doc.

24 The appropriate prices to use for this application are at the PG&E DLAP. The PG&E DLAP prices are generally influenced by generation and load over the entire CAISO, rather than just in the northern region. For example, the majority of solar generation is located in the southern part of CAISO; in the absence of significant congestion, this results in all such resources displacing thermal generation and therefore influencing marginal heat rates and prices throughout the CAISO’s footprint.

(PG&E-2)

2-13

1) Hourly Marginal Energy Costs 1

PG&E calculates MECs by calculating hourly prices for the 2

forecast period based on a forecast of hourly market heat rates at 3

the PG&E DLAP adjusted by additional factors that impact the 4

prices such as the cost of natural gas and GHG allowances, and 5

VOM.25 PG&E forecasted hourly EMHR by modeling the 6

relationship between EMHR and CAISO-wide ANL and other 7

variables using publicly-available data; then forecasting how ANL 8

and the other variables will change over the forecasting period; and 9

finally applying the heat rate model and forecasted gas and GHG 10

prices to yield forecasts of energy prices over the forecast period. 11

As shown in Figures 2-1 through 2-6, the average EMHR 12

forecasted in 2021 and 2025 closely matches the shape of the ANL, 13

but has a very different shape from the average raw load, and also a 14

different shape from the average measured or gross load.26 15

Specifically, the EMHR curves have troughs (corresponding to low 16

marginal cost hours) up to two hours earlier than the troughs of 17

ANL, while the peaks of EMHR and ANL line up almost perfectly. 18

The figures also show “WECC-Adjusted” and “WECC and Battery-19

Adjusted” ANL, which account for the impact of temperatures and 20

(gross) loads in the rest of the WECC and modeled FTM battery 21

operations described in Section B.1.i. PG&E believes that the slight 22

offset between ANL and forecasted EMHR relates to the “ramp 23

25 PG&E uses an Effective Market Heat Rate (EMHR), which adjusts the standard heat

rate for the cost of GHG allowances and VOM. Details of the calculation are explained below.

26 Raw load (also called customer demand) is equal to gross (metered) load plus rooftop photovoltaics (PV). The reason the analysis starts with raw instead of gross load, is that the hourly shape of raw load (that is, the load that would be present in the absence of rooftop PV) is likely to be more stable over time than the hourly shape of the gross load.

(PG&E-2)

2-14

effect” and temperature effect described below.27,28 On the other 1

hand, the peaks of metered load and especially raw load occur 2

earlier in the day than for EMHR in summer, while the troughs for 3

metered and raw load occur in early morning in all seasons, rather 4

than mid-day or in late morning as for EMHR. 5

Note that these Figures show forecasted data using a “weather 6

year” of 2013 (i.e., 8,760 hourly “shapes” of loads and renewable 7

generation corresponding to 2013 weather, but grossed up to match 8

forecasted annual loads and renewable generation as of 2021 and 9

2025, respectively).29 10

27 The ramp effect tends to shift EMHR earlier than ANL, because the maximum upward

ramp in ANL occurs before its peak, while the maximum downward ramp in ANL occurs before its trough. WECC-wide effects are clearest in summer, where high summer loads in the rest of the WECC also pull the EMHR earlier, as discussed below. Note that while battery operations tend to flatten prices and the EMHR curve, they do not shift peaks and troughs much because the battery charge-discharge pattern syncs up to the EMHR pattern so as to maximize energy arbitrage revenues.

28 Because EMHR is more directly related to ANL than is energy price (because the latter also depends on the price of natural gas, which varies seasonally), PG&E here compares the shape of ANL to that of EMHR. However, the hourly shape of energy price is identical to that of EMHR, since gas prices do not vary within a day. PG&E thus concludes that in reference to Principle 3 of D.17-01-006, PG&E’s 2021 and 2025 forecasts reveal no “significant differences between estimated high and low marginal cost hours and the [adjusted] net load shapes.”

29 For PG&E’s 2015 RDW and 2017 GRC Phase II applications, PG&E used hourly shapes taken from the 2014 LTPP proceeding, which corresponds to a “weather year” of 2006. For this application, PG&E used ten weather years from 2005-2014 as scenarios, with shapes taken from SERVM inputs posted by the CPUC and used in the 2018 IRP. The 2013 shape yielded EMHR forecasts that were closest to the hourly averages calculated over all ten scenarios and is thus used in these illustrations. However, PG&E’s MEC and MGCC analyses used the entire set of ten scenarios to calculate marginal costs.

(PG&E-2)

2-15

FIGURE 2-1 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, SUMMER 2021 (JUN-SEP; 2013 WEATHER)

FIGURE 2-2 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, SUMMER 2025 (JUN-SEP; 2013 WEATHER)

(PG&E-2)

2-16

FIGURE 2-3 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, WINTER 2021 (OCT-FEB; 2013 WEATHER)

FIGURE 2-4 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, WINTER 2025 (OCT-FEB; 2013 WEATHER)

(PG&E-2)

2-17

FIGURE 2-5 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, SPRING 2021 (MAR-MAY; 2013 WEATHER)

FIGURE 2-6 FORECASTED AVERAGE RAW, METERED, NET, AND ADJUSTED NET CAISO LOAD AND EFFECTIVE MARKET HEAT RATE, SPRING 2025 (MAR-MAY; 2013 WEATHER)

The significant forecasted solar production leads to an obvious 1

peak in average ANL (and EMHR, and therefore also MEC) in the 2

(PG&E-2)

2-18

evening, and a minimum in all three quantities just before mid-day in 1

all seasons. 2

2) Methodology 3

For a region such as California, where thermal (gas-fired) units 4

are generally on the margin,30 the heat rate of the marginal natural 5

gas-fired generator usually sets the energy price. Thus, an effective 6

way to model energy prices is via the marginal heat rate times an 7

assumed gas price, with further adjustments for VOM and GHG 8

costs, and both a “soft floor” and a “hard floor” to represent 9

curtailment of renewable and some baseload generation. 10

To calculate the MECs during the forecast period, PG&E 11

combined a set of forecasted hourly market heat rates with a natural 12

gas cost forecast and VOM costs. Specifically, PG&E calculated 13

energy prices for northern California for each hour in the forecast 14

period as: 15

Max {-15, Min [ P(ANL + Curtailmax), Max (0, P(ANL))]} 16

where 17

P(x) = [Hourly Market Heat Rate Forecast(x)] [Gas Cost + GHG] + 18

[VOM]. 19

Curtailmax is the maximum amount of utility-scale generation that 20

is forecast to be curtailable at a price of zero, Gas Cost is the gas 21

30 Except for hours in which the ANL is less than the minimum generation level or greater

than the maximum generation level of on-line thermal generators (including imports), a marginal additional MW of load will generally be met by increasing the generation at the on-line thermal generator with the lowest marginal heat rate. When the ANL is greater than the maximum generation level of on-line thermal generators additional load may be met by reducing other load via demand response, or in extreme situations via load shedding. When the ANL is less than the thermal generators’ minimum generation level (a situation the CAISO calls “oversupply”), additional load may reduce the renewable curtailment that is required during oversupply to balance the CAISO system. The situation is complicated by transmission constraints, which can result in thermal generators being on the margin in some zones while renewables are on the margin in others, with DLAP-wide heat rates settling between zero and the marginal heat rate for an efficient thermal generator. Thermal generators are forecasted to be on the margin for more than 89 percent of the hours in 2021 and approximately 81 percent of the hours in 2025 (assuming that prices less than or equal to zero indicate DLAP-wide renewable curtailment, while prices above $200/MWh indicate demand response or load shedding).

(PG&E-2)

2-19

price forecast at the burner-tip (including transportation cost), and 1

GHG is forecasted emission costs, converted to dollars per MMBtu. 2

This function is illustrated in Figure 2-7 below. 3

FIGURE 2-7 ILLUSTRATIVE PRICE CURVE WITH CURTAILMENT PRICE AND FLOOR PRICE

In Figure 2-7, the blue curve shows the price function P(ANL) 4

while the black curve shows the function P(ANL + Curtailmax), which 5

is shifted to the left by Curtailmax. With lower and lower ANL, as the 6

original price curve reaches the “soft floor” of zero31 the price stays 7

31 In the first six months of 2018, the RT price at the PG&E DLAP was between -$1 and -

$19 over 2422 intervals (of five minutes each), or 4.65 percent of intervals. The RT price was lower than -$19 in only 37 intervals (0.07%), and between -$1 and $+1 in 2457 intervals (4.71%). The first six months of 2019 show a similar pattern, with RTM prices between -$1 and -$15 2.3% of the time; less than -$15 0.09% of the time; and between -$1 and $1 7.4% of the time. PG&E’s conclusion is that RT curtailment currently occurs primarily when the RT price reaches approximately zero; and that additional curtailment occurs as the RT price drops to approximately -$15/MWh. Note that the situation appears to have shifted since 2016, when the RT price was rarely close to zero but more often close to -$15/MWh. In the absence of other information, PG&E assumes that this soft floor price stays constant in the future; while the proposed expansion of the Energy Imbalance Market (EIM) to a DA market referenced in footnote 46 would likely result in similar behavior in the DA market.

(PG&E-2)

2-20

at the curtailment price until the limiting amount of curtailment is 1

reached; then the price follows the shifted price curve until the floor 2

price is reached. PG&E applied a price floor of -$15/MWh (negative 3

$15/MWh) to consider the fact that additional renewable generation 4

could be curtailed when prices go below -$15/MWh, and that over 5

the long term, the CAISO market would adjust if prices went 6

below that level for a significant number of hours per year 7

(e.g., precipitating the building of new ES devices to take advantage 8

of such low prices). 9

3) Hourly Market Heat Rate Calibration and Forecasts 10

a) Historical Effective Market Heat Rates 11

To calculate hourly Market Heat Rate Forecasts, PG&E first 12

calculated EMHR for every hour in the historical period from 13

January 1, 2012 through December 31, 201832 as: 14

EMHR = [P – VOM] / [G + GTR + GHG] 15

where 16

– P is historical Locational Marginal Price for that hour at the 17

PG&E DLAP from CAISO market price data;33 18

32 For PG&E’s updated filing on May 1, 2020, the full model described in sections B.1.c –

B.1.e was recalibrated using data through December 31, 2019. 33 Obtained from the CAISO’s Open Access Same-Time Information System (OASIS)

website, http://oasis.caiso.com. While the vast majority of short-term energy is transacted in the DA market, between 5 and 10 percent of the energy is transacted in the 5-minute RTM. Also, most of the curtailment of renewables occurs in the RT market due to both contractual considerations and uncertainty in DA forecasts of load and non-dispatchable renewables. At the same time, both solar and wind generation are significantly under-scheduled in the DA market (see slides 19-21 in CAISO’s presentation at the May 17, 2016 Market Performance Planning Forum, available at http://www.caiso.com/Documents/Agenda_Presentation_MarketPerformance_PlanningForum_May172016.pdf). Finally, while the DA prices result from a CAISO market algorithm that considers DA forecasts of load and generation, the MEC price model is calibrated to actual load and generation that occur in RT (i.e., not the forecasted values). Based on all these factors, PG&E used a WAVG of DA and smoothed RT prices to compute the EMHR (with the RT weight increasing linearly from 10 percent in 2012 to 20 percent in 2018). Using these WAVG prices rather than the DA price alone improved the R-squared of the fit with the Simple Model described in Section b) from 62.0 to 64.1 percent and reduced the MAE from $5.72 to $5.29.

(PG&E-2)

2-21

– VOM is the Variable Operation and Maintenance Costs for 1

the corresponding year;34 2

– G is the historical natural gas prices at a blend of PG&E 3

Citygate and SoCal Citygate from the Intercontinental 4

Exchange for the corresponding day;35 and 5

– GTR is the historical gas transportation rates (G-EG and 6

G-SUR tariff for PG&E; GT-TLS for SoCal Gas) to burner-tip 7

for the corresponding month.36 8

b) Modeling Effective Market Heat Rates – Simple Model 9

Traditionally, the EMHR has been shown to follow a “hockey 10

stick” relationship with the load: higher load results in the 11

dispatch of a resource with higher variable cost, and as the load 12

gets closer to the available capacity, inefficient single cycle 13

generators turn on and the marginal heat rate increases much 14

more rapidly. However, in fact, EMHR would be more 15

correlated with the amount of fossil resources (including 16

imports) needed to generate in order to meet the load. To 17

derive the proxy for that amount for the purpose of modeling 18

EMHR, PG&E defines ANL as net load (gross load minus utility-19

scale wind and solar generation) minus nuclear, biomass and 20

34 This was set from the 2018 VOM cost of a generic combined cycle generator, escalated

by 2 percent annually to get 2021 costs of $0.87/MWh. The costs and inflation rates are chosen from the 2019 California Energy Commission (CEC) Cost of Generation Report, available at https://ww2.energy.ca.gov/2019publications/CEC-200-2019-005/CEC-200-2019-005.pdf.

35 In the 2017 GRC Phase II, PG&E used only PG&E Citygate gas prices when calculating the EMHR. However, after noting that disruptions to gas supply in Southern California due to the Aliso Canyon gas leak appeared to impact energy prices statewide, PG&E modified the model to use a WAVG of transportation-adjusted prices at PG&E Citygate and Southern California Citygate. The weights were optimized along with other model parameters; the optimal weight for Southern California Gas prices turned out to be 7.5 percent.

36 PG&E gas transport rates are at http://www.pge.com/nots/rates/tariffs/GRF.SHTML. In the 2017 GRC Phase II, PG&E used only non-backbone gas transportation charges when calculating the EMHR. However, after these non-backbone charges increased dramatically in August 2016, PG&E modified the model to use a WAVG of (lower) backbone and non-backbone transportation charges. The weightings were optimized along with other model parameters; the optimal weight for backbone transportation charges turned out to be 28.4 percent.

(PG&E-2)

2-22

geothermal generation and a factor h times the 25-day rolling 1

average of hydro generation.37 The 25-day rolling average of 2

daily hydro generation is intended to capture the average 3

amount of hydro generation that is available for dispatch in each 4

hour, while the factor h (which is greater than one) captures the 5

additional impact of hydro generation due to its flexibility 6

(e.g., provision of A/S such as spinning reserve and regulation). 7

Figure 2-8 shows the relationship of EMHR to ANL over the 8

historical period January 1, 2012 through December 31, 2018. 9

The graph shows that the EMHR slowly increases with ANL 10

when it is below about 10 MMBtu/MWh, while it rapidly 11

increases with ANL when it is above 10 MMBtu/MWh. Also, the 12

functional shape is a little steeper as you go to the far left on 13

ANL. This observation leads to modeling EMHR as the sum of 14

an exponential function (that dominates in the steep right part) 15

and a cubic function (that dominates in the middle and left parts) 16

of ANL. Specifically, PG&E used the following functional form to 17

model EMHR one hour at a time (i.e., without including ramping 18

or other intra-day effects): 19

EMHR= b1·Exp(b2·ANLc) + a3·ANLc3 + a1·ANLc + a0 + e 20

where 21

ANLc = Centered38 ANL, calculated as 22

ANLc = L – RS – US – W – h·H – N – c 23

where 24

L = hourly raw load in the CAISO system (i.e., assuming no 25

rooftop solar generation) 26

37 CAISO lists large hydro and small hydro separately; in this analysis they are summed to

represent the total hydro resources dispatched within the CAISO footprint. The 25-day average is computed using the average of (24-hr average) and (24-hr maximum) hydro generation, to reflect hydro capacity available to provide A/S.

38 The center parameter acts to shift the S curve to the right or left to match the characteristics of the thermal fleet. It also allows the use of a cubic term without a corresponding quadratic, leads to faster calibration, and keeps the magnitude of the other coefficients in the range of 0 to +10.

(PG&E-2)

2-23

– RS = hourly rooftop solar generation in the CAISO system39 1

– US = hourly utility-scale solar generation in the CAISO 2

system 3

– W = hourly wind generation in the CAISO system40 4

– H = 0.5 * 25-day moving average of (24-hour average + 5

24-hour maximum) hydro generation in the CAISO system 6

– N = hourly nuclear plus biomass/biogas plus geothermal 7

generation in the CAISO system 8

– h and c are adjustable parameters 9

– e = Residual 10

39 In the historical dataset, neither L nor RS are available, only their difference (L-RS),

which is equal to gross or metered load. In the forecast, both L and RS are available, and they are tracked separately.

40 US includes imported solar generation, and W includes imported wind generation. However, so-called firmed and shaped renewables are not included in these amounts, as they have a different effective generation profile from as-generated renewables. Firmed and shaped wind and (if any) solar generation are categorized as unspecified imports in both historical and forecasted datasets.

(PG&E-2)

2-24

FIGURE 2-8 RELATIONSHIP OF EFFECTIVE MARKET HEAT RATES TO ADJUSTED NET LOAD OVER THE PERIOD JANUARY 2012 THROUGH DECEMBER 2019 (SIMPLE MODEL)

The Simple ANL Model explains most of the variation in 1

EMHR, which is Energy Price minus VOM, divided by Gas Price 2

plus GHG adder. As shown above, PG&E’s Simple Model 3

yields modeled heat rates (shown in red in Figure 2-8) that 4

generally fit the historical data (shown as blue marks in 5

Figure 2-8). 6

c. Ramping Effects 7

In PG&E’s 2015 RDW Application (A.14-11-014), PG&E noted that 8

thermal power plants require some time to ramp-up generation, and 9

typically need to be ramped up ahead of an increase in load.41 As the 10

41 This condition applies whether or not there is sufficient ramping capacity to meet the

maximum three-hour ramp; i.e., it applies both on “Flex need days” and other days.

(PG&E-2)

2-25

CAISO explains in its description of the Duck Chart, the CAISO needs to 1

dispatch resources to follow two upward and two downward ramps per 2

day, with the most significant ramp being the upward ramp close to 3

sunset (see Figure 2-9). 4

FIGURE 2-9 NET LOAD IN JANUARY SHOWING UPWARD AND DOWNWARD RAMPS(a)

_______________

(a) Taken from Figure 1 in CAISO’s explanation of the Duck Chart, available at http://www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf.

The upward ramps tend to increase hourly energy prices 5

(i.e., marginal costs) when the load that those generators need to meet 6

(i.e., the ANL) is increasing quickly. Likewise, downward ramps in ANL 7

tend to depress prices. The impacts of increasing solar generation and 8

of the ramp on prices can be seen in Figure 2-10, which shows average 9

load, generation and EMHR during the summers of 2012, 2015 and 10

2018. 11

(PG&E-2)

2-26

FIGURE 2-10 RELATIONSHIP OF LOAD, GENERATION AND RAMPS WITH EMHR

IN 2012, 2015 AND 2018 (SUMMER, JUNE-SEPTEMBER)

2012 2015 2018

First, considering the impact of increasing solar generation, the red 1

circles show the timing of the peak in metered, or gross load (which 2

already accounts for the impact of distributed generation, of which PV 3

solar is an increasing part). The timing of that peak in 2012 occurs right 4

at the beginning of the 4 p.m. to 9 p.m. period outlined in tall blue 5

rectangles and shifts later by approximately half an hour by 2015, and 6

another half hour later in 2018. The slight “rightward shift” in the peak of 7

gross load is caused by increasing amounts of distributed PV, which 8

offsets load during the middle of the day (as can be seen by the 9

changing shape of the red gross load curve from concave downward in 10

2012, to almost straight in 2015, to straight with a mid-day downward 11

kink in 2018). 12

A more dramatic shift later in the day applies to ANL, which is 13

highlighted by the orange triangles in Figure 2-10. The utility-scale solar 14

generation (shown in yellow) displaces increasing amounts of thermal 15

generation in 2015 and 2018 during the middle of the day, impacting the 16

shape of the ANL curve (in bright pink) and shifting its peak later by 17

about three hours, towards the end of the 4 p.m. to 9 p.m. period 18

outlined by the rectangles. 19

Now when considering prices, the peak of the average prices 20

(indicated by the green circle and light green curve) also shifts later in 21

(PG&E-2)

2-27

the day, but with a lag compared to ANL—by 2015, the peak of average 1

summer prices occurs close to an hour earlier than the peak of ANL, 2

though by 2018 the peak of average summer prices has shifted later 3

and lines up with the peak of ANL.42 Thus a model that does not 4

incorporate the impact of ramping could potentially show prices peaking 5

too late.43 6

PG&E expanded the model described in section b) to include 7

ramping by adjusting the centered ANL proportional to the lagged 8

three-hour ramp.44 The formula for centered ANL then becomes: 9

ANLcr = L – RS – US – W – h·H – N – c + r·(ANLt – ANLt-3) 10

where 11

– ANLt-3 is the ANL three hours prior to the current hour 12

– ANLcr is the centered, ramp-modified ANL 13

– r is an adjustable parameter 14

d. Intra-Day Spread 15

As discussed in footnote 44, including a ramping effect improves the 16

model fit both in terms of sum of squared error (SSE) and MAE. The 17

model fit also improves in terms of reproducing the average spread 18

42 Figure 2-10 provides evidence that, in reference to Principle 3 of D.17-01-006, there are

no “significant differences between estimated high and low marginal cost hours and the [adjusted] net load shapes” in recent historical data, at least for summer months. Charts provided in Workpapers indicate that the same conclusion applies to the winter and spring seasons.

43 PG&E took this into account in its 2015 RDW Application (A.14-11-014) and chose a 4 p.m. – 9 p.m. peak rather than the 5 p.m. – 10 p.m. peak that was indicated based on its forecasted marginal costs, partly because its price model in the 2015 RDW application did not incorporate ramping effects. With PG&E’s 2017 GRC Phase II application, PG&E explicitly modeled ramping effects, as discussed here; using a later forecast period with more solar generation the peak indicated by updated marginal costs turned out to be 5 p.m. – 10 p.m., as well. However, PG&E settled with other parties on a 4 p.m. – 9 p.m. peak for Non-Residential customers, in part to align with SCE and SDG&E and reduce the potential for customer confusion with PG&E’s 4 p.m. – 9 p.m. peak for Residential customers.

44 PG&E tested all combinations of ramp length from two to four hours, and all combinations of ramp timing from fully lagged (i.e., from current hour to hour t-l, where l is the lag time) to centered (i.e., from hour t + l/2 to t – l/2). For the calibration period of 2012-2018, the three-hour lagged ramp yielded the best combination of model fit in terms of SSE for EMHR and MAE for prices. Inclusion of a ramp effect improved R-squared from 64.1 percent for the Simple Model to 66.3 percent, and reduced MAE from $5.29 for the Simple Model to $5.04.

(PG&E-2)

2-28

between maximum and minimum price over a day. However, the 1

modeled prices were still substantially flatter during the day than actual 2

prices over the calibration period. This is likely due to the requirement 3

for thermal generators to either start at least once per day or “ride 4

through” a number of hours in which marginal costs are less than 5

revenue (which now can occur during the middle of the day). Thus, 6

energy prices (and therefore MECs) are elevated during the top ANL 7

hours of each day and depressed during the lowest ANL hours of each 8

day, compared to how costs would develop in the absence of startup 9

costs. 10

To incorporate this intra-day effect, the price model “boosts” the 11

EMHR by a constant b in the top hour of each day (when a peaking unit 12

is more likely to be on the margin than in other hours), with lesser 13

boosts in the second, third, etc., top hours. (The other parameters of 14

the model adjust to keep the mean the same when this effect is added.) 15

Mathematically, the formula for EMHR in section c) is modified as 16

follows: 17

EMHR= b1·Exp(b2·ANLcr) + a3·ANLcr3 + a1·ANLcr + a0 + b·d(rankANLcr -1) 18

+ e, where 19

– ANLcr is the rank-adjusted centered ANL described above, 20

– rankANLcr is the rank of the ANLcr in the current hour out of all 21

24 hours of the day, 22

– and b and d are adjustable parameters (where the decay parameter 23

d must be between 0 and 1). 24

Inclusion of the intra-day effect improved model fit, with R-squared 25

increasing from 66.3 percent (for the Simple Model plus ramping) to 26

68.1 percent, and MAE decreasing from $5.04 to $4.72. An additional 27

term in the objective, called the spread bias, also improved from 5.68 for 28

Simple plus ramping, to 2.93.45 29

45 The spread bias measures the bias in the average difference, or spread, between

prices in HE17-21 and prices in HE9-15. This metric is particularly important when modeling the EGM of ES, since much of the margin from storage comes from energy arbitrage, which is directly related to this price spread.

(PG&E-2)

2-29

e. Expanded Influence of Conditions in Rest of Western Electricity 1

Coordinating Council 2

In November 2014, the CAISO implemented a RT EIM, which 3

expands the effective footprint of the CAISO and increases the influence 4

of conditions in the rest of the Western Electricity Coordinating Council 5

(WECC) on prices (and thus MECs) in the CAISO. The EIM allows 6

linked entities to trade energy in 15-minute and 5-minute markets with 7

the CAISO (there is as yet no “Day-Ahead EIM,” which would constitute 8

a completely linked market, although the CAISO has recently 9

announced plans to establish one).46 While initially only CAISO and 10

PacifiCorp were members of the EIM, more Balancing Authorities have 11

joined every year since then, and the EIM now includes the majority of 12

load in the WECC. PG&E models the increasing linkage between the 13

CAISO and the rest of the WECC by adjusting ANL in two ways: based 14

on extreme temperatures in Seattle and Phoenix, and average 15

streamflows on the Columbia River at The Dalles. 16

1) Modeling Load Impacts on Imports Via Temperatures 17

Since its inception, a significant proportion of load in the CAISO 18

has been met by imports.47 What has changed with the inception 19

and growth of the EIM is the flexibility of imports, and the 20

relationship between imports and conditions in California and the 21

rest of the WECC. For example, while imports were mostly set DA 22

and via longer-term contracts before 2014 and generally followed 23

hourly blocks, imports (and now sometimes exports) can be 24

changed in 5-minute intervals in response to conditions within and 25

outside California. In particular, if loads (and energy prices) are 26

relatively high in California and low elsewhere in the WECC, imports 27

will increase to “follow the money,” and vice versa if loads are 28

relatively low in California and high elsewhere in the WECC. 29

One measure of the increased linkage between the CAISO and the 30

46 See https://www.utilitydive.com/news/caiso-to-develop-regional-day-ahead-market-for-

renewables-after-participant/563480/. 47 The average level of imports was 7,155 MW between October 2010 and

December 2019, compared to average CAISO net load of 23,299 MW over that period.

(PG&E-2)

2-30

rest of the WECC is the proportion of the maximum daily three-hour 1

ramp met by changes in imports—in 2012, 20 percent of the 2

three-hour ramp was effectively met by imports, while by 2019 the 3

proportion met by imports had increased to 42 percent. 4

To avoid having to model loads and generators in the entire 5

WECC, PG&E uses temperatures in Phoenix (for the Southwest) 6

and Seattle-Tacoma (for the Pacific Northwest (PacNW)) as proxies 7

for loads and prices in those two areas. Specifically, hourly 8

temperatures in Phoenix above a high-temperature threshold are 9

assumed to decrease imports (and thus increase the effective 10

CAISO load) using a quadratic impact equation. The same 11

quadratic equation (but with different coefficients) increases 12

effective CAISO load when Seattle-Tacoma temperatures exceed a 13

high-temperature threshold or fall below a low-temperature 14

threshold.48 15

Moreover, as more Balancing Authorities have joined the EIM, 16

the linkage between CAISO and surrounding areas has increased 17

(as evidenced by the steady increase in percent of ramp covered by 18

imports). The model considers this effect by increasing the 19

temperature effects starting from 2013 through 2019. The 20

expansion of the (Real-Time (RT) only) EIM market is also a 21

justification for increasing the weight of RT prices in later years, as 22

discussed in footnote 33, above. 23

Note that only daily maximum and minimum temperatures were 24

available for the temperature stations used in the model (PHX for 25

Phoenix, and SEA-TAC for Seattle-Tacoma); hourly temperature 26

effects were estimated from the current and prior day’s maximum 27

and minimum temperatures using empirically-derived weights, with 28

formulae shown in workpapers. Note that in Summer and Fall, the 29

temperature effect moves the modeled peak ANL, and thus price, 30

earlier (by less than an hour), because modeled peak temperature-31

48 A significant amount of heating in the PacNW is via electric resistance heating, so very

cold temperatures increase loads in that region; this effect does not appear to be as significant in the Desert Southwest.

(PG&E-2)

2-31

dependent loads in Phoenix and Seattle occur earlier than peak 1

ANL.49 2

The addition of these temperature effects improved the model fit 3

dramatically; R-squared measured over 2012-2018 increased from 4

68.1 percent to 81.5 percent, while weighted MAE decreased from 5

$4.72 to $4.12 and spread bias decreased from 2.93 to 1.89. 6

2) Modeling Impacts on Imports via River Flows 7

The second model improvement made to incorporate 8

WECC-wide effects is the addition of hydro impacts from the 9

PacNW. As with California-based hydro, wet years (and seasons) in 10

the PacNW tend to reduce prices in CAISO, while prices tend to 11

increase during dry periods. The capacity and annual energy 12

production of the PacNW hydro system is actually much larger than 13

that of California hydro resources;50 but its impact on prices is 14

muted by local demand and transmission constraints. 15

While the PacNW hydro system is extensive, covering several 16

states and southern British Columbia, most of the water released 17

from PacNW hydro projects ends up flowing through a point on the 18

lower Columbia River called The Dalles. PG&E does not have 19

access to hourly or daily generation data for the PacNW hydro 20

system (which is shared among many owners), so daily streamflows 21

at The Dalles are used as a proxy for daily hydro generation. Also, 22

the CAISO does not rely on A/S provided by PacNW hydro 23

generation, so PG&E believes that average generation (or flows) are 24

a more appropriate measure of impact on CAISO prices than 25

maximum daily generation (or flows). 26

The MEC model takes into account PacNW flows by subtracting 27

an (adjustable) factor times the 25-day lagged average flow at The 28

49 High or low temperatures affect customer demand (raw load) and are therefore

unaffected by BTM or FTM renewables production. This effect, plus the time zone difference between California and Arizona, mean that summertime WECC-wide load impacts peak earlier than ANL does.

50 Hydroelectric generators on the Columbia River and its tributaries account for 29 gigawatts of capacity, and amounted to 44 percent of the total hydro generation in the United States in 2012 (see https://www.eia.gov/todayinenergy/detail.php?id=16891).

(PG&E-2)

2-32

Dalles from the ANL. The addition of flows at The Dalles yielded a 1

more modest improvement than the addition of extreme 2

temperatures discussed above; incorporating this additional variable 3

improved the R-squared value from 81.5 percent to 81.8 percent 4

over the period 2012-2018, and reduced the MAE from $4.12 to 5

$4.05 and spread bias from 1.89 to 1.52. 6

f. Calibration 7

In order to find the parameters of the above model to best fit the 8

function to the historical EMHR and ANL data, PG&E used optimization 9

to solve for the parameters that minimize a combination of: (1) the sum 10

of squared residuals e from the above equation over the calibration 11

period (January 1, 2012 through December 31, 2019), (2) the MAE of 12

energy prices over the calibration period,51 and 3) the spread bias 13

(described above). 14

The fitted parameters yield the following formula: 15

EMHR = 1.769 x 10-2 · Exp (3.628 ANLcr) + 0.487 ANLcr 3 + 2.172 x 16

ANLcr + 7.950 + 2.963 · 0.753(rankANLcr -1) + e, with the additional 17

parameters 18

– c = 1.926 (center parameter) 19

– h = 1.19 (hydro factor) 20

– r = 0.087 (ramp coefficient) 21

– hPacNW = 0.741 (hydro factor for flows at The Dalles) 22

– SEAfac = 1.143 (factor for impact of Seattle temperatures) 23

– PHXfac = 1.39 (factor for impact of Phoenix temperatures) 24

– Weight on SoCal Citygate prices = 0.092 25

– Weight on PG&E backbone transport cost = 0.235 26

This model captures approximately 75 percent of the variability in 27

EMHR over the calibration period (i.e., its weighted R-squared is 28

75 percent), while the fitted prices capture 82.6 percent of the variability 29

51 Both the SSE and the MAE are calculated such that more recent data are weighted

more heavily than the earlier data. The exact formulae are provided in Workpapers, which consist of Excel spreadsheets with all formulae intact.

(PG&E-2)

2-33

in DA PG&E DLAP prices.52 The average calibration error, measured 1

as MAE over the calibration period, is $4.21. 2

The effect of including ramping, intra-day boost and balance-of-3

WECC effects in the model is illustrated in Figures 2-11 and 2-12. 4

Figure 2-11 is similar to Figure 2-8, except that in this calibration the 5

modeled EMHR can have multiple values for the same ANL; this is 6

evidenced by the spread in the red modeled points around a general 7

S-curve. Figure 2-12 is constructed differently—in that figure, each 8

modeled (red) point has the ramping and boost adjustment removed (so 9

all the red points now fall on a smooth curve); and each corresponding 10

“actual” (blue) point has the same amount subtracted. Thus, the 11

differences between modeled and actual EMHR are portrayed 12

accurately in Figure 2-12, but both are adjusted to remove the WECC-13

wide, ramp and boost effects so that the red curve is identical to that in 14

Figure 2-8. 15

52 Hourly energy prices show higher R-squared because gas prices are an excellent

explanatory variable for energy prices, and gas prices had significant variability during the calibration period. However, this variation also implies that calibrating based on minimizing the errors in EMHR can be more robust (insensitive to outliers) than calibrating based on prices themselves. Likewise, minimizing the MAE of energy prices is more robust than minimizing the SSEs of prices, again because it de-emphasizes outliers.

(PG&E-2)

2-34

FIGURE 2-11 RELATIONSHIP OF EFFECTIVE MARKET HEAT RATES TO ADJUSTED NET LOAD

OVER THE PERIOD JANUARY 2012 THROUGH DECEMBER 2018 (INCLUDING RAMPING AND INTRA-DAY BOOST EFFECTS)

(PG&E-2)

2-35

FIGURE 2-12 RELATIONSHIP OF RAMPING AND BOOST-ADJUSTED ACTUAL EMHR

TO ADJUSTED NET LOAD OVER THE PERIOD JANUARY 2012 THROUGH DECEMBER 2018

By way of comparison, an identical model that uses gross load 1

instead of ANL as an explanatory variable (and is re-calibrated to find 2

the optimal coefficients), captures only 49 percent of the variability in 3

EMHR and 61 percent of the variability in price, and its MAE over the 4

period 2012-2018 is approximately 70 percent worse at $6.81. The 5

same (recalibrated) model that uses net load instead of ANL captures 6

72 percent of the variability in EMHR and 80 percent of the variability in 7

(PG&E-2)

2-36

price, with MAE over 2012-2018 approximately 17 percent worse than 1

the ANL-based model at $4.75. 2

g. Forecast 3

In order to develop the MEC forecasts, PG&E applies the formula 4

obtained in the calibration to the ANL forecast data for the period 5

2021-2030.53 Annual totals for load and non-hydro renewables are 6

taken from the 2019-2020 IRP’s RSP,54 while 8,760-hour shapes are 7

drawn from the most recent ten weather years available in the SERVM 8

dataset from the 2017-2018 IRP (2005-2014).55 Forecasts of Gas 9

prices and GHG costs are taken from the 2019-2020 IRP RSP. These 10

inputs are used to create ten scenarios of ANL, EMHR, and Energy 11

Price. The final MEC is equal to the average price over all ten 12

scenarios. Details of the development of hourly forecast data for hydro 13

and nuclear generation, and treatment of hydro and other renewable 14

curtailment are as follows: 15

Hydro and WECC-Wide Adjustments: Hydro scenarios for each 16

weather year and month were “bootstrapped” using the closest 17

monthly hydro scenario available in CAISO recorded data 18

(i.e., between May 2010 and December 2018), using the following 19

procedure. 20

– First, for each month and year in a weather scenario prior to the 21

CAISO hydro record (i.e., between January 2005 and 22

April 2010), the monthly average CAISO hydro generation was 23

53 Forecasted energy prices from 2021-2030 are used in the MGCC model to calculate

annual ACCs, as described below. In addition, prices from 2025 are used to compute MEC to assess TOU periods, while prices from 2021 are used to set rates.

54 The IRP RSP is specified for 2020-2024, 2026, and 2030. Annual values were interpolated for intervening years. See Workpapers for this Exhibit, which PG&E presents in fulfillment of the directive in Principle 3 of D.17-01-006 to “[a]s part of its TOU period analysis, … submit the latest data and assumptions, including those vetted in the LTPP and/or IRP or successor proceeding.”

55 8760 SERVM shapes from the 2019-2020 IRP are also available from the IRP website. However, due to the extremely large datasets involved and the necessity of cutting those files into pieces to fit within Excel’s million-row capacity, there was insufficient time to process the updated files to develop new load and renewables shapes in time for this update. PG&E considers that any imprecision resulting from the use of the earlier load and renewables shapes is mitigated by the use of ten weather scenarios.

(PG&E-2)

2-37

estimated based on a factor times the PG&E hydro generation 1

in that month. 2

– Second, the closest average CAISO hydro generation in the 3

historical record (for that month) was determined. Hourly 4

CAISO generation from that closest month was used to fill in 5

data for months and years prior to the historical record, while 6

actual hourly generation was used for the weather scenarios 7

between May 2010 and December 2014.56 Details of the 8

calculations are provided in workpapers. 9

– Finally, daily flows at The Dalles were available for all dates in 10

2005 through 2014 (except for short periods in which flows were 11

interpolated), so were used directly. Similarly, historical 12

temperatures in 2005 through 2014 at PHX and SEA-TAC were 13

used directly in the forecast model. 14

Nuclear: For this input, PG&E assumed that each nuclear unit 15

would run at full capacity except during refueling outages. Refueling 16

outages were assumed to follow the same pattern of outages 17

observed over the calibration period, with approximately 21 months 18

between outages for each unit and each outage lasting 46 days. 19

Curtailment: Much of hydro generation and/or utility-scale wind and 20

solar generation is assumed to be curtailed when the hourly price 21

drops to zero, as described in footnote 31, above. Considering that 22

curtailment at a price of zero is economic for large hydro at risk of 23

spilling and in the absence of (confidential) bid data on other 24

renewables, the maximum amount that is subject to curtailment at a 25

price of zero is set equal to the modeled CAISO hydro production. 26

Further curtailment up to the total renewables output is assumed at 27

the modeled price floor of -$15/MWh. 28

56 Note that the model actually has a better fit (higher R-squared) when the hydro effect

is calculated using the factor h times a 25-day average of (daily production plus daily maximum production)/two rather than hourly hydro production. This is fortunate, because otherwise the model would have to forecast hourly hydro generation in 2021-2030 assuming weather that aligns with the forecasted load and renewables outputs, and prices in the day-ahead market that reflect a later peak than occurred in the historical record.

(PG&E-2)

2-38

h. Computing MECs at the Transmission Voltage Level by TOU Period 1

This process yields an hourly price forecast for wholesale market 2

energy prices in northern California. PG&E’s forecast of the 2021 3

average Northern California energy price is $33.10 dollars per MWh for 4

power delivered at the transmission voltage level. 5

PG&E estimated the MEC at the transmission voltage level for each 6

of PG&E’s six TOU periods as the average of the hourly electric price 7

forecasts weighted by the forecasted hourly 2021 PG&E bundled loads 8

(time-shifted so that weekends line up between hourly prices and hourly 9

loads). 10

i. Adjusting MECs for the Impact of Energy Storage 11

During the calibration period, the vast majority of ES in the CAISO 12

consisted of pumped hydro (Helms Pump-Generating Plant being by far 13

the largest, with generation capacity of 1,212 MW). Additions to the 14

existing ES fleet consisted of the 40 MW Lake Hodges pumped storage 15

facility, and approximately 150 MW of Li-Ion batteries, mostly in 16

Southern California. While the operation of this ES fleet changed as the 17

peak of energy prices moved later and the off-peak transitioned from 18

overnight to mid-day, PG&E considers that the main impact of overall 19

ES operations in the CAISO was to provide A/S and flatten energy 20

prices somewhat (i.e., to reduce the spread between peak and off-peak 21

prices), and that the magnitude of that impact did not change sufficiently 22

during the calibration period to warrant detailed modeling. 23

However, the amount of grid-connected ES forecasted to be 24

deployed over the next decade (4,916 MW by 2024, and 10,491 MW by 25

2030 in the 2020 RSP) is significantly more than the currently-installed 26

capacity of ES. PG&E therefore modeled the impact of ES on energy 27

and A/S prices as described below. 28

As discussed above, the main impact of ES on MECs is to flatten 29

prices, by reducing the amount of load that must be met by thermal 30

generators and imports when the storage is discharging (which will 31

generally be when prices are high), and increasing the load met by 32

thermal generators and imports when the storage is charging (which will 33

(PG&E-2)

2-39

generally be when prices are low). PG&E modeled this effect using an 1

iterative process. 2

In the first iteration, MECs for all scenarios and forecast years were 3

generated using the methodology described in previous sections. As 4

described in section B.2.a.3.b, these initial MECs were used to develop 5

corresponding A/S prices, and both energy and A/S prices were then 6

input to the StorageVET ES operations model and charge/discharge 7

time series developed for energy-only and full A/S operation modes. 8

In the second iteration, these charge/discharge time series were 9

scaled up separately for the energy only and full A/S modes, based on 10

the amount of ES assumed to be operating in each mode in each 11

calendar year.57 The scaled charge/discharge time series were added 12

to/subtracted from ANL to generate an updated set of MECs with 13

significantly flatter prices,58 and the updated MECs and updated A/S 14

prices input to StorageVET for a second run. 15

The same process was repeated for two more iterations, until the 16

standard deviations of the MECs had stabilized, indicating a steady 17

state in terms of variability in energy prices. However, the 2025-2030 18

MECs from the third iteration were actually higher on average in HE24 19

than HE23 for some months, indicating that ES charge/discharge time 20

series were not converging to reasonable schedules.59 The issue was 21

57 The proportion of ES operating for A/S in each year modeled in RESOLVE is assumed

to be equal to the ratio RegUp_Spin MW/(RegUp_Spin MW + Energy MW) where RegUp_Spin MW is the maximum hourly average of the sum of Reg Up and Spin awards in that year’s RSP RESOLVE output, and Energy MW is the same quantity for energy charge or discharge. Proportions are interpolated between RESOLVE years. In an effort to improve convergence of the iterations, the amount of ES capacity modeled in iteration 1 was reduced in later years, down to only 50% in 2030. The 2030 reduction was set to only 10% for iteration 2, and zero for subsequent iterations.

58 The MECs developed in the second iteration had approximately 18% lower standard deviations than in the first iteration for year 2021, and 19% lower for year 2025, showing that the modeled operation of ES indeed “flattened” the MECs. Average MECs were more stable, with the second iteration MECs on average 2.3% higher than first iteration MECs for 2021, and 0.7% higher for 2025.

59 For example, MECs in one iteration could be highest in HE20-23, leading ES to discharge fully in those hours and not at all in surrounding ones, resulting in lower prices in HE20 than 19 and HE23 than 24 in the next iteration. This behavior could continue in following iterations with the ES “flip-flopping” between iterations and leading to unrealistic schedules and therefore MECs.

(PG&E-2)

2-40

mitigated by averaging the charge/discharge time series from 1

StorageVET over iterations and between adjacent hours. The final set 2

of MECs were taken from the fifth iteration of this process. 3

j. Computing MEC at the Distribution Voltage Levels Using Energy 4

Line Loss Factors 5

PG&E estimates the MECs at the primary and secondary 6

distribution voltage levels for each TOU period by multiplying the 7

appropriate energy line loss factors and the MEC at the transmission 8

voltage level.60 The line loss adjustment is necessary because the total 9

amount of energy PG&E procures must be equal to the sum of: (1) the 10

marginal energy consumed by customers; plus (2) the amount of energy 11

that will normally be lost in the system due to line losses. Table 2-6 12

displays PG&E’s current energy line loss factors. Table 2-7 shows how 13

energy line losses are used to create the 2020 MEC by voltage levels. 14

60 The energy line loss factors are the same as those used in PG&E’s TO-20 rate filing at

Federal Energy Regulatory Commission (FERC).

(PG&E-2)

2-41

TABLE 2-6 TRANSMISSION AND DISTRIBUTION

ENERGY LOSS FACTORS

Line No. Location

Percent Loss Factor

Energy Loss Factor

Cumulative Loss Factor

Meter to Generation

Generation to Meter

From Source Documents

1 / (1 - Percent Loss Factor)

Product of Loss Factors at Each

Level

Inverse of Cumulative

Loss Factors

1 Generator Bus Bar 0.000% 1.0000 1.0000 1.0000 2 Generation Tie 0.185% 1.0019 1.0019 0.9982 3 High Voltage Transmission 1.777% 1.0181 1.0200 0.9804 4 Low Voltage Transmission 1.544% 1.0157 1.0360 0.9653 5 Primary Distribution Output 1.847% 1.0188 1.0555 0.9474 6 Secondary Distribution 4.715% 1.0495 1.1077 0.9028

_______________

Note: Source Documents: Transmission losses from May 14, 2010 “Transmission Loss Factors.” Distribution losses from “Distribution Loss Values for the TO-8 Filing.”

TABLE 2-7 CALCULATION OF 2021 MARGINAL ENERGY COST

BY TOU RATE PERIOD AND VOLTAGE LEVEL (¢/KWH)

Line No. TOU Rate Period

Transmission MEC

(Based on 2020 Hourly Market

Price Forecast)

Multiplied by Primary

Distribution Energy Loss

Factor

Primary Distribution

MEC

Multiplied by Secondary Distribution

Energy Loss Factor

Secondary Distribution

MEC

1 Summer Peak 5.841 × 1.0188 = 5.951 × 1.0495 = 6.246 2 Summer Partial-Peak 4.378 × 1.0188 = 4.460 × 1.0495 = 4.681 3 Summer Off-Peak 2.783 × 1.0188 = 2.836 × 1.0495 = 2.976 4 Winter-Partial 5.013 × 1.0188 = 5.107 × 1.0495 = 5.360 5 Winter-Off 2.922 × 1.0188 = 2.976 × 1.0495 = 3.124 6 Spring Super Off-Peak (0.041) × 1.0188 = (0.041) × 1.0495 = (0.043)

2. Marginal Generation Capacity Costs 1

a. Generic Capacity Price Forecast 2

1) Model and Data Transparency 3

The methodology used by PG&E to develop its generic capacity 4

price forecast61 is based on a sharable version of its internal 5

avoided cost model, and the open-source StorageVET model 6

61 Generic capacity refers to System or Local Capacity, as distinct from Flex Capacity.

Flex capacity is discussed in Section B 2 c, below.

(PG&E-2)

2-42

developed by the Electric Power Research Institute (EPRI). These 1

models and their inputs and outputs can be shared with all GRC 2

Phase II parties.62 3

2) Principles 4

The MGCC represents the cost of a marginal resource for 5

capacity. In previous GRC Phase II proceedings, the Commission 6

indicated a preference for calculating MGCC using a longer-term 7

perspective and adopted a six-year planning horizon. Using that 8

precedent, PG&E has calculated its MGCC by levelizing six years of 9

annual MGCCs (which are here referred to as annual ACC to 10

reduce confusion). Specifically, for 2021 marginal costs, the ACC is 11

levelized over the six-year period 2021-2026. For 2025 marginal 12

costs (which are used to evaluate TOU periods in Chapter 11), the 13

ACC is levelized over 2025-2030. 14

During a forecasted period, the type of marginal resource for 15

capacity can change, primarily when existing resources can no 16

longer meet the demand for capacity and thus the market will need 17

a new generation resource in that year (i.e., a switch from short-run 18

to long-run ACC).63 However, it is also possible that market forces 19

will incent developers to build more generation in a future year than 20

is required to meet a capacity need, in which case the type of 21

marginal resource could switch back from long-run to short-run 22

(since the new market-incented resources are presumably making 23

enough to cover their going-forward costs and are therefore not at 24

risk of retirement). 25

In alignment with past GRCs, the short-run ACC for each year is 26

estimated by the going-forward annual fixed costs of an existing 27

62 PG&E has designated its models used to forecast MEC and MGCC as proprietary

intellectual property, so only parties that have a signed NDA with PG&E can access and run those models.

63 The first year in which new build is required to meet capacity requirements is generally known as the year of need, or alternatively the RBY.

(PG&E-2)

2-43

marginal generation resource net of its “energy gross margins”64 for 1

that year. The long-run ACC for each year on and following the 2

RBY is estimated by the levelized RECCs that annualize the total 3

fixed costs of a new marginal generation resource net of EGM over 4

the assumed asset life of the resource. 5

During the six-year period 2021-2026, PG&E assumes that the 6

marginal resource will be new ES, evaluated using the long-run 7

methodology. However, PG&E forecasts that the marginal resource 8

will transition back to short-run in 2029, as described later in this 9

chapter. While the latter transition does not affect the MGCC for 10

rates (described in this chapter), it does affect the MGCC used to 11

inform TOU periods (described in Chapter 11). PG&E here 12

describes its reasoning regarding marginal generation units, and 13

calculations of ACC and MGCC for both 2021-2026 and 2025-2030, 14

in order to keep all related discussions in the same place. The 15

methodologies for calculating both short-run and long-run 16

ACC based on thermal and storage resources are described first, 17

followed by a discussion of the modeling results. The resulting 18

annual and hourly MGCCs for both 2021 and 2025 are 19

discussed last. 20

3) Energy Gross Margins 21

a) Thermal Resources 22

To calculate the short-run or long-run cost of capacity, the 23

EGM that a marginal resource is expected to earn from the spot 24

energy market(s) is subtracted from the fixed cost. To 25

determine the EGMs of a marginal thermal resource,65 PG&E 26

used an options-based analysis to reflect uncertainties present 27

64 EGM is the expected market revenue net of variable cost. The variable cost includes

fuel and start-up (for thermal generators), battery charging (for ES), and VOM cost. 65 There are four types of thermal resources that can be on the margin for capacity in

California: Steam Turbines; Combustion Turbines; Combined Cycle Gas Turbines (CCGT); and Reciprocating Engines. Each of these uses natural gas as its primary fuel, though some of them can also use other fuels in emergency situations. Co-Generation thermal resources also provide power to the CAISO grid, but are not considered marginal for capacity.

(PG&E-2)

2-44

in the market prices of power and gas. This analysis is similar 1

to the analysis in PG&E’s 2017 GRC Phase II application, 2

except that: (1) nominal forecasts of average power and gas 3

prices were updated to match the values used in the MEC 4

model above; (2) volatilities and correlations were updated to 5

the most recent available values; and (3) the hours for energy 6

blocks were updated to better match hours in which CCGT 7

resources are expected to be online (i.e., hours in which 8

average energy prices often meet or exceed variable costs for a 9

CCGT). Variable costs were based on fuel costs, start-up 10

costs66 and VOM costs. The hourly forward prices for energy 11

were obtained from the same methodology used to derive the 12

MEC described above. 13

A generation resource was assumed to earn net energy 14

market revenues (i.e., energy market revenues less variable 15

costs) only when the average electric energy price over a block 16

of hours is higher than the total variable costs over those 17

hours.67 The EGM for each year was derived by taking the 18

expected values of net energy market revenues during the year. 19

Because thermal resources obtain a very small portion of their 20

revenues from providing A/S, the EGM for thermal resources 21

was based only on energy revenues. 22

b) Energy Storage 23

In contrast, ES resources (i.e., FTM batteries) currently 24

obtain most of their revenue from providing A/S, with some 25

66 Startup cost, heat rate, and various other assumptions were updated from values used

in 2017 GRC Phase II based on 2019-2020 IRP RESOLVE model inputs. 67 In the 2017 GRC analysis, PG&E used two blocks: an on-peak block from 4 p.m. to

Midnight and an off-peak block from Midnight to the following day’s 4 p.m. The current analysis uses three blocks, with an on-peak block from 5 p.m. to 11 p.m.; an overnight block from 11 p.m. to 7 a.m.; and a mid-day block from 7 a.m. to 5 p.m. PG&E assumes that a CCGT would only incur a startup in a day if net revenues exceeded costs (including startup costs) in the on-peak block; and would then stay on during the overnight block if energy revenues in that block exceeded costs (this time without a startup cost). Likewise, the units could also stay on during the mid-day block if energy revenues exceeded costs; the latter situation occurred very rarely.

(PG&E-2)

2-45

additional revenue from “energy arbitrage” (charging during low 1

price periods and discharging during high price periods).68 2

Thus, estimating the EGM for ES resources must use different 3

methodologies and models than for thermal resources.69 4

First, a forecast of A/S prices must be developed that aligns 5

hour-by-hour with the MEC forecasts. Then the ES operation is 6

modeled using two “operating regimes:” (1) assuming that the 7

ES operates only for energy arbitrage; and (2) assuming that 8

the ES can sell up to its entire capacity in the A/S market, 9

modeled as co-optimizing revenue from the Energy and A/S 10

markets (just as the CAISO co-optimizes Energy and A/S costs 11

in the DA and RTMs).70 The EGM is calculated assuming full 12

A/S operation for 2021, and energy arbitrage only thereafter. 13

These steps are described in turn below. 14

i) Ancillary Services Price Forecasts 15

The A/S price forecast was developed using a new 16

PG&E model calibrated from CAISO DA Energy and 17

Ancillary Service prices from 2010 through June 2019. The 18

model for Regulation Up(Reg Up) derives from the 19

observation that Reg Up prices tend to follow energy prices 20

at high heat rates (because the opportunity cost for 21

providing Reg Up is essentially the amount that a generator 22

can earn in the energy market by increasing its output, 23

68 This can be seen from the pattern of battery charge/discharge in the CAISO’s Daily

Outlook page at http://www.caiso.com/TodaysOutlook/Pages/supply.aspx. While they sometimes follow price trends, batteries are mostly responding to a regulation signal.

69 As discussed below, ES that can provide RA is assumed to obtain revenue only from energy arbitrage, starting in 2022. Other ES units are still assumed to operate to provide A/S after 2022, but their operation affects MGCC only indirectly through the MEC calculation that considers both energy arbitrage and A/S operations.

70 CAISO has five types of A/S: Reg Up; Regulation Down (Reg Down); Spin; Non-Spin; and Regulation Energy Management (REM). However, only the first three types of A/S were modeled—revenues from Non-Spin are negligible for batteries in the CAISO, since Non-Spin prices are always less than Spin prices while batteries can always provide as much Spin as Non-Spin (having zero start-up times); and the CAISO has limited need for REM so batteries already online or projected to be online by 2021 will likely saturate the REM market, leaving the other A/S products as marginal.

(PG&E-2)

2-46

which is zero at low heat rates but tracks energy prices 1

above a thermal generator’s marginal heat rate).71 2

Specifically, for Reg Up, the A/S prices are modeled as: 3

PRUt = Max(0, RU_Adder + RU_Mult1*MECt) [if EMHRt < 4

RU_Thresh] 5

RU_Mult1*(MEC@RU_Thresh) + [otherwise] 6

RU_Mult2*(MECt – MEC@RU_Thresh)) 7

where 8

– PRUt equals the Reg Up price in hour t, 9

– MECt equals the DA Energy Price in hour t, 10

– EMHRt is the Effective Market Heat Rate corresponding 11

to MECt, 12

– RU_Adder is the modeled Reg Up price when PE equals 13

zero, 14

– RU_Mult1 < RU_Mult2 are slopes of a piecewise linear 15

curve of PRU vs. MEC, 16

– RU_Thresh is the EMHR threshold above which the 17

slope increases, and 18

– MEC@RU_Thresh is the DA Energy Price 19

corresponding to EMHR = RU_Thresh. 20

The coefficients for this model were calibrated 21

separately for the years: (1) 2010 and 2012-2015, and 22

(2) 2011 and 2016 – June 2019, with an objective function 23

that minimized the weighted MAE72 while matching average 24

prices during each set of years. The first set of years 25

represent periods in which A/S prices were low relative to 26

energy prices (due to dry or average hydro conditions, and 27

relatively low renewable generation), while the second set of 28

71 The heat rate discussion assumes a thermal generator. The same concept applies for

ES, except that the amount the ES can earn in the energy market is essentially equal to the energy price less the charging price divided by round-trip efficiency, plus VOM – but the situation is complicated by state of charge (SOC) limits.

72 The weighted MAE used the same weights as the MEC model (i.e., greater for later years).

(PG&E-2)

2-47

prices represent periods of high A/S prices relative to 1

energy prices (due to wet hydro conditions and/or high 2

renewable generation).73 3

This model captures 69 percent of the variability in 4

Reg Up prices, i.e., its R-squared coefficient is 0.69. 5

For Reg Down, prices are highest when energy prices 6

are low, because the opportunity cost for providing Reg 7

Down is the amount a generator can earn by reducing its 8

output. This leads to the following model for Reg Down 9

prices (where coefficients were again calibrated separately 10

for the years 2010 and 2012-2015, and 2011 and 11

2016 – June 2019): 12

PRDt = Max(0, RD_Adder + RD_Mult1*GasGHGt * 13

(RD_Thresh-EMHRt) [if EMHRt > RD_Thresh] 14

RD_Mult2*GasGHGt*(RD_Thresh-EMHRt)) [otherwise] 15

where 16

– PRDt equals the Reg Down price in hour t, 17

– GasGHGt equals the combined gas and GHG Price in 18

hour t, 19

– EMHRt is the Effective Market Heat Rate in hour t, 20

– RD_Adder is the modeled Reg Down price when 21

EMHRt equals RD_Thresh, 22

– RD_Mult1 < RD_Mult2 are slopes of a piecewise linear 23

curve of PRD vs. EMHR, and 24

– RD_Thresh is the EMHR threshold above which the 25

slope decreases. 26

This model captures 29 percent of the variability in Reg 27

Down prices, i.e., its R-squared coefficient is 0.29. 28

73 Large hydro is an excellent source of A/S (which is why the coefficient h in the MEC

model is greater than one, reflecting hydro’s large “bang for the buck”). However, during wet years, hydro generators often run close to maximum capacity to avoid spilling, reducing their ability to provide both upward and downward A/S. The need for A/S generally increases with (non-dispatchable) renewable penetration, as the CAISO must then balance a system with greater uncontrolled variability.

(PG&E-2)

2-48

Finally, the model for Spin prices considers that the 1

correlation between Spin and Reg Up prices is extremely 2

high (0.957 over the period May 2010 through June 2019). 3

Thus, the Spin model is defined in terms of the Reg Up 4

price, as follows: 5

PSPt = Max(0, SP_Mult1* PRUt) [if PRUt < SP_Thresh] 6

SP_Mult1*(SP_Thresh) + [otherwise] 7

SP_Mult2*( PRUt – SP_Thresh)) 8

where 9

– PSPt equals the Spin price in hour t, 10

– SP_Mult1 < SP_Mult2 are slopes of a piecewise linear 11

curve of PSP vs. PRU, and 12

– SP_Thresh is the EMHR threshold above which the 13

slope increases. 14

This model captures 76% of the variability in Spin 15

prices, i.e., its R-squared coefficient is 0.76. 16

To develop forecast scenarios of Reg Up, Reg Down 17

and Spin prices, the formulae specified above were 18

evaluated using as inputs the energy prices (MEC) and 19

EMHR from each year and weather scenario, developing 20

Reg Up forecasts first so that they could be used for the 21

Spin price forecast. 22

ii) Modeling Energy Storage Operations and EGM 23

To estimate annual EGM for a four-hour battery, MECs 24

and A/S price forecasts were input to EPRI’s public domain 25

StorageVET tool (Version 2.1).74 This Valuation Estimating 26

Tool co-optimizes energy and A/S revenues from DA and 27

optionally RT energy markets given prices in those markets 28

and input characteristics of the ES device. While the tool 29

can calculate overall financial results based on tax structure, 30

debt/equity and other parameters, in this case StorageVET 31

provided annual revenues and variable costs, and those 32

74 Documentation and login instructions available at StorageVET.com.

(PG&E-2)

2-49

data were input to PG&E’s MGCC model to maintain 1

consistency with prior GRC practice. 2

Three operating regimes were modeled: (1) Full A/S 3

operations (offering Reg Up and Down, and Spin as well as 4

Energy Arbitrage) with upper and lower SOC bounds of 75 5

percent and 25 percent; (2) Spin and Energy Arbitrage, with 6

upper and lower SOC bounds of 100 percent and 0 percent; 7

and (3) Energy Arbitrage only, with upper and lower SOC 8

bounds of 100 percent and 0 percent. The reason that 9

modeled SOC bounds are tighter when modeling regulation 10

is that the actual amount of charge/discharge in each hour 11

is unknown when providing regulation, so the forecasted or 12

“expected” SOC for the ES device must be kept close to its 13

mid-range to avoid getting outside the 0 to 100 percent 14

operational limits in practice. Operations are much more 15

predictable when the ES is only offering spin and/or energy 16

arbitrage, so the modeled SOC limits can be wider for those 17

operating modes. Estimated annual EGM turned out to be 18

higher in Full A/S mode than Spin plus Energy and Energy-19

only modes for all forecast years.75 20

While A/S price model coefficients were estimated from 21

both “Low A/S” years (2010 and 2012-2015) and “High A/S” 22

years (2011 and 2016 – June 2019), only the “Low A/S” 23

model was used in the EGM calculations. The reason is 24

that by 2021, approximately 1,550 MW of four-hour FTM ES 25

is expected to be installed within CAISO, rising to over 26

1,800 MW by 2023 (see Figure 2-13, from the June 17, 27

2019 CPUC Presentation to the IRP Modeling Advisory 28

75 CAISO tariffs also specify that generators must be able to generate at the maximum or

minimum limit of their regulation range for an hour, and also must be able to generate at the upper limit of their spin range for 30 minutes. These limits were modeled by StorageVET in A/S mode. In all modes, RA calls were modeled by assuming that the ES generated at full power (with no A/S provision) for four hours during the peak four hours of the 12 highest-load days of each year.

(PG&E-2)

2-50

Group on 2019-2020 IRP Modeling Assumptions).76 This 1

Figure shows the “baseline” assumptions (i.e., the amount 2

that is expected to be built irrespective of model outputs); 3

the IRP RSP indicates that ES build will continue after 2024, 4

so the baseline quantities (blue bars) represent a minimum 5

that is likely to be exceeded significantly by 2030. 6

FIGURE 2-13 FTM ES BASELINE ASSUMPTIONS IN 2019-2020 IRP

PG&E considers that by 2021, even the baseline 7

quantity of ES capacity exceeds the total amount of Reg Up 8

plus Spin currently procured by the CAISO,77 so even if 9

(upward) A/S requirements were to increase by 2021 as a 10

76 Slide 29 of the Webinar Presentation, available at

https://www.cpuc.ca.gov/uploadedFiles/CPUCWebsite/Content/UtilitiesIndustries/Energy/EnergyPrograms/ElectPowerProcurementGeneration/irp/2018/IRP_MAG_20190617_CoreInputs.pdf. Note that the 2020 value includes approximately 570 MW of ES procured by PG&E in response to Resolution E-4949; the majority of that ES has expected online date of December 2020, so its first full year of operation would be 2021.

77 Between November 1-15, 2019, CAISO requirements for Spin varied between 600 and 900 MW; requirements for Reg Down varied between 400 and 800 MW; and requirements for Reg Up varied between 300 and 500 MW (except on November 3, when the Reg Up requirement was 600 MW for the first four hours to provide a larger buffer during the transition from Daylight Saving Time). CAISO data retrieved on November 15, 2019 from http://oasis.caiso.com/mrioasis/logon.do.

(PG&E-2)

2-51

result of increased renewables build, the ES plus existing 1

hydro resources would provide more than enough capacity 2

to meet those needs without any contribution from thermal 3

resources. This implies that (1) by 2021, A/S prices as a 4

proportion of energy prices are likely to fall back to their 5

levels experienced during the Low A/S calibration period; 6

and (2) starting in 2021, ES will be able to sell a decreasing 7

fraction of its capacity into the A/S markets (i.e., those 8

markets will be over-supplied).78 As described in section 9

B.1.i, above, in this update PG&E used outputs from the 10

2020 IRP RSP to estimate the (declining) proportion of ES 11

providing A/S in 2021 through 2030, when modeling MECs 12

and resulting A/S prices. However, in accordance with the 13

CAISO’s Third Revised Straw Proposal, PG&E assumed 14

that starting in 2022, ES providing RA would receive energy 15

market rents only from (DA) energy arbitrage. 16

4) Short-Run Avoided Cost of Capacity for 2021-2030 17

a) Methodology 18

For each year from 2021-2030, PG&E estimated the 19

short-run ACC as the annual going-forward fixed cost of an 20

existing generation resource net of EGM. PG&E assumed an 21

existing inefficient CCGT plant is the higher cost, and therefore 22

marginal unit79 in years where existing units can meet capacity 23

78 Due to the limited amount of A/S procured by the CAISO, as ES capacity increases

after 2021 batteries will be forced to compete for shares of the lucrative regulation market, and will need to rely more and more on energy arbitrage revenues. As described in a recent Utility Dive article, saturation of regulation markets has already happened in the PJM market, and is expected to spread to other markets such as CAISO. Indeed, PA Consulting “projects a decline in ancillary service revenues from over $70/kW-year in 2021, to just over $10/kW-year by 2027, as California’s regulation and spinning reserve markets become increasingly saturated with new battery storage.” See https://www.utilitydive.com/news/new-battery-storage-on-shaky-ground-in-ancillary-service-markets/567303/ accessed November 17, 2019.

79 For existing resources, the highest-cost technology (in terms of residual cost) is considered the most at risk of retirement, and therefore sets the marginal capacity cost. However, if new build is required, the lowest-cost technology (again in terms of residual cost) is considered the first to be built, and therefore sets the marginal capacity cost.

(PG&E-2)

2-52

requirements. PG&E used public data sources, including the 1

CEC 2019 Cost of Generation Report,80 the 2011 Market Price 2

Referent,81 the 2014 LTPP assumptions,82 and inputs from the 3

2019-2020 IRP RESOLVE model83 for inputs required to model 4

an existing CCGT. 5

b) Fixed Cost of a CCGT 6

The Going-Forward Fixed Cost consists of Fixed O&M, 7

insurance and property tax. Insurance and property tax are 8

estimated based on the capital costs. To estimate an existing 9

CCGT’s capital cost, PG&E used $1,300/kW in 2016 dollars 10

from the IRP RESOLVE model inputs. To estimate Fixed O&M 11

for an existing CCGT in the future years, PG&E used 12

$10.39/kW-year in $2016 from RESOLVE. 13

c) Results 14

Table 2-8 shows components in deriving short-run costs of 15

capacity for each year from 2021 through 2030. The resulting 16

short-run costs of capacity vary between $33.60/kW-year and 17

$35.81/kW-year for 2021 through 2030, respectively. This 18

represents the cost of an existing CCGT’s fixed costs above and 19

beyond what it could earn in the energy market. 20

5) Long-Run Avoided Cost of Capacity for 2023-2030 21

a) Marginal Generating Unit and Methodology 22

For each year from 2021 to 2030, PG&E estimated the 23

long-run ACC as the RECC of going-forward fixed cost of a new 24

Lithium-Ion (Li-Ion) battery resource net of EGM. 25

80 https://ww2.energy.ca.gov/2019publications/CEC-200-2019-005/CEC-200-2019-

005.pdf. 81 ftp://ftp2.cpuc.ca.gov/PG&E20150130ResponseToA1312012Ruling/2011/10/SB_GT&

0607901.pdf. 82 http://www.cpuc.ca.gov/WorkArea/DownloadAsset.aspx?id=9144. 83 https://www.cpuc.ca.gov/General.aspx?id=6442459770 RESOLVE includes two types

of CCGTs, PG&E used the less efficient (and therefore higher cost) CCGT2 unit.

(PG&E-2)

2-53

Consistent with SB 100, the recent IDER decision,84 and 1

results from the 2019-2020 IRP cycle, PG&E considers that new 2

natural gas generation is unlikely to be built in California in the 3

future, even if its net cost were lower than that of ES. If new 4

natural gas generation is unlikely to be built, it would be 5

inappropriate to use a gas-fired CT (or CCGT) as the marginal 6

new-build unit in California (although PG&E believes that 7

existing gas-fired resources will still be required to meet both 8

energy and capacity requirements for the foreseeable future). 9

b) Input Data 10

For the four-hour Li-Ion battery system, PG&E used cost 11

data from the most recent Lazard Levelized Cost of Storage 12

Reports (V. 4.0 and V. 5.0),85 which aggregate cost and 13

operational data from original equipment manufacturers and 14

energy storage developers, after validation from additional 15

Industry participants/energy storage users. The ES assumed 16

15 years for both financing and lifetime, which is halfway 17

between the 10 years assumed in the 2018 EPRI report that 18

was used in PG&E’s November 22, 2019 testimony, and the 19

assumption in the 2019-2020 IRP RSP.86 Tax rates, 20

debt-equity ratio and other financial factors were taken from 21

RESOLVE inputs where possible, or other public sources as 22

specified in workpapers. Other operating parameters used the 23

defaults for the grid-scale battery case in StorageVET, except 24

that the amount of regulation “take” (i.e., the average RT energy 25

84 D.20-04-010, issued April 24, 2020. 85 See https://www.lazard.com/media/450774/lazards-levelized-cost-of-storage-version-

40-vfinal.pdf and https://www.lazard.com/media/451087/lazards-levelized-cost-of-storage-version-50-vf.pdf Lazard’s V 4.0 report was used in the 2019-2020 IRP. V 5.0 provides a more up-to-date snapshot but provides only current year estimates rather than a ten-year forecast; PG&E therefore used the cost decline trajectory from the V 4.0 forecast through 2030, adjusted to match the V 5.0 cost estimate for 2019.

86 The 15-year life is also the assumption in the National Renewable Energy Lab’s 2019 Energy Storage Cost Report, available at https://www.nrel.gov/docs/fy19osti/73222.pdf

(PG&E-2)

2-54

deviation from DA schedule divided by regulation award) was 1

doubled to approximately 25 percent.87 2

c) Results 3

Table 2-9 shows components in deriving long-run ACC for 4

each year from 2021 through 2030. The resulting long-run 5

costs of capacity decrease from $127.01/kW-year in 2021 and 6

$121.65/kW-year in 202288 to $51.25/kW-year in 2028 and 7

$32.69/kW-year in 2030. These values represent the amount of 8

a new Li-Ion battery’s fixed costs above what it could earn in the 9

energy market. 10

The ACC for new ES is less than the ACC for an existing 11

CCGT in 2030; this implies that a battery project that was built in 12

2030 and sold RA could more than cover its lifetime costs with 13

RA plus EGM. 14

The consequence of these falling ACC costs in the late 15

2020s is that, as discussed previously and shown in 16

Figure 2-14, below, the marginal ACC shifts back to the value 17

for an existing CCGT for 2030. Starting in 2030, existing 18

CCGTs are assumed to be the marginal units as they are at risk 19

of retirement, while ES is “taking off.”89 20

87 ES is excellent at providing regulation, and from PG&E’s experience is operated more

aggressively under a regulation signal than conventional resources. This effect is shown in Figure 12 on page 41 of PG&E’s Final Report from the EPIC 2.02 – Distributed Energy Resource Management System Project (January 18, 2019), available at https://www.pge.com/pge_global/common/pdfs/about-pge/environment/what-we-are-doing/electric-program-investment-charge/PGE-EPIC-2.02.pdf.

88 The ACC for ES built in 2021 is only slightly higher than for 2022 because the 2021-built project can recover lucrative rents from A/S as well as energy arbitrage in its first year of operation; in all years after 2022 the ACC for new Li-Ion batteries declines over time more rapidly because capital costs decline while EGM increases slowly from 2022 through 2030 (except for 2024-26, when the retirement of DCPP leads to declining EGM).

89 Other technologies, such as long-duration ES (e.g., pumped storage) could also start to provide capacity by the late 2020s. These resources were not modeled, as they would affect only the weighting of capacity costs in setting TOU periods (in Chapter 11), and not rates established in this proceeding.

(PG&E-2)

2-55

6) Computing Levelized MGCC for 2021-2026 and 2025-2030 and 1

Adjusting by Voltage Level With Line Loss Factors 2

To calculate levelized MGCC, PG&E used ACCs taken from the 3

two sources described above to calculate a Net Present Value 4

(NPV) sum of the six years of ACCs and then converted this NPV to 5

a levelized value using its after-tax WAVG Cost of Capital of 6

7.04 percent. The resulting levelized MGCC for 2021-2026 is 7

$102.66/kW-year, the calculation for which is shown in Table 2-10 8

(where values outlined in yellow are chosen as the ACC in each 9

year); the levelized MGCC for 2025-2030 is $59.45/kW-year, the 10

calculation for which is shown in Table 2-11. Figure 2-14 provides a 11

graphical illustration of the data in Tables 2-10 and 2-11. 12

FIGURE 2-14 ANNUAL AVOIDED CAPACITY COST FOR EXISTING CCGT,

LOCAL RA, AND NEW LI-ION BATTERY, 2021-2030

To estimate the MGCC at three voltage levels, PG&E began 13

with the levelized MGCC at the transmission level and then grossed 14

that number up by the appropriate demand line loss factor90 to 15

90 The demand line loss factors are the same as those used in PG&E’s TO-20 rate filing at

FERC.

(PG&E-2)

2-56

account for expected line losses. This calculation is similar to how 1

the MEC was developed for each voltage level. Table 2-12 displays 2

PG&E’s current demand line loss factors. Tables 2-3 and 2-5, 3

above, show how line losses were used to estimate MGCCs by 4

voltage level. 5

b. Planning Reserve Margin 6

Since June 2006, PG&E and other Load Serving Entities (LSE) have 7

been required to meet a 15 percent planning reserve requirement.91 As 8

a result, the marginal capacity cost of serving a given customer demand 9

level, as calculated in PG&E’s revenue allocation model, should include 10

an additional 15 percent amount of capacity that LSEs need to procure 11

to serve that demand level plus the associated planning reserve margin. 12

C. Conclusion 13

The MEC reflects the value of the marginal energy unit to customers, while 14

the MGCC reflects the value of the marginal capacity unit to customers. 15

Economic theory holds that pricing energy and capacity at the marginal unit cost 16

results in an efficient allocation among customers of both energy and capacity. 17

PG&E requests the Commission adopt the generation marginal cost 18

estimates proposed by PG&E in this testimony. 19

TABLE 2-8 SHORT-RUN COST OF CAPACITY BASED ON EXISTING CCGT FOR 2021-2030

91 See D.04-10-035, Conclusion of Law 4.

Line No. Year Insurance

Cost Property

Taxes Fixed O&M

Going-Forward Fixed Cost of a CC

(Sum of previous three columns)

Equity Cost

Debt Payment Taxes

Energy Gross

Margin

Net Capacity Cost(Sum of previous

five columns)

1 2021 7.74 16.65 11.47 35.86 - - - (1.69) 34.17 2 2022 7.89 16.40 11.70 35.99 - - - (1.23) 34.75 3 2023 8.05 16.13 11.93 36.11 - - - (1.17) 34.94 4 2024 8.21 15.84 12.17 36.22 - - - (0.98) 35.25 5 2025 8.37 15.54 12.42 36.33 - - - (1.19) 35.13 6 2026 8.54 15.21 12.67 36.42 - - - (1.00) 35.42 7 2027 8.71 14.87 12.92 36.50 - - - (0.82) 35.68 8 2028 8.89 14.51 13.18 36.57 - - - (0.65) 35.92 9 2029 9.06 14.13 13.44 36.63 - - - (0.53) 36.11 10 2030 9.25 13.72 13.71 36.68 - - - (0.44) 36.24

(PG&E-2)

2-57

TABLE 2-9 LONG-RUN COST OF CAPACITY BASED ON 4-HR LI-ION BATTERY FOR 2021-2030

TABLE 2-10 CALCULATION OF MARGINAL GENERATION CAPACITY COST FOR 2021-2026

($/KW-YEAR)

Line No. Component Unit 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

1 Full Li-Ion Capital Cost $/kW 1,150.34 1,058.31 973.64 903.33 845.11 797.22 758.25 727.07 702.84 684.88

2 Insurance $/kW-year - - - - - - - - - -

3 Property Tax $/kW-year - - - - - - - - - -

4 Fixed O&M $/kW-year 18.07 16.96 15.92 15.06 14.37 13.83 13.42 13.12 12.94 12.76

5 Going Forward Fixed Costs(Sum of Line 2 through 4) $/kW-year 18.07 16.96 15.92 15.06 14.37 13.83 13.42 13.12 12.94 12.76

6 Equity Cost $/kW-year 172.08 127.00 123.25 114.96 104.77 96.84 91.28 89.21 84.62 80.27 7 Debt Payment $/kW-year 54.11 50.77 47.65 45.09 43.03 41.40 40.16 39.28 38.73 38.19

8 Taxes $/kW-year 17.47 2.99 4.39 3.51 1.43 (0.17) (1.20) (1.20) (2.48) (5.13)

9 Energy Gross Margin (Including Ancillary Services) $/kW-year (134.72) (76.07) (85.38) (85.11) (81.05) (80.18) (82.46) (89.17) (91.26) (93.39)

10 Net Capacity Cost(Sum of Lines 5 through 9) $/kW-year 127.01 121.65 105.82 93.50 82.54 71.72 61.20 51.25 42.56 32.69

(PG&E-2)

2-58

TABLE 2-11 CALCULATION OF MARGINAL GENERATION CAPACITY COST FOR 2025-2030

($/KW-YEAR)

TABLE 2-12 PACIFIC GAS AND ELECTRIC COMPANY

TRANSMISSION AND DISTRIBUTION DEMAND LOSS FACTORS

Line No. Location

Percent Loss

Factor Demand Loss

Factor Cumulative Loss Factor

From Source

Documents

1/ (1 – Percent Loss Factor)

Meter to Generation

Generation to Meter

Product of Loss Factors at Each

Level

Inverse of Cumulative Loss

Factors

1 Generator Bus Bar 0.000% 1.0000 1.0000 1.0000 2 Generation Tie 0.211% 1.0021 1.0021 0.9979 3 High Voltage Transmission 2.061% 1.0210 1.0232 0.9773 4 Low Voltage Transmission 1.946% 1.0198 1.0435 0.9583 5 Primary Distribution Output 2.852% 1.0294 1.0741 0.9310 6 Secondary Distribution 5.651% 1.0599 1.1385 0.8784

_______________

Note: Source Documents: Transmission losses from May 14, 2010 “Transmission Loss Factors.” Distribution losses from “Distribution Loss Values for the TO-8 Filing.”

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 3

GENERATION ENERGY AND CAPACITY COST OF SERVICE

(PG&E-2)

3-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 3

GENERATION ENERGY AND CAPACITY COST OF SERVICE

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 3-1

B. Overview of Methodology ................................................................................. 3-2

1. Definition of NEM ....................................................................................... 3-4

2. Definition of TOU Periods .......................................................................... 3-4

3. Load Profile Differences and their Alignment with ANL .............................. 3-4

C. Concepts of “Cost” and “Benefit” ...................................................................... 3-8

1. Illustration of Energy Related “Cost” and “Benefit” for NEM Load Profiles ............................................................................................ 3-10

2. Illustration of Energy Related “Cost” and “Benefit” for Non-NEM Load Profiles ............................................................................................ 3-13

3. Illustration of Generic Capacity Related “Cost” and “Benefit” ................... 3-17

D. Generation Cost of Service Results................................................................ 3-18

1. Generation Marginal Energy Cost Revenue Per kWh .............................. 3-19

a. Annual Level Comparison of Average GMECR/kWh ......................... 3-19

b. Comparison of Average GMECR/kWh in Summer ............................ 3-21

c. Comparison of Average GMECR/kWh in Winter ............................... 3-23

2. Marginal Generation Capacity Cost Revenue Per kWh ........................... 3-26

a. Annual Level Comparison of Average MGCCR/kWh ......................... 3-26

b. Comparison of Average MGCCR/kWh in Summer ............................ 3-27

c. Comparison of Average MGCCR/kWh in Winter ............................... 3-29

3. Total Generation Marginal Cost Revenue Per kWh ................................. 3-31

a. Annual Level Comparison of Total GMCR/kWh ................................. 3-31

b. Comparison of Total GMCR Per kWh for Summer TOU Periods ...... 3-33

c. Comparison of Total GMCR per kWh Winter TOU Periods ............... 3-35

4. Cost of Service Comparison by Customer Class ..................................... 3-37

(PG&E-2) PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 3 GENERATION ENERGY AND CAPACITY COST OF SERVICE

TABLE OF CONTENTS

(CONTINUED)

3-ii

a. Cost of Service Differences Within Residential Customer Class ....... 3-37

b. Cost of Service Difference Within Commercial Customers ................ 3-38

c. Cost of Service Difference Within Agricultural Customers ................. 3-40

d. Cost of Service Difference Within Large Commercial and Industrial Customers .......................................................................... 3-41

e. Additional Explanation for the Difference in GMCR/kWh between Non-NEM and NEM ........................................................................... 3-41

E. Conclusion ...................................................................................................... 3-43

(PG&E-2)

3-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 3 2

GENERATION ENERGY AND CAPACITY COST OF SERVICE 3

A. Introduction 4

In this chapter, Pacific Gas and Electric Company (PG&E) describes its 5

generation energy and capacity cost of service methodology and the results 6

presented in terms of Marginal Cost Revenue (MCR) per kilowatt-hour (kWh). 7

PG&E’s improved generation energy and capacity cost of service analysis 8

estimates cost of service for Net Energy Metering (NEM)1 and Non-NEM 9

customer sub-groups, which leads to more accurate cost of service results 10

overall. PG&E requests the California Public Utilities Commission’s (CPUC or 11

Commission) approval of the methodology and results presented herein. 12

Implementing an improved generation cost of service framework is critical 13

because of the way PG&E’s customers’ load profiles have been evolving with 14

the adoption of technologies such as solar Photovoltaic (PV) and storage, which 15

carries flow of electricity not only from utility grid to the customer service point 16

(called “delivered” flow), but also from the customer service point back to the 17

grid (called “received” flow). The following points summarize the impacts in 18

this regard: 19

1. Technology adoptions have led to a change to load shapes, including shift of 20

peak hours; 21

2. NEM customer load profiles are significantly different than Non-NEM 22

customers due to solar generation from NEM customers during the day; 23

3. Generation from solar PV causes the net load to decrease significantly 24

during the day, but the hourly load during the system peak hours (occurring 25

in the late afternoon and evening) remains the same and is little impacted 26

due to the generation from solar PV; and 27

4 For NEM customers, the average MCR per kWh increases due to relatively 28

higher load during the costlier hours, while load decreases during the 29

significantly less expensive hours of a day due to the adoption of solar PV. 30

1 NEM refers to Net Energy Metering customers having on-premise generation

resources, such as solar PV.

(PG&E-2)

3-2

The generation energy and capacity cost of service methodology is 1

discussed in detail in the next sections. 2

B. Overview of Methodology 3

The metric proposed by PG&E to analyze cost of service and subsequently 4

perform revenue allocation is MCR per kWh. The appropriateness of using 5

MCR per kWh to analyze cost of service has been discussed in Chapter 1, 6

Section A of Exhibit (PG&E-2). MCR is obtained by multiplying the marginal 7

cost by the appropriate cost driver. There are three components of the 8

generation marginal costs, described in detail in Chapter 2 of Exhibit (PG&E-2), 9

that also apply to the generation cost of service analysis: Marginal Energy 10

Costs (MEC), generic Marginal Generation Capacity Costs (MGCC), and flexible 11

MGCCs.2 Therefore, the Generation Marginal Cost Revenue (GMCR) is 12

obtained by adding three revenue components: Generation Marginal Energy 13

Cost Revenue (GMECR); generic Marginal Generation Capacity Cost Revenue 14

(MGCCR); and flexible MGCCR. Figure 3-1 shows the steps involved in 15

generation cost of service analysis.3 16

2 As defined in Chapter 2 of Exhibit (PG&E-2), “generic” capacity is where the peak

demand is defined as the systemwide or local demand and “flexible” capacity is where the demand is measured based on the three-hour positive ramp in system net load.

3 Flexible MGCCs are forecasted to be very small compared to MEC and generic MGCCs, comprising an insignificant portion of the total generation cost of service. For that reason, going forward the term MGCC in this chapter refers to generic MGCC components, unless specified otherwise.

(PG&E-2)

3-3

FIGURE 3-1 GENERATION COST OF SERVICE ANALYSIS STEPS

The GMECR for a bundled4 customer class is calculated by multiplying the 1

hourly MEC by the corresponding hourly load of that bundled customer class. 2

This calculation is done for each customer class, and, within each customer 3

class, separately for NEM and Non-NEM customers. 4

The hourly MGCCR is calculated by using the hourly percentage allocator 5

developed based on the California Independent System Operator (CAISO) 6

corporation’s hourly Adjusted Net Load (ANL) profile.5 The generation Peak 7

Capacity Allocation Factor (PCAF) method6 used by PG&E in its 2017 General 8

Rate Case (GRC) Phase II filing is used for this purpose. It provides hourly 9

percentage allocators for the top ANL hours that fall within 80 percent of annual 10

peak hourly ANL. For the remaining hours, the percentage allocator is zero, 11

i.e., no cost is allocated to those hours. The annual MGCC, expressed in 12

$/kilowatt-year (kW-year), is multiplied by the hourly percent allocator and the 13

corresponding hourly load to get hourly MGCCR. Hourly MGCCR is summed up 14

by customer class and by time of use (TOU) period separately for the customers 15

with and without NEM. Understanding of the overall generation cost of service 16

4 Only bundled PG&E customers are considered for the generation Customer Opinion

Survey analysis. 5 The ANL is defined as the gross load minus the renewable (solar, wind, biomass and

geothermal), hydro and nuclear generation resources. 6 This method has been presented in Attachment A of this chapter.

(PG&E-2)

3-4

comes from analyzing GMCR per kWh for customers with and without NEM 1

within customer classes, and during different TOU periods. The “Generation 2

Cost of Service Results” section in this chapter presents PG&E’s cost related 3

findings and form the basis of PG&E’s request to the Commission to approve the 4

methodology used in calculating the cost of service. 5

1. Definition of NEM 6

PG&E has used the NEM indicator from billing data to identify NEM 7

customers. The different components of NEM customers are described in 8

detail in Chapter 1. Streetlights and Standby customer classes are excluded 9

from the analysis because there are no NEM customers in these 10

customer classes. 11

2. Definition of TOU Periods 12

TOU periods are the same for all rate classes in this cost of service 13

comparison analysis. Summer months are June through September, and 14

the rest of the months in a year are considered winter months. Both 15

summer and winter have a peak period from 4 p.m.to 9 p.m. Summer has 16

two part-peak periods: from 2 p.m. to 4 p.m. and from 9 p.m. to 11 p.m. 17

Winter has a super-off-peak period during months March through May 18

starting at 9 a.m. and ending at 2 p.m. The rest of the hours are considered 19

off-peak periods in both summer and winter. 20

3. Load Profile Differences and their Alignment with ANL 21

The CAISO’s ANL is the main driver of generation costs.7 Therefore, a 22

customer class’s cost of service depends on how that customer class’s load 23

profile aligns with the ANL profile. 24

In this section, load profile differences between NEM and Non-NEM 25

bundled customers during summer, winter, and spring months (using 2017 26

recorded data) are discussed. Figures 3-2A and 3-2B show the hourly 27

average load profile during a summer month (July) for weekdays and 28

7 Please refer to Chapter 2, Section B.1.a.

(PG&E-2)

3-5

weekends respectively for PG&E’s bundled customers.8 For purposes of 1

comparison, the load profiles have been normalized.9 In addition, the ANL 2

at the hourly level has been shown in each of these figures. The ANL profile 3

provides insight as to when the MECs and MGCCs are relatively high, 4

because higher ANL generally results in a higher cost. The ANL is the 5

highest during hour ending 18 through hour ending 22, that is, from 5 p.m. to 6

10 p.m., for both weekdays and weekends. NEM load profiles peak during 7

these hours as well, while the decrease in load due to NEM generation 8

usually happens earlier in the day. Non-NEM load profiles during both 9

weekdays and the weekends are significantly different than the 10

corresponding NEM load profiles. When evaluated along with ANL, PG&E 11

has found that NEM load profiles are costlier on average than the Non-NEM 12

during the summer months. The comparison of bundled NEM and 13

Non-NEM load profiles for weekdays and weekends during a winter month 14

(December) are shown in Figures 3-3A and 3-3B. The resulting average 15

cost of NEM load profile is higher than the corresponding Non-NEM load 16

profile in winter months as well. 17

The comparison of NEM load profiles and Non-NEM load profiles for a 18

spring month (April) for weekdays and weekends are shown in Figures 3-4A 19

and 3-4B, respectively. Here again, similar load profile differences can be 20

seen between NEM and Non-NEM. The spring months experience 21

significantly higher over-generation compared to summer and winter 22

months, and it is even more pronounced during the weekends as can be 23

seen from the range of variation in normalized NEM load during the day. 24

The scale used during the weekend is -8 to +8 which is much wider than the 25

scales for NEM load profiles for other scenarios. 26

The estimated costs are presented later in this chapter. 27

8 Streetlight customers have been excluded from these figures since there are no NEM

customers within the Streetlight class. Additionally, Streetlights customer load profiles have very different drivers, given that Streetlight load is on in the nighttime rather than the daytime, when compared to other customer classes.

9 Normalization is performed at the daily level, i.e., hourly loads are divided by the total daily load to obtain normalized load.

(PG&E-2)

3-6

FIGURE 3-2A BUNDLED LOAD SHAPE IN A SUMMER MONTH (JULY), WEEKDAYS

FIGURE 3-2B BUNDLED LOAD SHAPE IN A SUMMER MONTH (JULY), WEEKENDS

(PG&E-2)

3-7

FIGURE 3-3A BUNDLED LOAD SHAPE IN A WINTER MONTH (DECEMBER), WEEKDAYS

FIGURE 3-3B BUNDLED LOAD SHAPE IN A WINTER MONTH (DECEMBER), WEEKENDS

(PG&E-2)

3-8

FIGURE 3-4A BUNDLED LOAD SHAPE IN A SPRING MONTH (APRIL), WEEKDAYS

FIGURE 3-4B BUNDLED LOAD SHAPE IN A SPRING MONTH (APRIL), WEEKENDS

C. Concepts of “Cost” and “Benefit” 1

PG&E’s methodology for allocating generation energy and capacity costs in 2

this filing is similar to what was used in its 2017 GRC Phase II filing, except that 3

the “costs” and “benefits” have first been estimated at the hourly level and 4

aggregated at the TOU period level for NEM and Non-NEM customers, 5

separately. 6

(PG&E-2)

3-9

As shown in Figure 3-5 below, separate rules apply for energy and capacity 1

for the purpose of assigning “cost” and “benefit”. For energy, MEC can either be 2

positive or negative. During a particular hour, if the MEC is positive, then the 3

“delivered” flow of a customer class will result in a “cost” for that customer class. 4

In addition, if the MEC in a particular hour is negative (as frequently seen to 5

occur from March through May), and the load of a customer class is “received” 6

for that hour (i.e., electricity is flowing back to the grid), then the negative MEC 7

also results in a “cost.” On the contrary, if the MEC is negative for an hour while 8

the load of a customer class is “delivered,” the negative MEC will result in a 9

“benefit” for that customer class. Also, during a positive MEC hour, a customer 10

class’s “received” load will result in a “benefit” for that customer class. 11

For capacity-related “cost” and “benefit,” the rules are a bit different than that 12

of energy as depicted in Figure 3-5. PCAF allocated costs for an hour are either 13

zero (no cost allocated) or positive. Hence, the “delivered” load of a customer 14

class will result in a “cost” for that customer class, and the “received” load will 15

result in a “benefit” for that customer class. 16

FIGURE 3-5 “COST” AND “BENEFIT” ASSIGNMENT RULES

(PG&E-2)

3-10

1. Illustration of Energy Related “Cost” and “Benefit” for NEM 1

Load Profiles 2

This section demonstrates how positive or negative load, combined 3

with positive or negative MECs, can result in either a cost or a benefit 4

(i.e., negative cost) at the hourly level during a typical summer day, winter 5

day and spring day. Because negative load cannot occur from a Non-NEM 6

customer, the focus here is on NEM customers. 7

The figures below illustrate the costs and benefits from both positive and 8

negative load for different seasons, and for both residential and 9

non-residential NEM customers. 10

Figures 3-6A and 3-6B show a representative summer day (July 14, 11

2017). Here, because MEC is positive during all hours, the cost is negative 12

during the hours when load is negative, i.e., electricity flows back to the grid. 13

This occurs during hour ending 10 through hour ending 16 for residential 14

customers, and during 11 through 16 for non-residential customers. 15

Negative costs are considered “benefit.” 16

Figures 3-7A and 3-7B show a representative winter day (February 10, 17

2017). Similar to the summer day, the MEC is positive during all hours; cost 18

is, therefore, negative during the hours when load is negative, i.e., electricity 19

flows back to the grid. This happens during hour ending 12 through hour 20

ending 15 for residential customers. For the non-residential customers, cost 21

is positive for all hours. 22

Figures 3-8A and 3-8B show a representative spring day (April 3, 2017). 23

It is important to note that MEC is negative during hour ending 12 through 24

hour ending 16. Because load is also negative during these hours, the cost 25

is positive. This applies to both residential and non-residential customers. 26

For all other hours where MEC is positive, cost is positive when load is 27

positive, and negative when load is negative. 28

(PG&E-2)

3-11

FIGURE 3-6A COST AND BENEFIT HOURS OF RESIDENTIAL NEM LOAD PROFILE ON A SUMMER DAY

FIGURE 3-6BCOST AND BENEFIT HOURS OF NON-RESIDENTIAL NEM LOAD PROFILE ON A SUMMER DAY

(PG&E-2)

3-12

FIGURE 3-7A COST AND BENEFIT HOURS OF RESIDENTIAL NEM LOAD PROFILE ON A WINTER DAY

FIGURE 3-7BCOST AND BENEFIT HOURS OF NON-RESIDENTIAL NEM LOAD PROFILE ON A WINTER DAY

(PG&E-2)

3-13

FIGURE 3-8A COST AND BENEFIT HOURS OF RESIDENTIAL NEM LOAD PROFILE ON A SPRING DAY

FIGURE 3-8B COST AND BENEFIT HOURS OF NON-RESIDENTIAL NEM LOAD PROFILE ON A SPRING DAY

2. Illustration of Energy Related “Cost” and “Benefit” for Non-NEM 1

Load Profiles 2

This section focuses on Non-NEM customers, and demonstrates how 3

positive load, combined with positive or negative MEC gives either cost or 4

benefit (i.e., negative cost) at the hourly level during a typical summer day, 5

(PG&E-2)

3-14

winter day and spring day. It is important to note that, unlike NEM load, 1

Non-NEM load is always positive. 2

The figures below illustrate the costs and benefits from the positive load 3

for different seasons, and for both residential and non-residential Non-NEM 4

customers. 5

Figures 3-9A and 3-9B show a representative summer day (July 14, 6

2017). Because MEC is positive during all hours, and their load is also 7

positive, their costs are positive for all hours of the day. This is true for both 8

residential and non-residential Non-NEM customers. The same observation 9

can be made from Figures 3-10A and 3-10B, which illustrate costs during a 10

typical winter day (February 10, 2017). However, during a typical spring day 11

(April 3, 2017), the MEC is negative during hour ending 12 through 16. 12

Therefore, cost is negative during these hours, as can be seen in 13

Figures 3-11A and 3-11B. 14

FIGURE 3-9A COST AND BENEFIT HOURS OF RESIDENTIAL NON-NEM LOAD PROFILE ON A SUMMER DAY

(PG&E-2)

3-15

FIGURE 3-9B COST AND BENEFIT HOURS OF NON-RESIDENTIAL NON-NEM LOAD PROFILE ON A

SUMMER DAY

FIGURE 3-10ACOST AND BENEFIT HOURS OF RESIDENTIAL NON-NEM LOAD PROFILE ON A WINTER DAY

(PG&E-2)

3-16

FIGURE 3-10B COST AND BENEFIT HOURS OF NON-RESIDENTIAL NON-NEM LOAD PROFILE ON A

WINTER DAY

FIGURE 3-11ACOST AND BENEFIT HOURS OF RESIDENTIAL NON-NEM LOAD PROFILE ON A

SPRING DAY

(PG&E-2)

3-17

FIGURE 3-11B COST AND BENEFIT HOURS OF NON-RESIDENTIAL NON-NEM LOAD PROFILE ON A

SPRING DAY

3. Illustration of Generic Capacity Related “Cost” and “Benefit” 1

This section discusses generic capacity related “cost” and “benefit” 2

hours. As mentioned earlier, annual level MGCC is allocated at the hourly 3

level using the hourly ANL profile. The hourly cost allocated in this manner 4

is always either zero or positive.10 Therefore, the “cost” and “benefit” hours 5

are determined by whether the kWh is “delivered” or “received.” A cost is 6

assigned for “delivered” electricity hours, and a “benefit” is assigned for 7

“received” electricity hours. Note that “received” electricity exists only for 8

NEM customers. Non-NEM customers will see a “cost” for all hours in a 9

year. The illustration of such “cost” and “benefit” hours have been shown in 10

Figures 3-12A and 3-12B for a typical summer day (July 14, 2017). The 11

illustrations for winter and spring are similar to Figures 3-12A and 3-12B, but 12

with smaller PCAF weighted costs. 13

10 This is unlike the MEC at hourly level, which can either be positive, zero or negative.

(PG&E-2)

3-18

FIGURE 3-12A COST AND BENEFIT HOURS RESULTING FROM ONLY CAPACITY COST

ALLOCATION OF RESIDENTIAL NON-NEM LOAD PROFILE ON A SUMMER DAY

FIGURE 3-12B COST AND BENEFIT HOURS RESULTING FROM ONLY CAPACITY COST

ALLOCATION OF RESIDENTIAL NEM LOAD PROFILE ON A SUMMER DAY

D. Generation Cost of Service Results 1

The metric of MCR per kWh (or in the case of generation, GMCR per kWh) 2

is appropriate to evaluate cost of service, as stated earlier. PG&E concludes, 3

based on analysis of the data, that the differences in cost of service are driven 4

by (1) the differences in customer class load profiles; and (2) how the “delivered” 5

(PG&E-2)

3-19

and “received” energy flows from NEM customers impact NEM load profiles. 1

The load profile differences between Non-NEM and NEM customers have 2

already been discussed in Section B.3 of this chapter. 3

First, the results of the analysis are discussed separately for the two main 4

components of generation cost of service, MEC and MGCC. Next, the findings 5

related to total generation marginal cost per kWh are discussed. These results 6

form the basis of PG&E’s proposal that a separate cost analysis for Non-NEM 7

and NEM customers is necessary now and for future proceedings. 8

Finally, PG&E shows that Non-NEM and NEM customers have different cost 9

of service within the following main customer classes: Residential, Commercial, 10

Agricultural, and Large Commercial and Industrial.11 11

1. Generation Marginal Energy Cost Revenue Per kWh 12

This section presents the results and findings on the energy-related cost 13

of service differences between Non-NEM and NEM bundled customers. 14

Results at annual level as well as seasonal/TOU period level have 15

been discussed. 16

a. Annual Level Comparison of Average GMECR/kWh 17

Tables 3-1A and 3-1B present the cost of service comparison of 18

Non-NEM bundled customers with that of NEM bundled customers for 19

the years 2017 (recorded data) and 2021 (forecast data). Based on 20

the 2017 recorded data (Table 3-1A), the net MEC revenue per kWh, 21

i.e., GMECR per kWh of NEM customers is 5.58 cents/kWh, which 22

is significantly higher than the net GMECR of 3.99 cents/kWh for 23

Non-NEM customers. For the 2021 forecast year (Table 3-1B), the 24

per-unit cost for NEM bundled customers is also higher at 6 cents/kWh, 25

compared to 3.65 cents/kWh for the Non-NEM bundled customers.12 26

11 The Residential customer class includes all rate schedules available for the residential

customers. The Commercial customer class includes the rate schedules A-1, A-10, A-6, E-19P, E-19PV, E-19S, E-19SV, E-19T and E-19TV. The Agricultural customer class includes the rate groups American Gas Association, AGBLg and AGBSm. Large Commercial and Industrial customers include rate schedules E-20P, E-20S, E-20T. Standby and Streetlights have been excluded since these customers do not have any NEM.

12 The 2021 forecast includes the forecast of Direct Access/Community Choice Aggregator migration as well as higher NEM adoption.

(PG&E-2)

3-20

TABLE 3-1A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2017 ANNUAL

TABLE 3-1B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2021 ANNUAL

Tables 3-1A and 3-1B also present the cost and benefit estimates 1

separately for the “delivered” and “received” energy hours. Non-NEM 2

customers do not have any received load, while NEM customers have 3

both “delivered” and “received” load. It is important to note that NEM 4

customers’ GMECR/kWh for “delivered load” is 4.386 cents per kWh in 5

2017, which is 10 percent higher than that of Non-NEM customers in 6

2017. In addition, the GMECR/kWh is forecasted to be 4.28 cents 7

per kWh in 2021, which is 17.4 percent higher than the forecast for 8

Non-NEM customers of 3.65 cents per kWh. These results reveal that 9

the “delivered” portion of their load profile alone has higher average 10

MEC for NEM customers. On a net basis, NEM customers had a 11

40 percent higher average MEC than Non-NEM customers in 2017 and 12

are forecasted to have 64.6 percent higher average MEC than 13

Non-NEM customers in 2021. The percent differences are higher on a 14

net basis. This is because the “received” load usually occurs during 15

Formula Cost Benefit Net Cost Benefit NetDelivered a $2,167,946,724 ($2,877,236) $2,165,069,488 $184,603,036 ($76,042) $184,526,994Received b $0 $0 $0 $462,842 ($43,706,622) ($43,243,780)Net a+b $2,165,069,488 $141,283,214Delivered c 53,605,124,775 606,173,847 54,211,298,622 4,191,044,736 16,383,881 4,207,428,617Received d 0 0 0 109,650,911 1,565,629,475 1,675,280,387Net c-d 54,211,298,622 2,532,148,230Delivered a/c 0.04044 -0.00475 0.03994 0.04405 -0.00464 0.04386Received b/d 0 0 0 0.00422 -0.02792 -0.02581Net (a+b)/abs(c-d) 0.03994 0.05580

Non-NEM NEMAnnual Generation Energy Cost of Service of All Bundled Customers

GMCR

kWh

GMCR per kWh($/kWh)

Formula Cost Benefit Net Cost Benefit NetDelivered a $1,089,707,485 ($7,845,891) $1,081,861,594 $229,127,544 ($571,914) $228,555,630Received b $0 $0 $0 $2,273,152 ($31,357,509) ($29,084,357)Net a+b $1,081,861,594 $199,471,273Delivered c 28,251,081,013 1,407,658,698 29,658,739,711 5,233,548,333 103,933,555 5,337,481,888Received d 0 0 0 384,964,959 1,629,452,575 2,014,417,534Net c-d 29,658,739,711 3,323,064,354Delivered a/c 0.03857 -0.00557 0.03648 0.04378 -0.0055 0.04282Received b/d 0 0 0 0.0059 -0.01924 -0.01444Net (a+b)/abs(c-d) 0.03648 0.06003

GMCR

kWh

GMCR per kWh($/kWh)

Annual Generation Energy Cost of Service of All Bundled CustomersNon-NEM NEM

(PG&E-2)

3-21

relatively low-cost hours of the day along with the fact that the “net” load 1

is smaller than the “delivered” load.13 2

b. Comparison of Average GMECR/kWh in Summer 3

Tables 3-2A and 3-2B compare the GMECR, usage (kWh), and 4

average GMECR/kWh comparison between Non-NEM and NEM 5

customers for the summer TOU periods, i.e., summer peak, summer 6

part-peak and summer off-peak. Table 3-2A shows 2017 summer data 7

and Table 3-2B shows the forecast. 8

During the summer peak period, NEM customers have a smaller 9

percentage of “received” load, when compared to summer part-peak 10

and summer off-peak periods. A relatively small percentage of 11

“received” load causes the net average GMECR/kWh to be driven 12

mostly by the “delivered” GMECR/kWh. In 2017, NEM customers had a 13

14 percent higher net GMECR/kWh than that of Non-NEM customers. 14

In 2021, NEM customers are forecasted to have 11 percent higher 15

GMECR/kWh compared to Non-NEM customers. 16

13 Section 4.e of this chapter provides additional explanation for the differences in cost of

service between NEM and Non-NEM customers.

(PG&E-2)

3-22

TABLE 3-2A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2017 SUMMER

During the summer part-peak and off-peak periods, NEM customers 1

have significant “received” load and the net GMECR/kWh are 2

significantly higher than that of the Non-NEM customers. In 2017, NEM 3

customers had 15.2 percent and 7 percent higher net GMECR/kWh than 4

that of Non-NEM customers during part-peak and off-peak periods, 5

respectively. However, in 2021, the GMECR/kWh for NEM customers 6

are forecasted to be much higher than Non-NEM customers: 56 percent 7

higher during the part-peak period, and 58 percent higher during the 8

off-peak period. 9

Summer Generation Energy Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $427,111,072 $0 $427,111,072 $40,760,663 $0 $40,760,663Received $0 $0 $0 $0 ($3,874,529) ($3,874,529)Net $427,111,072 $36,886,134Delivered 5,571,417,845 0 5,571,417,845 495,321,066 0 495,321,066Received 0 0 0 590,640 73,481,188 74,071,827Net 5,571,417,845 421,249,239Delivered 0.07666 0 0.07666 0.08229 0 0.08229Received 0 0 0 0 -0.05273 -0.05231Net 0.07666 0.08756

Cost Benefit Net Cost Benefit NetDelivered $190,483,676 ($100,407) $190,383,269 $14,417,983 ($2,989) $14,414,995Received $0 $0 $0 $17,015 ($6,078,555) ($6,061,540)Net $190,383,269 $8,353,454Delivered 4,123,389,466 11,060,331 4,134,449,797 307,935,603 332,023 308,267,626Received 0 0 0 3,958,436 146,790,772 150,749,208Net 4,134,449,797 157,518,418Delivered 0.0462 -0.00908 0.04605 0.04682 -0.009 0.04676Received 0 0 0 0.0043 -0.04141 -0.04021Net 0.04605 0.05303

Cost Benefit Net Cost Benefit NetDelivered $390,078,751 ($271,850) $389,806,901 $28,949,206 ($7,976) $28,941,230Received $0 $0 $0 $43,503 ($13,394,046) ($13,350,544)Net $389,806,901 $15,590,686Delivered 11,881,358,415 44,825,714 11,926,184,129 876,499,195 1,311,530 877,810,725Received 0 0 0 8,218,130 425,529,706 433,747,836Net 11,926,184,129 444,062,889Delivered 0.03283 -0.00606 0.03268 0.03303 -0.00608 0.03297Received 0 0 0 0.00529 -0.03148 -0.03078Net 0.03268 0.03511

Summer Part-peak: NEMSummer Part-peak: Non-NEM

GMCR

kWh

GMCR per kWh($/kWh)

Summer Peak: Non-NEM Summer Peak: NEM

GMCR

kWh

GMCR per kWh($/kWh)

GMCR

kWh

GMCR per kWh($/kWh)

Summer Off-peak: Non-NEM Summer Off-peak: NEM

(PG&E-2)

3-23

TABLE 3-2B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2021 SUMMER

c. Comparison of Average GMECR/kWh in Winter 1

Table 3-3A and 3-3B compare the GMECR, usage (kWh), and 2

average GMECR/kWh for the winter TOU periods: winter peak, winter 3

off-peak and winter super-off-peak. 4

During the winter peak period, NEM customers have a smaller 5

percentage of “received” load, when compared to winter off-peak and 6

winter super-off-peak periods. A relatively small percentage of 7

“received” load causes the net average GMECR/kWh to be driven 8

mostly by the “delivered” GMECR/kWh. In 2017, the net GMECR/kWh 9

for NEM customers were 15 percent higher than that of the Non-NEM 10

customers. In 2021, the GMECR/kWh for NEM customers are 11

forecasted to be 18 percent higher. 12

Cost Benefit Net Cost Benefit NetDelivered $189,061,103 $0 $189,061,103 $41,047,074 $0 $41,047,074Received $0 $0 $0 $0 ($4,093,754) ($4,093,754)Net $189,061,103 $36,953,320Delivered 3,056,579,281 0 3,056,579,281 626,590,547 0 626,590,547Received 0 0 0 0 88,158,394 88,158,394Net 3,056,579,281 538,432,152Delivered 0.06185 0 0.06185 0.06551 0 0.06551Received 0 0 0 0 -0.04644 -0.04644Net 0.06185 0.06863

Cost Benefit Net Cost Benefit NetDelivered $101,939,340 ($59,898) $101,879,442 $20,261,004 ($5,186) $20,255,817Received $0 $0 $0 $11,021 ($5,286,056) ($5,275,036)Net $101,879,442 $14,980,782Delivered 2,262,531,935 8,895,361 2,271,427,296 391,088,093 745,070 391,833,162Received 0 0 0 1,931,327 175,594,705 177,526,032Net 2,271,427,296 214,307,130Delivered 0.04506 -0.00673 0.04485 0.05181 -0.00696 0.0517Received 0 0 0 0.00571 -0.0301 -0.02971Net 0.04485 0.0699

Cost Benefit Net Cost Benefit NetDelivered $191,494,666 ($703,146) $190,791,520 $37,768,994 ($57,646) $37,711,348Received $0 $0 $0 $139,032 ($9,289,553) ($9,150,521)Net $190,791,520 $28,560,827Delivered 6,535,901,726 125,456,875 6,661,358,601 1,133,216,182 9,825,928 1,143,042,111Received 0 0 0 27,184,025 483,064,457 510,248,483Net 6,661,358,601 632,793,628Delivered 0.0293 -0.0056 0.02864 0.03333 -0.00587 0.03299Received 0 0 0 0.00511 -0.01923 -0.01793Net 0.02864 0.04513

GMCR

kWh

GMCR per kWh($/kWh)

GMCR

Summer Off-peak: Non-NEM Summer Off-peak: NEM

GMCR

kWh

GMCR per kWh($/kWh)

Summer Part-peak: Non-NEM Summer Part-peak: NEM

GMCR per kWh($/kWh)

kWh

Summer Generation Energy Cost of Service of All Bundled CustomersSummer Peak: Non-NEM Summer Peak: NEM

(PG&E-2)

3-24

TABLE 3-3A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2017 WINTER

During the winter off-peak period, NEM customers have relatively 1

large “received” load, and the net GMECR/kWh are significantly higher 2

than that of the Non-NEM customers. In 2017, the net GMECR/kWh for 3

NEM customers was 22 percent higher than that of the Non-NEM 4

customers. In 2021, NEM customers’ forecasted GMECR/kWh is 5

expected to be much higher—52 percent higher than that of Non-NEM 6

customers. 7

The winter super-off-peak period is different from the rest of the 8

TOU periods because energy prices can be negative during quite a few 9

hours, and it is relatively low for the rest of the hours. The average 10

GMECR/kWh is, therefore, relatively low for both Non-NEM and NEM 11

customers. The 2017 data shows a negative GMECR (-$0.00892/kWh) 12

for NEM customers. PG&E investigated the hourly 2017 super-off-peak 13

Winter Generation Energy Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $421,363,367 ($77,923) $421,285,443 $40,193,625 ($2,303) $40,191,322Received $0 $0 $0 $8,463 ($2,114,392) ($2,105,930)Net $421,285,443 $38,085,392Delivered 7,705,480,273 20,874,873 7,726,355,146 687,572,602 660,154 688,232,757Received 0 0 0 3,313,031 75,383,377 78,696,408Net 7,726,355,146 609,536,348Delivered 0.05468 -0.00373 0.05453 0.05846 -0.00349 0.0584Received 0 0 0 0.00255 -0.02805 -0.02676Net 0.05453 0.06248

Cost Benefit Net Cost Benefit NetDelivered $702,946,146 ($801,037) $702,145,109 $59,080,713 ($21,170) $59,059,544Received $0 $0 $0 $124,278 ($14,627,090) ($14,502,812)Net $702,145,109 $44,556,732Delivered 21,819,548,890 173,851,867 21,993,400,757 1,743,997,189 4,716,293 1,748,713,482Received 0 0 0 32,607,957 570,831,752 603,439,709Net 21,993,400,757 1,145,273,773Delivered 0.03222 -0.00461 0.03193 0.03388 -0.00449 0.03377Received 0 0 0 0.00381 -0.02562 -0.02403Net 0.03193 0.0389

Cost Benefit Net Cost Benefit NetDelivered $35,963,712 ($1,626,020) $34,337,693 $1,200,846 ($41,604) $1,159,241Received $0 $0 $0 $269,584 ($3,618,009) ($3,348,426)Net $34,337,693 ($2,189,184)Delivered 2,503,929,886 355,561,063 2,859,490,949 79,719,080 9,363,881 89,082,961Received 0 0 0 60,962,717 273,612,681 334,575,398Net 2,859,490,949 -245,492,437Delivered 0.01436 -0.00457 0.01201 0.01506 -0.00444 0.01301Received 0 0 0 0.00442 -0.01322 -0.01001Net 0.01201 -0.00892

Winter Peak: Non-NEM Winter Peak: NEM

GMCR per kWh($/kWh)

Winter Off-peak: Non-NEM Winter Off-peak: NEM

GMCR

kWh

Super Off-peak: NEM

GMCR

kWh

GMCR per kWh($/kWh)

GMCR

kWh

GMCR per kWh($/kWh)

Super Off-peak: Non-NEM

(PG&E-2)

3-25

usage and marginal cost data and found that the negative GMECR/kWh 1

is driven by mostly positive hourly energy marginal cost, while the load 2

is “received” in almost all hours during this TOU period. Non-NEM 3

customers, on the other hand, have only “delivered” load and, therefore, 4

positive average GMECR/kWh in 2017. In 2021, however, MEC is 5

forecasted to be negative during many of the super-off-peak hours. As 6

a result, NEM customers’ “received” load will cause a significantly higher 7

average GMECR/kWh. Table 3-3B shows that GMECR/kWh is 8

0.24 cents per kWh for NEM customers, while it is negative (-0.07 cents 9

per kWh) for Non-NEM customers. 10

TABLE 3-3B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION ENERGY, 2021 WINTER

Cost Benefit Net Cost Benefit NetDelivered $217,410,543 ($145,064) $217,265,479 $48,388,903 ($13,829) $48,375,074Received $0 $0 $0 $26,872 ($1,826,127) ($1,799,255)Net $217,265,479 $46,575,819Delivered 4,145,066,731 42,447,167 4,187,513,898 856,951,107 3,986,769 860,937,876Received 0 0 0 7,271,064 93,108,043 100,379,107Net 4,187,513,898 760,558,768Delivered 0.05245 -0.00342 0.05188 0.05647 -0.00347 0.05619Received 0 0 0 0.0037 -0.01961 -0.01792Net 0.05188 0.06124

Cost Benefit Net Cost Benefit NetDelivered $386,203,075 ($2,241,403) $383,961,673 $81,359,832 ($167,935) $81,191,898Received $0 $0 $0 $628,558 ($10,124,765) ($9,496,206)Net $383,961,673 $71,695,692Delivered 11,462,460,635 440,712,398 11,903,173,033 2,162,894,141 34,096,154 2,196,990,295Received 0 0 0 116,147,015 616,175,044 732,322,059Net 11,903,173,033 1,464,668,235Delivered 0.03369 -0.00509 0.03226 0.03762 -0.00493 0.03696Received 0 0 0 0.00541 -0.01643 -0.01297Net 0.03226 0.04895

Cost Benefit Net Cost Benefit NetDelivered $3,598,759 ($4,696,381) ($1,097,622) $301,737 ($327,318) ($25,581)Received $0 $0 $0 $1,467,669 ($737,255) $730,414Net ($1,097,622) $704,833Delivered 788,540,705 790,146,897 1,578,687,602 62,808,264 55,279,634 118,087,898Received 0 0 0 232,431,528 173,351,931 405,783,458Net 1,578,687,602 -287,695,560Delivered 0.00456 -0.00594 -0.0007 0.0048 -0.00592 -0.00022Received 0 0 0 0.00631 -0.00425 0.0018Net -0.0007 0.00245

GMCR

kWh

Winter Off-peak: Non-NEM Winter Off-peak: NEM

GMCR

GMCR per kWh($/kWh)

Super Off-peak: Non-NEM Super Off-peak: NEM

GMCR

kWh

GMCR per kWh($/kWh)

GMCR per kWh($/kWh)

kWh

Winter Generation Energy Cost of Service of All Bundled CustomersWinter Peak: Non-NEM Winter Peak: NEM

(PG&E-2)

3-26

2. Marginal Generation Capacity Cost Revenue Per kWh 1

This section presents the comparison of average MGCC between the 2

Non-NEM and NEM bundled14 customers, including results on an annual 3

basis and results for summer and winter TOU periods. The average 4

marginal capacity cost is measured as the MGCCR/kWh. Although 5

MGCCR/kWh is lower than the GMECR/kWh discussed previously, the 6

conclusion is consistent in that the MGCCR/kWh is higher for bundled NEM 7

customers when compared with bundled Non-NEM customers. 8

a. Annual Level Comparison of Average MGCCR/kWh 9

The comparison, on an annual basis, of MGCCR/kWh between 10

Non-NEM and NEM bundled customers is shown in Tables 3-4A and 11

3-4B. Table 3-4A shows the results based on the 2017 recorded data 12

and Table 3-4B shows the results based on 2021 forecasts. Annual 13

MGCCs are allocated at the hourly level using PG&E’s generation PCAF 14

methodology, which allocates marginal costs at the hourly level based 15

on coincident demand with respect to ANL. The load profile of NEM 16

customers shows that their peak is more coincident to the ANL than the 17

load profile of Non-NEM customers, and therefore, the average 18

MGCCR/kWh is higher for the NEM customers. This is evident from 19

both 2017 (Table 3-4A) and 2021 forecast (Table 3-4B) results. As 20

shown in Table 3-4A, the net MGCCR/kWh is 0.643 cents per kWh for 21

Non-NEM customers, and 1.376 cents per kWh for NEM customers in 22

2017. Table 3-4B shows that the net MGCCR/kWh becomes 2 cents 23

per kWh for Non-NEM customers, and 4.74 cents per kWh for NEM 24

customers in 2021. 25

14 Except the Streetlights and Standby customers, as mentioned above.

(PG&E-2)

3-27

TABLE 3-4A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2017 ANNUAL

TABLE 3-4B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2021 ANNUAL

b. Comparison of Average MGCCR/kWh in Summer 1

The comparison of MGCCR/kWh between the Non-NEM and NEM 2

bundled customers for the summer TOU periods is shown in 3

Tables 3-5A and 3-5B. Table 3-5A shows the results based on the 4

2017 recorded data and Table 3-5B shows the results based on 2021 5

forecasts. There are three summer TOU periods: summer peak, 6

summer part-peak and summer off-peak. There is no allocation of 7

capacity cost for summer off-peak period since the allocation is 8

essentially based on ANL coincident peak hours and there are no such 9

hours during these TOU periods. 10

For the summer peak period, Table 3-5A shows that the net 11

MGCCR/kWh is 7.606 cents per kWh for NEM customers and 12

5.547 cents per kWh for Non-NEM customers. NEM customers’ 13

Annual Generation Capacity Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $348,689,984 $0 $348,689,984 $36,784,279 $0 $36,784,279Received $0 $0 $0 $0 ($1,936,484) ($1,936,484)Net $348,689,984 $34,847,795Delivered 53,605,124,775 0 54,211,298,622 4,191,044,736 0 4,207,428,617Received 0 0 0 109,650,911 1,565,629,475 1,675,280,387Net 54,211,298,622 2,532,148,230Delivered 0.0065 0 0.00643 0.00878 0 0.00874Received 0 0 0 0 -0.00124 -0.00116Net 0.00643 0.01376

Non-NEM NEM

MGCCR

kWh

MGCCR per kWh($/kWh)

Cost Benefit Net Cost Benefit NetDelivered $595,345,813 $0 $595,345,813 $158,283,188 $0 $158,283,188Received $0 $0 $0 $0 ($816,047) ($816,047)Net $595,345,813 $157,467,141Delivered 28,251,081,013 0 29,658,739,711 5,233,548,333 0 5,337,481,888Received 0 0 0 384,964,959 1,629,452,575 2,014,417,534Net 29,658,739,711 3,323,064,354Delivered 0.02107 0 0.02007 0.03024 0 0.02966Received 0 0 0 0 -0.0005 -0.00041Net 0.02007 0.04739

Annual Generation Capacity Cost of Service of All Bundled CustomersNon-NEM NEM

MGCCR

kWh

MGCCR per kWh($/kWh)

(PG&E-2)

3-28

MGCCR/kWh is higher during this TOU period because NEM load 1

profiles are more skewed towards the higher cost hours with summer 2

peak period. A similar result is shown in Table 3-5B based on 2021 3

forecast data: the MGCCR/kWh for NEM customers is 22.23 cents per 4

kWh, and for Non-NEM customers it is 14.75 cents per kWh. 5

For the summer part-peak period in Table 3-5A, which is based 6

on 2017 recorded data, the net MGCCR/kWh is 1.728 cents per kWh 7

for NEM customers and 0.94 cents for Non-NEM customers. This 8

comparison is similar to the summer peak period. Table 3-5B, based 9

on 2021 forecast data, shows that the MGCCR/kWh for NEM customers 10

is 8.51 cents per kWh, whereas for Non-NEM customers it is 11

3.02 cents per kWh. 12

TABLE 3-5A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2017 SUMMER

Summer Generation Capacity Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $309,069,738 $0 $309,069,738 $33,340,222 $0 $33,340,222Received $0 $0 $0 $0 ($1,302,000) ($1,302,000)Net $309,069,738 $32,038,222Delivered 5,571,417,845 0 5,571,417,845 495,321,066 0 495,321,066Received 0 0 0 590,640 73,481,188 74,071,827Net 5,571,417,845 421,249,239Delivered 0.05547 0 0.05547 0.06731 0 0.06731Received 0 0 0 0 -0.01772 -0.01758Net 0.05547 0.07606

Cost Benefit Net Cost Benefit NetDelivered $38,867,592 $0 $38,867,592 $3,356,139 $0 $3,356,139Received $0 $0 $0 $0 ($634,389) ($634,389)Net $38,867,592 $2,721,750Delivered 4,123,389,466 0 4,134,449,797 307,935,603 0 308,267,626Received 0 0 0 3,958,436 146,790,772 150,749,208Net 4,134,449,797 157,518,418Delivered 0.00943 0 0.0094 0.0109 0 0.01089Received 0 0 0 0 -0.00432 -0.00421Net 0.0094 0.01728

Cost Benefit Net Cost Benefit NetDelivered $0 $0 $0 $0 $0 $0Received $0 $0 $0 $0 $0 $0Net $0 $0Delivered 11,881,358,415 0 11,926,184,129 876,499,195 0 877,810,725Received 0 0 0 8,218,130 0 433,747,836Net 11,926,184,129 444,062,889Delivered 0 0 0 0 0 0Received 0 0 0 0 0 0Net 0 0

MGCCR per kWh($/kWh)

Summer Off-peak: Non-NEM Summer Off-peak: NEM

MGCCR

Summer Peak: Non-NEM Summer Peak: NEM

MGCCR

kWh

MGCCR per kWh($/kWh)

kWh

MGCCR per kWh($/kWh)

Summer Part-peak: Non-NEM Summer Part-peak: NEM

MGCCR

kWh

(PG&E-2)

3-29

TABLE 3-5B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2021 SUMMER

c. Comparison of Average MGCCR/kWh in Winter 1

The comparison of MGCCR/kWh between Non-NEM and NEM 2

bundled customers for the winter TOU periods is shown in Tables 3-6A 3

and 3-6B. Table 3-6A shows the results based on the 2017 recorded 4

data and Table 3-5B shows the results based on 2021 forecasts. There 5

are three winter TOU periods: winter peak, winter off-peak and winter 6

super-off-peak. There is no allocation of generic capacity cost for winter 7

off-peak and winter super off-peak periods since the allocation is 8

essentially based on ANL coincident peak hours, and there are no such 9

hours during these TOU periods. For the winter peak period, the 10

allocated costs are quite low. Hence, the comparison of the numbers is 11

not meaningful. 12

Cost Benefit Net Cost Benefit NetDelivered $450,756,487 $0 $450,756,487 $120,412,014 $0 $120,412,014Received $0 $0 $0 $0 ($733,441) ($733,441)Net $450,756,487 $119,678,573Delivered 3,056,579,281 0 3,056,579,281 626,590,547 0 626,590,547Received 0 0 0 0 88,158,394 88,158,394Net 3,056,579,281 538,432,152Delivered 0.14747 0 0.14747 0.19217 0 0.19217Received 0 0 0 0 -0.00832 -0.00832Net 0.14747 0.22227

Cost Benefit Net Cost Benefit NetDelivered $68,504,788 $0 $68,504,788 $18,271,594 $0 $18,271,594Received $0 $0 $0 $0 ($39,932) ($39,932)Net $68,504,788 $18,231,663Delivered 2,262,531,935 0 2,271,427,296 391,088,093 0 391,833,162Received 0 0 0 1,931,327 175,594,705 177,526,032Net 2,271,427,296 214,307,130Delivered 0.03028 0 0.03016 0.04672 0 0.04663Received 0 0 0 0 -0.00023 -0.00022Net 0.03016 0.08507

Cost Benefit Net Cost Benefit NetDelivered $0 $0 $0 $0 $0 $0Received $0 $0 $0 $0 $0 $0Net $0 $0Delivered 6,535,901,726 0 6,661,358,601 1,133,216,182 0 1,143,042,111Received 0 0 0 27,184,025 0 510,248,483Net 6,661,358,601 632,793,628Delivered 0 0 0 0 0 0Received 0 0 0 0 0 0Net 0 0

Summer Generation Capacity Cost of Service of All Bundled CustomersSummer Peak: Non-NEM Summer Peak: NEM

MGCCR

kWh

MGCCR

kWh

MGCCR per kWh($/kWh)

Summer Part-peak: Non-NEM Summer Part-peak: NEM

MGCCR per kWh($/kWh)

MGCCR per kWh($/kWh)

Summer Off-peak: Non-NEM Summer Off-peak: NEM

MGCCR

kWh

(PG&E-2)

3-30

TABLE 3-6A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2017 WINTER

Winter Generation Capacity Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $752,654 $0 $752,654 $87,918 $0 $87,918Received $0 $0 $0 $0 ($95) ($95)Net $752,654 $87,823Delivered 7,705,480,273 0 7,726,355,146 687,572,602 0 688,232,757Received 0 0 0 3,313,031 75,383,377 78,696,408Net 7,726,355,146 609,536,348Delivered 0.0001 0 0.0001 0.00013 0 0.00013Received 0 0 0 0 0 0Net 0.0001 0.00014

Cost Benefit Net Cost Benefit NetDelivered $0 $0 $0 $0 $0 $0Received $0 $0 $0 $0 $0 $0Net $0 $0Delivered 21,819,548,890 0 21,993,400,757 1,743,997,189 0 1,748,713,482Received 0 0 0 32,607,957 0 603,439,709Net 21,993,400,757 1,145,273,773Delivered 0 0 0 0 0 0Received 0 0 0 0 0 0Net 0 0

Cost Benefit Net Cost Benefit NetDelivered $0 $0 $0 $0 $0 $0Received $0 $0 $0 $0 $0 $0Net $0 $0Delivered 2,503,929,886 0 2,859,490,949 79,719,080 0 89,082,961Received 0 0 0 60,962,717 0 334,575,398Net 2,859,490,949 -245,492,437Delivered 0 0 0 0 0 0Received 0 0 0 0 0 0Net 0 0

MGCCR per kWh($/kWh)

MGCCR

Winter Peak: Non-NEM Winter Peak: NEM

MGCCR

kWh

MGCCR per kWh($/kWh)

kWh

MGCCR per kWh($/kWh)

Super Off-peak: Non-NEM Super Off-peak: NEM

Winter Off-peak: Non-NEM Winter Off-peak: NEM

MGCCR

kWh

(PG&E-2)

3-31

TABLE 3-6B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

GENERATION CAPACITY, 2021 WINTER

3. Total Generation Marginal Cost Revenue Per kWh 1

This section discusses the findings related the total GMCR15 per kWh 2

for Non-NEM and NEM bundled customers. Results on an annual basis as 3

well as for seasonal/TOU periods are presented. 4

a. Annual Level Comparison of Total GMCR/kWh 5

Table 3-7A shows the comparison, on an annual basis, of total 6

GMCR per kWh based on 2017 recorded data. For the net usage, the 7

total GMCR/kWh for NEM customers is 6.956 cents per kWh, while the 8

total GMCR/kWh for Non-NEM customers is 4.637 cents per kWh. In 9

other words, GMCR/kwh is 2.319 cents per kWh higher for NEM 10

customers than for Non-NEM customers on an annual basis. For 11

15 GMCR is the sum of GMECR and MGCCR.

Cost Benefit Net Cost Benefit NetDelivered $74,206,214 $0 $74,206,214 $19,143,640 $0 $19,143,640Received $0 $0 $0 $0 ($41,621) ($41,621)Net $74,206,214 $19,102,019Delivered 4,145,066,731 0 4,187,513,898 856,951,107 0 860,937,876Received 0 0 0 7,271,064 93,108,043 100,379,107Net 4,187,513,898 760,558,768Delivered 0.0179 0 0.01772 0.02234 0 0.02224Received 0 0 0 0 -0.00045 -0.00041Net 0.01772 0.02512

Cost Benefit Net Cost Benefit NetDelivered $1,878,324 $0 $1,878,324 $455,940 $0 $455,940Received $0 $0 $0 $0 ($1,053) ($1,053)Net $1,878,324 $454,887Delivered 11,462,460,635 0 11,903,173,033 2,162,894,141 0 2,196,990,295Received 0 0 0 116,147,015 616,175,044 732,322,059Net 11,903,173,033 1,464,668,235Delivered 0.00016 0 0.00016 0.00021 0 0.00021Received 0 0 0 0 0 0Net 0.00016 0.00031

Cost Benefit Net Cost Benefit NetDelivered $0 $0 $0 $0 $0 $0Received $0 $0 $0 $0 $0 $0Net $0 $0Delivered 607,371,183 0 1,578,687,602 49,016,591 0 118,087,898Received 0 0 0 275,622,015 0 405,783,458Net 1,578,687,602 -287,695,560Delivered 0 0 0 0 0 0Received 0 0 0 0 0 0Net 0 0

Winter Generation Capacity Cost of Service of All Bundled CustomersWinter Peak: Non-NEM Winter Peak: NEM

MGCCR

kWh

MGCCR

Super Off-peak: Non-NEM Super Off-peak: NEM

Winter Off-peak: Non-NEM Winter Off-peak: NEM

kWh

MGCCR per kWh($/kWh)

kWh

MGCCR per kWh($/kWh)

MGCCR per kWh($/kWh)

MGCCR

(PG&E-2)

3-32

“delivered” usage, GMCR/kWh (5.26 cent per kWh) for NEM customers 1

is higher than Non-NEM customers’ GMCR/kWh (4.637 cents per kWh). 2

This is consistent with what we expect, because the NEM customers’ 3

“delivered” kWh occurs mostly during the highest cost hours 4

(i.e., summer peak and winter peak hours), while the “delivered” kWh 5

for Non NEM customers occurs throughout all 24 hours. Additionally, 6

since NEM customers’ “received” kWh (electricity sent back to the grid) 7

occurs during the daytime when hourly marginal costs are relatively low, 8

the GMCR/kwh for “received” kWh is less than that for the “delivered” 9

load, as seen in Table 3-7A. The GMCR/kWh for “received” kWh 10

is -2.697 cent per kWh. 11

TABLE 3-7A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2017 ANNUAL

Table 3-7B shows the annual level comparison of total GMCR per 12

kWh based on the 2021 forecasted data. GMCR/kWh comparisons are 13

similar to those observed in Table 3-7A. For the “delivered” kWh, 14

GMCR/kWh is higher for NEM customers when compared to that of 15

Non-NEM customers. Also, the GMCR/kWh for the “received” kWh is 16

lower than the GMCR/kWh for the “delivered” kWh for the 17

NEM customers. 18

Annual Total Generation Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $2,516,636,707 ($2,877,236) $2,513,759,471 $221,387,316 ($76,042) $221,311,274Received $0 $0 $0 $462,842 ($45,643,107) ($45,180,265)Net $2,513,759,471 $176,131,009Delivered 53,605,124,775 606,173,847 54,211,298,622 4,191,044,736 16,383,881 4,207,428,617Received 0 0 0 109,650,911 1,565,629,475 1,675,280,387Net 54,211,298,622 2,532,148,230Delivered 0.04695 -0.00475 0.04637 0.05282 -0.00464 0.0526Received 0 0 0 0.00422 -0.02915 -0.02697Net 0.04637 0.06956

Non-NEM NEM

GMCR

kWh

GMCR per kWh($/kWh)

(PG&E-2)

3-33

TABLE 3-7B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2021 ANNUAL

b. Comparison of Total GMCR Per kWh for Summer TOU Periods 1

Table 3-8A shows a comparison of GMCR per kWh between 2

Non-NEM and NEM customers for the summer TOU periods using 3

the 2017 recorded data. For net usage, the GMCR per kWh for NEM 4

customers is higher for all TOU periods. The “received” usage is low in 5

the summer peak period, and it is higher for the summer part-peak and 6

off-peak periods. As a result, the difference in GMCR per kWh on net 7

usage basis is the highest for the summer peak period. For “delivered” 8

usage, the GMCR/kWh for NEM customers is significantly higher than 9

that of Non-NEM customers for the summer peak period (1.746 cents 10

per kWh higher), while the difference in GMCR/kWh is relatively lower 11

for the part-peak and off-peak period. 12

Table 3-8B shows the same comparison as Table 3-8A, using the 13

2021 forecast data. 14

Cost Benefit Net Cost Benefit NetDelivered $1,685,053,299 ($7,845,891) $1,677,207,407 $387,410,732 ($571,914) $386,838,818Received $0 $0 $0 $2,273,152 ($32,173,556) ($29,900,404)Net $1,677,207,407 $356,938,414Delivered 28,251,081,013 1,407,658,698 29,658,739,711 5,233,548,333 103,933,555 5,337,481,888Received 0 0 0 384,964,959 1,629,452,575 2,014,417,534Net 29,658,739,711 3,323,064,354Delivered 0.05965 -0.00557 0.05655 0.07402 -0.0055 0.07248Received 0 0 0 0.0059 -0.01975 -0.01484Net 0.05655 0.10741

Annual Total Generation Cost of Service of All Bundled Customers

GMCR per kWh($/kWh)

Non-NEM NEM

GMCR

kWh

(PG&E-2)

3-34

TABLE 3-8A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2017 SUMMER

Summer Total Generation Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $736,180,810 $0 $736,180,810 $74,100,886 $0 $74,100,886Received $0 $0 $0 $0 ($5,176,529) ($5,176,529)Net $736,180,810 $68,924,356Delivered 5,571,417,845 0 5,571,417,845 495,321,066 0 495,321,066Received 0 0 0 590,640 73,481,188 74,071,827Net 5,571,417,845 421,249,239Delivered 0.13214 0 0.13214 0.1496 0 0.1496Received 0 0 0 0 -0.07045 -0.06989Net 0.13214 0.16362

Cost Benefit Net Cost Benefit NetDelivered $229,351,268 ($100,407) $229,250,861 $17,774,123 ($2,989) $17,771,134Received $0 $0 $0 $17,015 ($6,712,944) ($6,695,929)Net $229,250,861 $11,075,205Delivered 4,123,389,466 11,060,331 4,134,449,797 307,935,603 332,023 308,267,626Received 0 0 0 3,958,436 146,790,772 150,749,208Net 4,134,449,797 157,518,418Delivered 0.05562 -0.00908 0.05545 0.05772 -0.009 0.05765Received 0 0 0 0.0043 -0.04573 -0.04442Net 0.05545 0.07031

Cost Benefit Net Cost Benefit NetGMCR Delivered $390,078,751 ($271,850) $389,806,901 $28,949,206 ($7,976) $28,941,230

Received $0 $0 $0 $43,503 ($13,394,046) ($13,350,544)Net $389,806,901 $15,590,686Delivered 11,881,358,415 44,825,714 11,926,184,129 876,499,195 1,311,530 877,810,725Received 0 0 0 8,218,130 425,529,706 433,747,836Net 11,926,184,129 444,062,889Delivered 0.03283 -0.00606 0.03268 0.03303 -0.00608 0.03297Received 0 0 0 0.00529 -0.03148 -0.03078Net 0.03268 0.03511

GMCR per kWh($/kWh)

kWh

GMCR per kWh($/kWh)

GMCR per kWh($/kWh)

Summer Part-peak: Non-NEM Summer Part-peak: NEM

GMCR

kWh

Summer Off-peak: Non-NEM Summer Off-peak: NEM

Summer Peak: Non-NEM Summer Peak: NEM

GMCR

kWh

(PG&E-2)

3-35

TABLE 3-8B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2021 SUMMER

c. Comparison of Total GMCR per kWh Winter TOU Periods 1

Tables 3-9A and 3-9B show the comparison of GMCR/kWh between 2

Non-NEM and NEM customers for the winter TOU periods, based on 3

2017 recorded data and 2021 forecasted data, respectively. The results 4

shown here are consistent with the results shown in previous 5

subsections. 6

Cost Benefit Net Cost Benefit NetDelivered $639,817,590 $0 $639,817,590 $161,459,088 $0 $161,459,088Received $0 $0 $0 $0 ($4,827,195) ($4,827,195)Net $639,817,590 $156,631,893Delivered 3,056,579,281 0 3,056,579,281 626,590,547 0 626,590,547Received 0 0 0 0 88,158,394 88,158,394Net 3,056,579,281 538,432,152Delivered 0.20932 0 0.20932 0.25768 0 0.25768Received 0 0 0 0 -0.05476 -0.05476Net 0.20932 0.2909

Cost Benefit Net Cost Benefit NetDelivered $170,444,128 ($59,898) $170,384,230 $38,532,598 ($5,186) $38,527,412Received $0 $0 $0 $11,021 ($5,325,988) ($5,314,967)Net $170,384,230 $33,212,444Delivered 2,262,531,935 8,895,361 2,271,427,296 391,088,093 745,070 391,833,162Received 0 0 0 1,931,327 175,594,705 177,526,032Net 2,271,427,296 214,307,130Delivered 0.07533 -0.00673 0.07501 0.09853 -0.00696 0.09833Received 0 0 0 0.00571 -0.03033 -0.02994Net 0.07501 0.15498

Cost Benefit Net Cost Benefit NetGMCR Delivered $191,494,666 ($703,146) $190,791,520 $37,768,994 ($57,646) $37,711,348

Received $0 $0 $0 $139,032 ($9,289,553) ($9,150,521)Net $190,791,520 $28,560,827Delivered 6,535,901,726 125,456,875 6,661,358,601 1,133,216,182 9,825,928 1,143,042,111Received 0 0 0 27,184,025 483,064,457 510,248,483Net 6,661,358,601 632,793,628Delivered 0.0293 -0.0056 0.02864 0.03333 -0.00587 0.03299Received 0 0 0 0.00511 -0.01923 -0.01793Net 0.02864 0.04513

Summer Total Generation Cost of Service of All Bundled Customers

GMCR per kWh($/kWh)

Summer Peak: Non-NEM Summer Peak: NEM

GMCR

kWh

GMCR

kWh

Summer Part-peak: Non-NEM Summer Part-peak: NEM

GMCR per kWh($/kWh)

kWh

GMCR per kWh($/kWh)

Summer Off-peak: Non-NEM Summer Off-peak: NEM

(PG&E-2)

3-36

TABLE 3-9A COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2017 WINTER

Winter Total Generation Cost of Service of All Bundled Customers

Cost Benefit Net Cost Benefit NetDelivered $422,116,020 ($77,923) $422,038,097 $40,281,543 ($2,303) $40,279,240Received $0 $0 $0 $8,463 ($2,114,488) ($2,106,025)Net $422,038,097 $38,173,215Delivered 7,705,480,273 20,874,873 7,726,355,146 687,572,602 660,154 688,232,757Received 0 0 0 3,313,031 75,383,377 78,696,408Net 7,726,355,146 609,536,348Delivered 0.05478 -0.00373 0.05462 0.05859 -0.00349 0.05853Received 0 0 0 0.00255 -0.02805 -0.02676Net 0.05462 0.06263

Cost Benefit Net Cost Benefit NetDelivered $702,946,146 ($801,037) $702,145,109 $59,080,713 ($21,170) $59,059,544Received $0 $0 $0 $124,278 ($14,627,090) ($14,502,812)Net $702,145,109 $44,556,732Delivered 21,819,548,890 173,851,867 21,993,400,757 1,743,997,189 4,716,293 1,748,713,482Received 0 0 0 32,607,957 570,831,752 603,439,709Net 21,993,400,757 1,145,273,773Delivered 0.03222 -0.00461 0.03193 0.03388 -0.00449 0.03377Received 0 0 0 0.00381 -0.02562 -0.02403Net 0.03193 0.0389

Cost Benefit Net Cost Benefit NetGMCR Delivered $35,963,712 ($1,626,020) $34,337,693 $1,200,846 ($41,604) $1,159,241

Received $0 $0 $0 $269,584 ($3,618,009) ($3,348,426)Net $34,337,693 ($2,189,184)Delivered 2,503,929,886 355,561,063 2,859,490,949 79,719,080 9,363,881 89,082,961Received 0 0 0 60,962,717 273,612,681 334,575,398Net 2,859,490,949 -245,492,437Delivered 0.01436 -0.00457 0.01201 0.01506 -0.00444 0.01301Received 0 0 0 0.00442 -0.01322 -0.01001Net 0.01201 -0.00892

GMCR per kWh($/kWh)

Super Off-peak: Non-NEM Super Off-peak: NEM

kWh

GMCR per kWh($/kWh)

GMCR per kWh($/kWh)

Winter Off-peak: Non-NEM Winter Off-peak: NEM

GMCR

kWh

Winter Peak: Non-NEM Winter Peak: NEM

GMCR

kWh

(PG&E-2)

3-37

TABLE 3-9B COST OF SERVICE COMPARISON BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS

TOTAL GENERATION, 2021 WINTER

4. Cost of Service Comparison by Customer Class 1

To provide a comparison of the total GMCR per kWh at the customer 2

class level, Tables 3-10A and 3-10B present the annual level comparison of 3

Total GMCR, kWh and GMCR/kWh separately for Residential, Commercial, 4

Agricultural, and Large Commercial and Industrial customers. Table 3-10A 5

shows the results based on 2017 recorded data and Table 3-10B shows the 6

results based on the 2021 forecast. The following subsections discuss 7

these customer class-specific comparisons. 8

a. Cost of Service Differences Within Residential Customer Class 9

Table 3-10A shows that, for the Residential customer class, there is 10

a difference in GMCR/kWh of 4.528 cents per kWh at annual level 11

between Non-NEM and NEM customer groups, based on the 2017 12

recorded data. 4.528 cents per kWh is the greatest difference between 13

Cost Benefit Net Cost Benefit NetDelivered $291,616,757 ($145,064) $291,471,693 $67,532,543 ($13,829) $67,518,714Received $0 $0 $0 $26,872 ($1,867,748) ($1,840,876)Net $291,471,693 $65,677,838Delivered 4,145,066,731 42,447,167 4,187,513,898 856,951,107 3,986,769 860,937,876Received 0 0 0 7,271,064 93,108,043 100,379,107Net 4,187,513,898 760,558,768Delivered 0.07035 -0.00342 0.0696 0.07881 -0.00347 0.07842Received 0 0 0 0.0037 -0.02006 -0.01834Net 0.0696 0.08635

Cost Benefit Net Cost Benefit NetDelivered $388,081,399 ($2,241,403) $385,839,996 $81,815,773 ($167,935) $81,647,838Received $0 $0 $0 $628,558 ($10,125,818) ($9,497,259)Net $385,839,996 $72,150,579Delivered 11,462,460,635 440,712,398 11,903,173,033 2,162,894,141 34,096,154 2,196,990,295Received 0 0 0 116,147,015 616,175,044 732,322,059Net 11,903,173,033 1,464,668,235Delivered 0.03386 -0.00509 0.03241 0.03783 -0.00493 0.03716Received 0 0 0 0.00541 -0.01643 -0.01297Net 0.03241 0.04926

Cost Benefit Net Cost Benefit NetGMCR Delivered $3,598,759 ($4,696,381) ($1,097,622) $301,737 ($327,318) ($25,581)

Received $0 $0 $0 $1,467,669 ($737,255) $730,414Net ($1,097,622) $704,833Delivered 788,540,705 790,146,897 1,578,687,602 62,808,264 55,279,634 118,087,898Received 0 0 0 232,431,528 173,351,931 405,783,458Net 1,578,687,602 -287,695,560Delivered 0.00456 -0.00594 -0.0007 0.0048 -0.00592 -0.00022Received 0 0 0 0.00631 -0.00425 0.0018Net -0.0007 0.00245

Winter Peak: Non-NEM Winter Peak: NEMWinter Total Generation Cost of Service of All Bundled Customers

GMCR per kWh($/kWh)

GMCR

kWh

Winter Off-peak: NEM

GMCR

kWh

Winter Off-peak: Non-NEM

GMCR per kWh($/kWh)

Super Off-peak: NEM

kWh

GMCR per kWh($/kWh)

Super Off-peak: Non-NEM

(PG&E-2)

3-38

NEM and Non-NEM customers of all four customer classes shown, 1

causing NEM Total GMCR/kWh to be 90 percent higher than that for 2

Non-NEM. The results are similar when using 2021 forecast data, as 3

shown in Table 3-10B. The 2021 forecast shows that NEM customers 4

will have a net total GMCR/kWh of 14.65 cents per kWh, which is 5

8.4 cents per kWh higher than the net GMCR/kWh of Non-NEM 6

customers representing a 135 percent difference. 7

It is important to discuss what causes Residential NEM customers to 8

have a significantly higher annual level GMCR/kWh than that of 9

Residential Non-NEM customers. Both energy and capacity 10

components of marginal costs contribute to this difference in total 11

GMCR/kWh. The energy portion shows a difference of 2.98 cents per 12

kWh based on 2017 recorded data, and a difference of 3.84 cents per 13

kWh based on the 2021 forecast.16 As for the capacity portion, the 14

difference is 1.54 cents per kWh in 2017, while the 2021 forecasted 15

difference is 4.59 cents per kWh. Unlike the Non-Residential classes’ 16

loads, Residential load peaks during the high cost summer and winter 17

peak periods starting at 4 p.m. and ending at 9 p.m., while solar 18

generation takes place mostly during the relatively low-cost daytime 19

hours and tapers off in the period before sundown. This pattern drives 20

the large difference in GMCR/kWh between the Residential NEM and 21

Non-NEM customers.17 22

b. Cost of Service Difference Within Commercial Customers 23

For the Commercial class, the GMCR/kWh of NEM customers are 24

higher than the GMCR/kWh of Non-NEM customers by 1.05 cents per 25

kWh in 2017 and is forecasted to be 2.6 cents per kWh higher in 2021. 26

This difference is significant, but lower than the difference observed for 27

the Residential class. The results, based on 2017 actual data and 2021 28

forecasts, are shown in Table 3-10A and Table 3-10B, respectively. 29

16 These details will be available in the related workpapers. 17 Section 4.e of this chapter provides additional explanation.

(PG&E-2)

3-39

TABLE 3-10A CUSTOMER CLASS SPECIFIC COST OF SERVICE COMPARISON

BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS TOTAL GENERATION, 2017 ANNUAL

Annual Total Generation Cost of Service of Residential Customers

Cost Benefit Net Cost Benefit NetDelivered $1,129,600,402 ($1,215,189) $1,128,385,214 $128,957,607 ($31,185) $128,926,422Received $0 $0 $0 $321,276 ($33,721,432) ($33,400,156)Net $1,128,385,214 $95,526,265Delivered 22,058,421,065 249,172,513 22,307,593,578 2,232,649,438 6,696,153 2,239,345,591Received 0 0 0 76,857,786 1,166,020,490 1,242,878,276Net 22,307,593,578 996,467,315Delivered 0.05121 -0.00488 0.05058 0.05776 -0.00466 0.05757Received 0 0 0 0.00418 -0.02892 -0.02687Net 0.05058 0.09586

Annual Total Generation Cost of Service of Commercial Customers

Cost Benefit Net Cost Benefit NetDelivered $870,153,126 ($1,054,345) $869,098,781 $54,322,172 ($23,776) $54,298,396Received $0 $0 $0 $101,962 ($7,888,369) ($7,786,408)Net $869,098,781 $46,511,989Delivered 19,693,594,775 227,324,761 19,920,919,535 1,145,379,196 5,139,360 1,150,518,556Received 0 0 0 23,346,891 268,183,703 291,530,594Net 19,920,919,535 858,987,962Delivered 0.04418 -0.00464 0.04363 0.04743 -0.00463 0.04719Received 0 0 0 0.00437 -0.02941 -0.02671Net 0.04363 0.05415

Annual Total Generation Cost of Service of Agricultural Customers

Cost Benefit Net Cost Benefit NetDelivered $209,950,474 ($207,411) $209,743,063 $15,824,932 ($5,442) $15,819,491Received $0 $0 $0 $35,269 ($3,705,283) ($3,670,015)Net $209,743,063 $12,149,476Delivered 4,633,162,624 45,339,942 4,678,502,566 331,021,215 1,179,923 332,201,138Received 0 0 0 8,459,136 120,852,084 129,311,220Net 4,678,502,566 202,889,918Delivered 0.04531 -0.00457 0.04483 0.04781 -0.00461 0.04762Received 0 0 0 0.00417 -0.03066 -0.02838Net 0.04483 0.05988

Annual Total Generation Cost of Service of Large Commercial and Industrial Customers

Cost Benefit Net Cost Benefit NetDelivered $291,984,662 ($380,468) $291,604,195 $21,164,848 ($14,907) $21,149,941Received $0 $0 $0 $4,078 ($306,842) ($302,765)Net $291,604,195 $20,847,176Delivered 7,219,946,311 84,336,632 7,304,282,943 481,994,887 3,368,446 485,363,332Received 0 0 0 987,099 10,573,199 11,560,297Net 7,304,282,943 473,803,035Delivered 0.04044 -0.00451 0.03992 0.04391 -0.00443 0.04358Received 0 0 0 0.00413 -0.02902 -0.02619Net 0.03992 0.04400

Non-NEM NEM

GMCR

kWh

GMCR per kWh($/kWh)

Non-NEM NEM

GMCR

kWh

GMCR per kWh($/kWh)

Non-NEM NEM

GMCR

kWh

GMCR per kWh($/kWh)

Non-NEM NEM

GMCR

kWh

GMCR per kWh($/kWh)

(PG&E-2)

3-40

TABLE 3-10B CUSTOMER CLASS SPECIFIC COST OF SERVICE COMPARISON

BETWEEN NON-NEM AND NEM BUNDLED CUSTOMERS TOTAL GENERATION, 2021 ANNUAL

c. Cost of Service Difference Within Agricultural Customers 1

For the Agricultural class, the GMCR/kWh of NEM customers is 2

higher than that of Non-NEM customers by 1.5 cents per kWh in 2017 3

and forecasted to be 4.1 cents per kWh higher in 2021. This difference 4

is significant, but lower than the difference observed in the Residential 5

Cost Benefit Net Cost Benefit NetDelivered $764,143,182 ($3,225,755) $760,917,427 $216,935,986 ($1,121,000) $215,814,986Received $0 $0 $0 $1,449,337 ($21,259,996) ($19,810,659)Net $760,917,427 $196,004,328Delivered 11,678,191,118 539,118,831 12,217,309,949 2,677,450,727 21,807,349 2,699,258,076Received 0 0 0 252,927,304 1,108,830,648 1,361,757,953Net 12,217,309,949 1,337,500,123Delivered 0.06543 -0.00598 0.06228 0.08102 -0.0514 0.07995Received 0 0 0 0.00573 -0.01917 -0.01455Net 0.06228 0.14655

Cost Benefit Net Cost Benefit NetDelivered $486,677,819 ($2,484,981) $484,192,838 $82,855,159 ($485,516) $82,369,643Received $0 $0 $0 $416,652 ($4,997,547) ($4,580,895)Net $484,192,838 $77,788,748Delivered 8,896,299,386 468,881,734 9,365,181,120 1,283,467,904 21,007,614 1,304,475,517Received 0 0 0 65,617,133 237,267,450 302,884,583Net 9,365,181,120 1,001,590,934Delivered 0.05471 -0.0053 0.0517 0.06456 -0.02311 0.06314Received 0 0 0 0.00635 -0.02106 -0.01512Net 0.0517 0.07767

Cost Benefit Net Cost Benefit NetDelivered $226,593,090 ($1,089,923) $225,503,167 $65,505,375 ($92,593) $65,412,782Received $0 $0 $0 $366,963 ($5,376,603) ($5,009,640)Net $225,503,167 $60,403,143Delivered 3,881,495,967 208,465,051 4,089,961,018 930,350,310 18,920,808 949,271,118Received 0 0 0 60,401,634 262,589,097 322,990,731Net 4,089,961,018 626,280,387Delivered 0.05838 -0.00523 0.05514 0.07041 -0.00489 0.06891Received 0 0 0 0.00608 -0.02048 -0.01551Net 0.05514 0.09645

Cost Benefit Net Cost Benefit NetDelivered $196,804,356 ($994,559) $195,809,797 $21,968,681 ($60,305) $21,908,377Received $0 $0 $0 $38,314 ($514,546) ($476,232)Net $195,809,797 $21,432,145Delivered 3,795,094,542 191,193,081 3,986,287,623 372,359,442 12,117,735 384,477,177Received 0 0 0 6,018,888 20,765,379 26,784,267Net 3,986,287,623 357,692,910Delivered 0.05186 -0.0052 0.04912 0.059 -0.00498 0.05698Received 0 0 0 0.00637 -0.02478 -0.01778Net 0.04912 0.05992

kWh

GMCR per kWh($/kWh)

Annual Total Generation Cost of Service of Large Commercial and Industrial CustomersNon-NEM NEM

GMCR

GMCR

kWh

GMCR per kWh($/kWh)

GMCR per kWh($/kWh)

Annual Total Generation Cost of Service of Agricultural CustomersNon-NEM NEM

Annual Total Generation Cost of Service of Residential CustomersNon-NEM NEM

GMCR

GMCR

kWh

kWh

GMCR per kWh($/kWh)

Annual Total Generation Cost of Service of Commercial CustomersNon-NEM NEM

(PG&E-2)

3-41

class. The results, based on 2017 data and 2021 forecast, are shown in 1

Table 3-10A and Table 3-10B, respectively. 2

d. Cost of Service Difference Within Large Commercial and Industrial 3

Customers 4

For the Large Commercial and Industrial class, the GMCR/kWh for 5

NEM customers are higher than that of Non-NEM customers by 6

0.4 cents per kWh in 2017 and forecast to be 1.1 cents per kWh higher 7

in 2021. This difference is significant, but much lower than the 8

difference observed in the Residential class. The results, based on 9

2017 data and 2021 forecast, are shown in Table 3-10A and 10

Table 3-10B, respectively. 11

e. Additional Explanation for the Difference in GMCR/kWh between 12

Non-NEM and NEM 13

To get insight on what drives the difference in cost of service 14

between Non-NEM and NEM customers, we need to review the 15

expressions of Net GMCR per kWh for Non-NEM and NEM customers 16

as shown below: 17

For Non-NEM customers: 18 ℎ = " " " " ℎ 19

For NEM customers: 20 ℎ = " " + " " (" " ℎ − " " ℎ) 21

As apparent from the equation for Non-NEM customers shown 22

above, Non-NEM customers have only “delivered” load and no 23

“received” load, which makes their Net GMCR per kWh to be the ratio of 24

their “delivered” GMCR to their “delivered” kWh. However, NEM 25

customers can have both “delivered” and “received” load, so their Net 26

GMCR per kWh is calculated by dividing the sum of the “delivered” 27

GMCR and “received” GMCR, as shown in the numerator above, by the 28

(PG&E-2)

3-42

absolute value18 of the net of ”delivered” kWh and “received” kWh, as 1

shown in the denominator above. Presence of both “delivered” and 2

“received” components creates complexity and drives the differences in 3

the Net GMCR per kWh values between NEM and Non-NEM customers. 4

Additionally, during which hours the ”delivered” load and the “received” 5

load occur, determines how different the GMCR/kWh will be between 6

Non-NEM and NEM customers. The following points explain this point: 7

Effect of “received” kWh: The subtraction of the “received” kWh in 8

the denominator of the net GMCR per kWh calculation reduces the 9

denominator’s value, thereby increasing the Net GMCR per kWh 10

value for NEM customers. The greater the ratio of the “received” 11

load to the net load for a NEM customer, the greater this effect will 12

be for that customer. For example, Residential NEM customers 13

have a higher proportion of “received” load (50 percent of net load in 14

2017) than any other customer class and, therefore, the Net GMCR 15

per kWh differential between NEM and Non-NEM customers is 16

expected to be the highest for the Residential customer class 17

compared to the rest of the customer classes. 18

Effect of “received” GMCR: GMCR, as has been described before, 19

is calculated on an hourly basis incurring either a cost (positive 20

GMCR) or a benefit (negative GMCR) for each hour. The net effect 21

of that is typically a negative GMCR for the “received” load. This 22

“received” GMCR is also a lot smaller (in absolute value) than the 23

“delivered” GMCR since the “received” load typically occurs during 24

relatively low-cost hours of the day. Therefore, when the “received” 25

GMCR (which is usually negative) is added to the “delivered” GMCR 26

in the numerator of the Net GMCR per kWh calculation, the 27

numerator will typically decrease by only a small amount. Since the 28

denominator in the Net GMCR per kWh calculations for NEM 29

customers typically reduces by a significant amount (for example, by 30

50% for residential NEM customers), the relatively smaller reduction 31

18 The absolute value of the denominator in the GMCR per kWh calculation is taken to

preserve the sign of the GMCR in the numerator which corresponds to “cost” when positive and “benefit” when negative.

(PG&E-2)

3-43

in the numerator caused by the “received” GMCR still results in a 1

higher Net GMCR per kWh for NEM customers compared to the 2

“delivered” GMCR per kWh for Non-NEM customers and even 3

compared to that of the NEM customers themselves. 4

Effect of “received” load on the “delivered” GMCR for NEM 5

Customers: Load for the Residential class peaks during the late 6

afternoon and evening period, which are also the highest cost hours. 7

The “received” load timing is independent of customers class; 8

occurring during the day (early morning till late afternoon). 9

Therefore, the “received” load of the Residential class offsets 10

relatively low-usage hours of the day, which are also relatively less 11

costly, causing the proportion of the daily load during the high-cost 12

hours to be large. As a result, the net GMCR per kWh increases 13

significantly for the NEM customers, when compared to Non-NEM 14

customers. 15

The load profile for the Commercial class is significantly 16

different from the load profile of Residential class. For the 17

Commercial class, load is the highest during the daytime hours. 18

Therefore, the “received” load that occur during the day-time can 19

offset a relatively large proportion of daily load, thereby benefitting 20

the customers more in comparison to the Residential customers. 21

Consequently, the difference in GMCR per kWh between Non-NEM 22

and NEM customers are smaller than what we observe in the case 23

of Residential class. 24

E. Conclusion 25

PG&E is seeking CPUC’s approval for the methodology described in this 26

chapter and its results summarized as the MCR per kWh estimates in the 27

Marginal Cost Table19 used for revenue allocation. PG&E has found the 28

following key conclusions for its bundled customers (excluding those in the 29

Standby and Streetlights customer classes): 30

1) A significant difference in total GMCR per kWh between the NEM and Non-31

NEM customers existed in 2017 and is forecasted to exist in 2021. The total 32

19 The Marginal Cost Table is included in the workpapers for Exhibit (PG&E-2).

(PG&E-2)

3-44

GMCR/kWh for NEM customers is consistently higher than the GMCR/kWh 1

for Non-NEM customers. 2

2) Higher total GMCR/kWh for the NEM customers is driven by the 3

phenomenon that solar generation happens during the daytime, while the 4

CAISO’s ANL as well as the generation MEC peaks during the peak period 5

starting at 4 p.m. and ending at 9 p.m. (Note that CAISO’s ANL was used to 6

allocate generic capacity cost at hourly level.)20 7

3) When net usage is considered, the Residential class shows the biggest 8

difference in total GMCR/kWh between NEM and Non-NEM customers. The 9

difference is 4.528 cents per kWh in 2017 and is forecasted to be 8.4 cents 10

per kWh in 2021. The GMCR/kWh for NEM customers in other customer 11

classes is higher, as well, compared to the GMCR/kWh for Non-NEM 12

customers. 13

4) In general, greater the proportion of load offset by NEM customer-owned 14

generation, greater the difference in total GMCR/kWh between NEM and 15

Non-NEM customers. 16

5) When “delivered” and “received” usages are considered separately, the 17

following is observed: 18

a. Analysis based on 2017 recorded data shows Total GMCR per kWh for 19

NEM customers to be higher than that of Non-NEM customers for the 20

“delivered” usage for the Residential, Commercial, Agricultural, and 21

Large Commercial and Industrial customer classes. This happens 22

because NEM customers’ “delivered” usage occurs primarily during the 23

summer peak and winter peak hours when the hourly MEC and hourly 24

allocated MGCC are the highest, whereas Non-NEM customers 25

“delivered” usage occurs throughout all 24 hours. This finding holds for 26

the total GMCR/kWh estimates based on 2021 forecast, as well. 27

b. The “received” kWh from NEM customers happens mostly during 28

summer part-peak and off-peak hours, and during winter off-peak and 29

super off-peak hours, when the hourly MEC and hourly allocated MGCC 30

are relatively low in comparison to the summer peak and winter peak 31

20 Generic capacity costs are allocated to the highest ANL hours that have ANL above the

threshold of 80 percent of max hourly ANL in a year.

(PG&E-2)

3-45

hours. As a result, total GMCR/kWh is lower for “received” kWh than for 1

“delivered” load. This is true for all of the analyzed customer classes, 2

based on 2017 recorded data as well as 2021 forecasted data. 3

In summary, NEM customers (especially those that offset a large 4

proportion of their load) clearly have higher total MCR per kWh than 5

Non-NEM customers, because NEM customers’ “delivered” usage 6

occurs mostly during the highest cost hours, while “received” kWh 7

occurs during other hours. 8

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 3

ATTACHMENT A

GENERATION PEAK CAPACITY ALLOCATION FACTOR

ANALYSIS

(PG&E-2)

3-AtchA-1

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 3

ATTACHMENT A GENERATION PEAK CAPACITY ALLOCATION FACTOR ANALYSIS

A. Introduction This attachment describes Pacific Gas and Electric Company’s (PG&E)

methodology to calculate generation Peak Capacity Allocation Factors (PCAF) which are hourly factors. These hourly factors are used to compute customer class-specific generation PCAF-weighted loads that are needed for allocating the generation capacity cost responsibilities across all customer classes. As in PG&E’s 2017 General Rate Case Phase II, PG&E continues to use the Adjusted Net Load (ANL), defined as the gross load minus the renewable (solar and wind), hydro and nuclear generation resources, to identify the PCAF hours. Only the PCAF hours are subject to capacity cost allocations; the rest of the hours in a year get zero capacity cost allocation.

B. Methodology Calculation of generation PCAF-weighted loads for different customer classes is

performed in following steps: 1) Identify the PCAF hours: Generation PCAF hours are the hours during which

the ANL is above 80 percent of the maximum hourly ANL in a calendar year. Based on this 80 percent cut-off, if more than 800 hours are identified as PCAF hours, then only the top 800 hours are considered as PCAF hours. Alternatively, if less than 10 hours are identified as PCAF hours based on 80 percent cut-off, then the top 10 hours are considered as PCAF hours.

2) Assigning appropriate weights to the identified PCAF hours in Step 1: For each identified PCAF hour in the previous step, the corresponding PCAF weight ( ) is calculated using the following formula: = −∑ ( − ) Where: : ANL load at PCAF hour i, : The total number of identified PCAF hours,

(PG&E-2)

3-AtchA-2

: 80 percent of ANL maximum load. If less than 10 PCAF hours are identified in Step 1, TL is the 11th largest system load.

3) Calculating Generation PCAF-weighted load for each customer class: Generation PCAF-weighted load for each customer class by TOU period is calculated as the weighted sum of the corresponding hourly customer class load at the identified PCAF hours that fall within respective TOU periods.

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 4

DEFERRABLE TRANSMISSION CAPACITY PROJECTS

(PG&E-2)

4-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 4

DEFERRABLE TRANSMISSION CAPACITY PROJECTS

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 4-1

B. PG&E’s 2019 Electric Transmission Grid Expansion Plan ............................... 4-1

C. Screening Process ........................................................................................... 4-3

D. Results ............................................................................................................. 4-4

E. Conclusion ........................................................................................................ 4-4

(PG&E-2)

4-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 4 2

DEFERRABLE TRANSMISSION CAPACITY PROJECTS 3

A. Introduction 4

This chapter describes the process Pacific Gas and Electric Company 5

(PG&E) used to identify projects contained in PG&E’s 2019 Electric 6

Transmission Grid Expansion Plan1 that can be deferred if electric demand 7

growth fails to materialize as projected over the 10-year study period that is 8

discussed in Section B below. In this chapter, Section D also provides the list of 9

projects from the grid expansion plan that are identified as deferrable using the 10

screening process described below. 11

Deferrable transmission projects are effectively demand-related marginal 12

transmission investments. These projects are a key input into the calculation of 13

PG&E’s marginal transmission capacity cost, which is discussed in Chapter 5 of 14

Exhibit (PG&E-2), “Transmission Cost of Service.” As part of PG&E’s annual 15

planning process, PG&E coordinates with the California Independent System 16

Operator (CAISO) to discuss the status of projects previously approved by the 17

CAISO. During these meetings, PG&E notifies the CAISO of any projects that 18

have been deferred and the reasoning for deferral. 19

The remainder of this chapter is organized as follows: 20

Section B – PG&E’s 2019 Electric Transmission Grid Expansion Plan 21

Section C – Screening Process 22

Section D – Results 23

Section E – Conclusion 24

B. PG&E’s 2019 Electric Transmission Grid Expansion Plan 25

The PG&E transmission grid consists of about 18,500 miles of electric 26

transmission lines and cables and 200 high voltage substations with nominal 27

voltages of 500, 230, 115, 70 and 60 kilovolts (kV). 28

As the registered Transmission Planner for PG&E’s transmission system, 29

PG&E is required to annually develop and submit a transmission plan that 30

1 PG&E’s 2019 Electric Transmission Grid Expansion Plan is available upon request to

Parties who sign a non-disclosure agreement.

(PG&E-2)

4-2

addresses the near term (years one through five) and longer term (years six 1

through ten) planning horizons to the CAISO Corporation for approval. 2

Development of this transmission plan is coordinated in accordance with the 3

CAISO’s Section 24 Tariff, as well as the business rules set forth in the CAISO 4

Business Practice Manual for CAISO Transmission Planning Process (TPP). 5

The CAISO’s TPP is an annual integrated, open, participatory, and transparent 6

process that focuses on ensuring reliable, economically efficient, and 7

nondiscriminatory use of the transmission system. Specifically, the CAISO’s 8

TPP entails three phases: (a) development of Unified Planning Assumptions 9

and CAISO Study Plan; (b) technical studies and transmission plan 10

development;2 and (c) if needed, competitive solicitation to build and own 11

transmission projects identified in the CAISO Board-approved plan. 12

Development of PG&E’s annual transmission system assessment and 13

transmission expansion plan is coordinated along the CAISO’s TPP. PG&E is 14

required to participate and submit annually electric transmission facility 15

expansion plans to address identified reliability concerns. PG&E’s transmission 16

expansion plan is submitted to the CAISO on an individual project basis during 17

the annually open Request Window. PG&E’s plan identifies the electric 18

transmission facilities in PG&E’s service territory that are projected to not meet 19

North American Electric Reliability Corporation, Western Electricity Coordinating 20

Council, and CAISO Planning Standards and Criteria during the 10-year 21

planning horizon. As part of the CAISO TPP, market participants are 22

encouraged to participate and provide comments and input on PG&E’s proposed 23

transmission plans. 24

PG&E’s latest transmission expansion plan is documented in the “PG&E 25

2019 Electric Transmission Grid Expansion Plan,” dated March 2019. PG&E’s 26

2019 plan covers the 10-year planning horizon from 2019 to 2028. It identifies 27

specific transmission expansion upgrades needed in the near-term (5-year 28

period) and in the long-term (10-year period) to address the applicable reliability 29

standards and criteria. The new PG&E projects identified in this cycle to meet 30

2 The second phase in the TPP entails the following activities: performing technical

studies to assess system reliability and various other purposes; documenting technical study results and development of transmission proposals; and development of the annual CAISO transmission plan.

(PG&E-2)

4-3

reliability and regulatory obligations were submitted to the CAISO in 1

September 2018. Then, as part of the TPP process, the CAISO staff evaluated 2

all received project proposals and developed a Transmission Plan 3

recommending the approval of those projects found to be most effective at 4

mitigating the identified transmission problems. The CAISO Transmission Plan 5

was then submitted to the CAISO Board for approval. The PG&E projects that 6

CAISO recommended for approval are included in the CAISO Transmission 7

Plan. The CAISO Board subsequently approved the various project proposals 8

identified in the CAISO Transmission Plan in March 2019. 9

PG&E submitted seven new project proposals during the CAISO’s Request 10

Window for approval, and eight CAISO-generated projects were added to the 11

PG&E plan. With PG&E’s previous CAISO-approved projects, there are now 12

88 transmission projects in PG&E’s 2019 Electric Transmission Grid Expansion 13

Plan that will allow PG&E to continue to be in full compliance with all applicable 14

transmission planning criteria and standards. These projects are expected to 15

require a capital investment of $2.8 billion to $3.5 billion, depending on final 16

project cost, scope and schedule. PG&E’s 2019 Electric Transmission Grid 17

Expansion Plan also identifies several previously approved projects that have 18

had their scope revised or have been cancelled by the CAISO in the 2018-2019 19

TPP as a result of system condition changes, including, the load forecast. 20

C. Screening Process 21

In 2019, PG&E evaluated the 73 previously approved projects in its 2019 22

Electric Transmission Grid Expansion Plan to determine whether a project could 23

be deferred if electric demand growth in the area served by the project failed to 24

materialize as projected over the 10-year study period. The following principles 25

were applied in the evaluation: 26

1. Projects needed to meet regulatory, contractual, or safety requirements are 27

non-deferrable. 28

2. Projects that improve system efficiency, such as those that reduce Local 29

Capacity Adequacy Requirements or cost-effectively reduce customer 30

outage time, are non-deferrable. 31

3. Projects that address greater than 10 percent capacity deficiency are 32

non-deferrable. 33

(PG&E-2)

4-4

4. Projects requiring to be initiated as soon as possible to mitigate an existing 1

thermal or voltage problem are non-deferrable. 2

D. Results 3

Of the 73 previously approved projects in PG&E’s 2019 Electric 4

Transmission Grid Expansion Plan, six projects were found to be deferrable. 5

These six projects are listed in Table 4-1 below. 6

TABLE 4-1 DEFERRABLE TRANSMISSION CAPACITY PROJECTS

Line No. Project Title

Cost Range (Million)

Targeted In-

Service Date

Can Be Deferred by 5-10% Demand

Reduction

1 Pittsburg 230/115 kV Transformer Capacity Increase $10-$20 May-22 Yes 2 Ignacio Area Upgrade Project $30-$40 Dec-23 Yes 3 Vierra 115 kV Looping Project $40-$50 Feb-23 Yes 4 Cascade 115/60 kV No.2 Transformer Project $40-$60 Jan-22 Yes 5 Vaca Dixon - Lakeville Corridor Series Compensation Project $20-$30 Aug-22 Yes 6 Wheeler Ridge Voltage Support $40-$50 Apr-21 Yes

E. Conclusion 7

PG&E develops its Electric Transmission Grid Expansion Plan on an annual 8

basis. The process allows for continuing review of the need for identified system 9

facilities during the 10-year planning horizon. If a given facility load is reduced 10

considerably prior to the implementation date due to system or demand 11

forecasts changes, then the implementation date of the project may be deferred 12

accordingly. Of the projects developed in PG&E’s 2019 Electric Transmission 13

Grid Expansion Plan, six projects were found to be deferrable if the forecasted 14

load does not materialize. 15

The six transmission projects found to be deferrable using the process 16

discussed in this chapter are projects mainly driven by demand, and are not 17

required by regulatory, safety, contractual, efficiency or other reasons listed in 18

the screening criteria in Section C above. Therefore, these deferrable 19

transmission projects are the marginal investments from which PG&E can 20

estimate the marginal transmission capacity cost. The calculation of PG&E’s 21

marginal transmission capacity cost is discussed in Chapter 5, “Marginal 22

Transmission Capacity Costs and Cost Causation Study,” of Exhibit (PG&E-2). 23

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 5

MARGINAL TRANSMISSION CAPACITY COSTS AND

COST CAUSATION STUDY

(PG&E-2)

5-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 5

MARGINAL TRANSMISSION CAPACITY COSTS AND COST CAUSATION STUDY

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 5-1

B. MTCC Method and Estimate ............................................................................ 5-2

C. Transmission Cost Causation Study ................................................................. 5-3

D. Summary of Findings and Conclusions ............................................................ 5-8

(PG&E-2)

5-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 5 2

MARGINAL TRANSMISSION CAPACITY COSTS AND 3

COST CAUSATION STUDY 4

A. Introduction 5

This chapter presents Pacific Gas and Electric Company’s (PG&E) 6

demand-related marginal transmission capacity cost (MTCC) estimate. PG&E’s 7

transmission rates are regulated by the Federal Energy Regulatory Commission 8

(FERC). Therefore, an MTCC estimate is not required for rate-setting in this 9

proceeding. Although PG&E’s retail transmission rates are regulated by the 10

FERC1 and set on an embedded cost basis, PG&E requires an MTCC estimate 11

for purposes other than determining electric transmission marginal cost 12

revenues and class allocations of electric transmission revenue requirements for 13

rate setting. In particular, PG&E requires an MTCC estimate for use in marginal 14

cost price floor calculations and for use in other proceedings where an MTCC 15

estimate may be required. The result of PG&E’s estimate of MTCC is a value of 16

$12.46 per kilowatt (kW) per year. 17

In addition to PG&E’s MTCC estimate, the California Public Utilities 18

Commission’s (CPUC) final decision in PG&E’s 2017 General Rate Case (GRC) 19

Phase II2 ordered that PG&E file a transmission cost causation study examining 20

the appropriate allocation of transmission costs between non-coincident demand 21

charges and system-peak demand charges in PG&E’s 2020 GRC Phase II 22

application. Accordingly, PG&E conducted such a study. The details and 23

findings of the study are presented in this chapter. 24

The remainder of this chapter is organized as follows: 25

Section B – MTCC Method and Estimate 26

Section C – Transmission Cost-Causation Study 27

Section D – Summary of Findings and Conclusions 28

1 PG&E filed its first case, Docket No. ER97-2358, in 1997 at the FERC. 2 Decision (D.) 18-08-013, Ordering Paragraph 6.

(PG&E-2)

5-2

B. MTCC Method and Estimate 1

PG&E estimated a systemwide forward-looking MTCC using the Discounted 2

Total Investment Method (DTIM)3 with a test year of 2021. The DTIM 3

calculation includes an adjustment for the time value of money and calculates an 4

average investment cost for streams of lumpy4 investments needed to serve 5

additional loads. The MTCC, using the DTIM, is calculated in two basic steps. 6

The first step is to calculate the marginal investment per megawatt (MW) of 7

demand. The second step is to calculate an annualized marginal cost in dollars 8

per MW per year using a Real Economic Carrying Charge (RECC) factor with 9

marginal cost loaders and financial factors. 10

The first step for application of the DTIM to a marginal capacity cost 11

calculation is the calculation of the marginal investment: the marginal investment 12

is the present value of deferrable investments divided by the present value of 13

forecasted load growth.5 The result is the marginal investment per MW. The 14

marginal investments are projects in PG&E’s 2019 Electric Transmission Grid 15

Expansion Plan identified as deferrable in Exhibit (PG&E-2), Chapter 4, 16

“Deferrable Transmission Capacity Projects.”6 The forecasted load growth is 17

from PG&E’s Line 09 Peak Forecast MW 1-in-10 recurrence interval used as 18

part of the transmission planning process.7 19

3 The mathematical equation for DTIM is discussed in Exhibit (PG&E-2) Chapter 7,

“Marginal Distribution Capacity Costs.” 4 An investment stream is said to be “lumpy” when there is a large year-to-year variation

in the size of the investments. 5 The apparent discounting of load arises from the discounting of incremental cash flows

needed to invest in capacity where the discounted cash flows are equal to the application of a marginal cost per unit of capacity (essentially a levelized price) applied to the quantities of incremental load capacity. This discounting of load is discussed further in Exhibit (PG&E-2), Chapter 7.

6 As described in Exhibit (PG&E-2), Chapter 4, “Deferrable Transmission Capacity Projects,” PG&E’s 2019 Electric Transmission Grid Expansion Plan was submitted to the California Independent System Operator (CAISO) in September 2018. The CAISO approved various project proposals identified in that plan in March 2019.

7 The systemwide and geographical Line 09 1-in-10 recurrence interval peak MW demand forecast used to calculate demand growth for the DTIM calculation is taken from the 2019 peak demand forecast incorporated in PG&E’s 2019 Electric Transmission Grid Expansion Plan. This forecast was accepted by CAISO for developing a base case load assessment for PG&E’s bulk transmission planning.

(PG&E-2)

5-3

The second step is to annualize the marginal investment. The marginal 1

investment is multiplied by an annualization factor that levelizes the marginal 2

investment in real dollars over its lifetime. The resulting marginal cost reflects 3

the annualized cost of investment (as well as other capital costs) and annual 4

expenses associated with adding capacity to the transmission system. The 5

annualization factor includes a RECC and marginal cost loaders and financial 6

factors for Operations and Maintenance and Administrative and General 7

expenses, common and general plant, cash working capital and franchise fees 8

and uncollectibles. The development of the RECC and the loaders and financial 9

factors is presented in Exhibit (PG&E-2), Chapter 10, “Marginal Cost Loaders 10

and Financial Factors.” 11

C. Transmission Cost Causation Study 12

Today, PG&E’s retail transmission rates, which are regulated by the FERC, 13

are not differentiated by time of day. PG&E charges its smaller customers 14

(i.e., residential and small commercial customers) a constant dollar per 15

kilowatt hour rate and its larger customers (i.e., large commercial and industrial 16

customers) a constant dollar per kW rate for transmission service. In 17

D.18-08-013 the CPUC stated its desire to investigate the extent to which 18

transmission rates could be differentiated by time of day. OP 6 of D.18-08-103 19

states that “Pacific Gas and Electric shall file a transmission cost causation 20

study with its next GRC Phase II application. This study must examine the 21

appropriate allocation of transmission costs between non-coincident demand 22

charges and system peak demand charges.” To answer this question, PG&E 23

analyzed its transmission costs from its most recent Transmission Owner (TO) 24

tariff rate filing TO 20 using a method similar to the method San Diego Gas & 25

Electric Company (SDG&E) presented to the FERC.8 However, PG&E’s 26

method differs from SDG&E’s in some of its details. This section first presents 27

PG&E’s method for determining the causes of its transmission costs and for 28

determining the appropriate allocation of these costs between system-peak 29

demand charges and non-coincident demand charges. Lastly, this section 30

presents the results of PG&E’s analysis. 31

8 Reference SDG&E filing (Docket No. ER19-221-000).

(PG&E-2)

5-4

Like SDG&E’s method, PG&E’s method consists of two main steps. First, 1

PG&E classified its more recent functional transmission plant (that is, 2

transmission investments) into categories describing the principle drivers 3

(i.e., the main purposes of the plant) to determine the percentage of costs driven 4

by capacity requirements. For the second step, PG&E used the results from the 5

first step and customers’ interval loads to examine the appropriate allocation of 6

its transmission revenue requirement among system-peak demand charges and 7

non-coincident demand charges based on the principle of cost-causation. In 8

contrast to SDG&E, PG&E did not use system-level energy usage by 9

time-of-use (TOU) period. Instead, PG&E used its system-level peak capacity 10

allocation factor (PCAF) loads i.e., generation PCAF loads,9 because they more 11

accurately measure the expected relative capacity10 required by customers from 12

the transmission grid during each interval.11 PG&E’s application of the two 13

main steps are described in detail below. 14

Step One 15

For the first step, PG&E analyzed its functional transmission plant in service 16

as of the end of 2017 and forecasted transmission plant additions for 2018 and 17

2019 as shown in its TO20 workpapers.12 The total cost of PG&E’s functional 18

transmission plant in service as of the end of 2017 is $10,910,651, which 19

reconciles to the amount of $11,404,072 in box 207.58g of PG&E’s 2017 FERC 20

Form 1 statement. PG&E’s forecasted transmission plant additions for 2018 and 21

2019 are the output of PG&E’s integrated planning process and electric 22

9 See Chapter 3 “Generation Energy and Capacity Cost of Service” for an explanation of

how the generation PCAF loads were calculated. PCAF loads for the transmission system were not developed as a part of this study. Because peak load on the transmission system can vary by transmission planning area, PCAFs developed on that basis would likely show a smaller concentration of costs in the peak period.

10 Generation PCAF loads are a reasonable estimate of the required capacity of the transmission grid because the transmission grid is highly networked and directly connected with generation facilities.

11 PG&E introduced its original PCAF costing methodology in its 1993 GRC, which was litigated before the CPUC. In D.92-12-057, the CPUC adopted PG&E’s proposed PCAF methodology. Once again, PG&E utilized its adopted PCAF methodology in its 1996 GRC.

12 Please refer to the workpapers supporting this chapter for the costs of PG&E’s functional transmission plant in service as of the end of 2017 and the costs of PG&E’s forecasted transmission plant additions for 2018 and 2019.

(PG&E-2)

5-5

transmission capital investment and execution planning process.13 1

Furthermore, all projects to increase capacity are reviewed and approved by 2

CAISO through its transmission planning process. 3

Unlike SDG&E, PG&E did not review the descriptions of its transmission 4

projects to determine their drivers because PG&E already organizes its 5

transmission investments by Major Work Category (MWC) as part of its planning 6

processing. MWC is PG&E’s internal accounting method for identifying the 7

purpose of a project, and, since 2010, PG&E has used it to identify the purposes 8

of capital additions.14 Using MWC, PG&E grouped its in-service functional 9

transmission plant added after 2010 and forecasted plant additions by whether 10

they were built to meet capacity requirements. Because MWCs 60 and 61 11

identify capacity-related plant, PG&E grouped the costs associated with 12

MWC 60 and 61 together and, separately, the costs associated with the other 13

MWCs together. The other MWCs identify investments primarily made for 14

non-capacity reasons but include some of PG&E’s most important investments. 15

For example, safety is a major priority to PG&E. MWCs 3F, 63, 67, and 94 16

identify infrastructure enhancement projects involving system design 17

modifications to improve system safety, reliability, resiliency, performance and/or 18

operational flexibility. Lastly, PG&E divided the capacity-related costs by the 19

total cost to determine the percentage of costs driven by capacity requirements 20

and the percentage of costs driven by non-capacity requirements. 21

Step Two 22

For the second step, PG&E determined how much of its Transmission 23

Revenue Requirement (TRR) to allocate between system-peak demand charges 24

and non-coincident demand charges. First, PG&E multiplied the transmission 25

revenue requirement by the percentage of transmission capacity-related costs 26

calculated in step one to determine the dollar amount of transmission 27

capacity-related costs. Next, in accordance with the established principle of 28

13 PG&E’s integrated planning process and electric transmission capital investment and

execution planning process are the processes by which PG&E systematically identifies the projects to meet transmission system requirements established by regulatory bodies, including the FERC, CPUC, CAISO, North American Electric Reliability Corporation, and Western Electricity Coordinating Council.

14 For a complete description of transmission MWCs and PG&E transmission planning process, see Exhibit (PG&E-2) of PG&E’s TO-20 testimony.

(PG&E-2)

5-6

cost-causation, PG&E assigned the dollar amount of transmission 1

capacity-related costs to be recovered through system-peak demand charges. 2

Recall that system-peak demand charges are charges based on demands 3

during various TOU periods.15 Table 5-1 below lists the definitions of the TOU 4

periods that PG&E is proposing in its 2020 GRC Phase II application.16 To 5

divide up the dollar amount of transmission capacity-related costs among the 6

specific TOU periods, PG&E used it hourly, system-level PCAF loads since they 7

estimate the expected relative peak loads17 on the transmission system. PG&E 8

simply allocated the total dollar amount of transmission capacity-related costs to 9

the TOU periods in direct proportion to the PCAF loads summed up by TOU 10

period. Lastly, PG&E allocated the remainder of the transmission revenue 11

requirement to non-coincident demand charges, once again in accordance with 12

the established principle of cost-causation. In summary, the calculations were 13

as follows: 14

TOU Allocation = TRR ⋅ CP ⋅ ∑(Hourly PCAF Loads in TOU Period) / 15 ∑(Hourly PCAF Loads) 16

where 17

TRR = transmission revenue requirement and 18

CP = capacity %. 19

NCDMD Allocation = TRR ⋅ (1 – CP) 20

where 21

NCDMD = non-coincident demand, 22

TRR = transmission revenue requirement and 23

CP = capacity %. 24

15 The TOU periods are summer on-peak, summer part-peak, summer off-peak, winter

part-peak, winter off-peak and winter super off-peak. 16 See Chapter 11, “Time-Of-Use Period Assessment and Analysis,” of Exhibit (PG&E-2)

for a complete discussion of how the proposed TOU periods were derived. 17 Technically, loads and capacity are different things. However, peak load is a

reasonable estimate of capacity because at least that much capacity is required from the system.

(PG&E-2)

5-7

TABLE 5-1 PROPOSED TOU PERIODS

Line No. TOU Period Definition

1 Summer On-Peak June – September, 4 p.m. – 9 p.m., All days of the week

2 Summer Part-Peak June – September, 2 p.m. – 4 p.m. & 9 p.m. – 11pm, All days of the week

3 Summer Off-Peak June – September, All other hours

4 Winter Part-Peak October – May, 4 p.m. – 9 p.m., All days of the week

5 Winter Super Off-Peak March – May, 9 a.m. – 2 p.m., All days of the week.

6 Winter Off-Peak All other hours

Table 5-2 below shows the costs of post-2010 capital additions in-service as 1

of the end of 2017 by category. As can be seen, capacity-related projects are 2

27.19 percent of the total. 3

TABLE 5-2 IN-SERVICE FUNCTIONAL NETWORK TRANSMISSION PLANT

POST 2010 CAPITAL ADDITIONS – EOY 2017

Line No. Description MWCs Cost ($000s) Percentage

1 Capacity Increase 60, 61 $1,671,012 27.19% 2 Emergency Response and Replacement 65, 92 310,775 5.06% 3 Infrastructure Enhancement 3F, 63, 67, 94 1,416,546 23.05% 4 Infrastructure Replacement – Substations 64, 66, 68, 23 821,651 13.37% 5 Infrastructure Replacement – Lines 70, 71, 72, 93 1,141,538 18.57% 6 Operations Support 5, 12 2,672 0.04% 7 WRO and New Customer Interconnection 82 640,391 10.42% 8 Build IT Applications and Infrastructure 2F 82 0.00% 9 Other 141,460 2.30%

10 Total $6,146,126 100.00%

Table 5-3 below shows the costs of forecasted plant additions for 2018 and 4

2019 as of the end of 2019. Similarly, capacity-related projects are only 5

27.33 percent of the total. 6

(PG&E-2)

5-8

TABLE 5-3 FORECASTED PLANT ADDITIONS

INCREMENTAL 2018 AND 2019 – EOY 2019

Line No. Description MWCs Cost ($000s) Percentage

1 Capacity Increase 60, 61 $611,270 27.33% 2 Emergency Response and Replacement 65, 92 164,032 7.34% 3 Infrastructure Enhancement 3F, 63, 67, 94 581,701 26.01% 4 Infrastructure Replacement - Substations 64, 66, 68, 23 261,041 11.67% 5 Infrastructure Replacement - Lines 70, 71, 72, 93 521,234 23.31% 6 Operations Support 5, 12 – 0.00% 7 WRO and New Customer Interconnection 82 96,959 4.34% 8 Build IT Applications and Infrastructure 2F – 0.00%

9 Total $2,236,238 100.00%

Table 5-4 below presents the allocation of the transmission revenue 1

requirement between system-peak demand charges and non-coincident 2

demand charges. 3

TABLE 5-4 ALLOCATION OF THE REVENUE REQUIREMENT (RRQ) TO SYSTEM PEAK AND

NON-COINCIDENT DEMAND CHARGES

Line No. Description

RRQ ($000s)

PCAF Percentage PCAF Load

1 Retail Base Transmission revenue requirement $1,963,060 2 Percentage Capacity-Related 27.26% 3 Capacity-Related revenue requirement $535,158 4 Summer On-Peak $475,891 88.93% 21,860,109 5 Summer Part-Peak $58,100 10.86% 2,668,854 6 Summer Off-Peak $0 0.00% 0 7 Winter Part-Peak $1,167 0.22% 53,601 8 Winter Super Off-Peak $0 0.00% 0 9 Winter Off-Peak $0 0.00% 0

10 Total $535,158 100.00% 24,582,563

10 Non-Capacity-Related revenue requirement $1,427,902 _______________

Note: The percentage of capacity-related costs is the average of 27.19 percent and 27.33 percent.

D. Summary of Findings and Conclusions 4

PG&E respectfully requests that the CPUC adopt its MTCC estimate of 5

$12.46 per kW per year. PG&E also respectfully requests that the CPUC permit 6

PG&E to use its MTCC estimate for setting marginal cost-based price floors 7

under tariff E 31 and for use in other proceeding where an MTCC estimate may 8

(PG&E-2)

5-9

be needed. Finally, PG&E has also provided its compliant transmission cost-1

causation study as required by D.18-08-013. 2

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 6

DISTRIBUTION EXPANSION PLANNING PROCESS AND

PROJECT COSTS

(PG&E-2)

6-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 6

DISTRIBUTION EXPANSION PLANNING PROCESS AND PROJECT COSTS

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 6-1

B. Selection of Distribution Study Areas ............................................................... 6-2

C. Load Forecasting .............................................................................................. 6-3

1. Overview .................................................................................................... 6-3

2. Load Forecasting Tool – LoadSEER .......................................................... 6-4

a. Load Forecast ...................................................................................... 6-4

1) Input Data ...................................................................................... 6-5

2) Load Adjustments ......................................................................... 6-6

b. Regression Models .............................................................................. 6-7

1) R-Squared ..................................................................................... 6-8

2) Adjusted R-Squared ...................................................................... 6-8

c. Forecast Selection ............................................................................... 6-8

d. DER Forecast ...................................................................................... 6-9

D. Capability of Facilities ..................................................................................... 6-10

E. Distribution Expansion Plans and Costs ......................................................... 6-11

1. Normal Criteria ......................................................................................... 6-11

2. Emergency Criteria .................................................................................. 6-11

F. Distribution Capacity Planning DER Alternative Analysis ............................... 6-12

G. Emergency Capacity 2020 GRC Phase I Testimony ...................................... 6-13

H. Non-Marginality of Distribution O&M and Replacement Costs ....................... 6-14

I. Results and Conclusion .................................................................................. 6-14

(PG&E-2)

6-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 6 2

DISTRIBUTION EXPANSION PLANNING PROCESS AND PROJECT 3

COSTS 4

A. Introduction 5

This chapter describes the planning process that Pacific Gas and Electric 6

Company (PG&E) uses to develop circuit and bank load forecasts and 7

distribution expansion plans, both of which are used to develop area distribution 8

marginal costs. 9

PG&E’s distribution system is defined as facilities operated at voltages less 10

than 50 kilovolts (kV). This system is further divided into two parts: (1) the 11

primary distribution system; and (2) the secondary distribution system. Any 12

equipment operating at or above 4 kV is considered part of the primary system 13

and is covered under this chapter. This chapter addresses the planning process 14

for the primary distribution system.1 15

The purpose of the planning process is to provide sufficient substation 16

and circuit capacity so that equipment is not overloaded and so that 17

service-operating parameters (e.g., voltage limits and adequate reliability levels) 18

are maintained under both normal and emergency operating conditions. 19

The basic distribution planning approach described in this chapter is 20

generally similar to those described in PG&E’s previous General Rate Cases 21

(GRC) beginning with the 1993 GRC. As with those previous rate cases, the 22

location and time-specific distribution plans resulting from PG&E’s planning 23

process continue to support a marginal cost methodology that is forward-looking, 24

captures the timing and magnitude of planned investments, and reflects cost 25

differences by geographic area (18 divisions).2 26

1 Secondary systems operate at less than 600 volts. In contrast to primary distribution,

most secondary distribution investments are made for the purpose of connecting new customers to the distribution grid or as small projects to correct secondary voltage issues. Since most secondary distribution system investments come about because of customer notifications to PG&E, no formal planning process for secondary distribution investments is needed.

2 For purposes of distribution planning, Humboldt and Sonoma are evaluated as one Division. For evaluation of marginal costs in the subsequent chapters, these operating divisions are shown separately.

(PG&E-2)

6-2

However, while the basic distribution planning approach has remained 1

generally consistent over time, in 2016 PG&E revised the load forecasting 2

process to include a Distributed Energy Resource (DER) growth scenario as a 3

layer in the forecasting process. This change is discussed in Section C, below. 4

The remainder of this chapter is organized as follows: 5

Section B – Selection of Distribution Study Areas 6

Section C – Load Forecasting 7

Section D – Capability of Facilities 8

Section E – Distribution Expansion Plans and Costs 9

Section F – Emergency Capacity 2020 GRC Phase I Testimony 10

Section G – Non-Marginality of Distribution Operations and Maintenance 11

(O&M) and Replacement Costs 12

Section H – Distribution Capacity Planning – DER Alternative Analysis 13

Section I – Results and Conclusion 14

The basic elements listed above are the same as those presented during 15

Phase II of PG&E’s 2017 GRC. The workpapers supporting this chapter include 16

a copy of Electric Distribution Engineering and Planning Drawing 050864, Guide 17

for Planning Area Distribution Facilities. 18

B. Selection of Distribution Study Areas 19

PG&E divides its distribution system into specific geographic areas or 20

Distribution Planning Areas (DPA). Prior to 2012, PG&E had forecasted 21

distribution expansion based on DPAs and analyzed load and capacity at the 22

DPA level. However, since the adoption of the LoadSEER forecasting software 23

in 2012, forecasts have been performed at a more granular level (the distribution 24

substation transformer bank (Bank) and circuit level) so the DPA no longer had 25

significance in PG&E’s planning or forecasting process. 26

Ideally, a DPA has uniform load distribution, uniform load growth rate, 27

a single primary distribution voltage, strong distribution ties among substations 28

inside the area, and limited ties to substations outside the area. Although 29

perfectly ideal DPAs are not encountered in practice, DPAs are defined as 30

nearly as practicable to that ideal. Currently, there are 243 DPAs in PG&E’s 31

service territory. Although not used for planning or forecasting, DPA designation 32

and boundaries have been retained as a way to identify the location of a group 33

(PG&E-2)

6-3

of substations and is generally used to assign work to PG&E’s Distribution 1

Planning engineering offices. 2

C. Load Forecasting 3

PG&E’s current process forecasts load growth at the bank and circuit level, 4

aggregated to the DPA level. PG&E’s forecasting process and additional 5

information about the forecasting program are described below. 6

1. Overview 7

PG&E’s Distribution Engineers (Distribution Engineers) annually 8

forecast the magnitude and location of load and DER growth projections to 9

ensure that sufficient distribution capacity is available in time to meet 10

demand growth in each area. Forecasting future loads and DER growth and 11

assigning those adjustments to specific facilities allows adequate time to 12

address capacity deficiencies to prevent overloading of facilities. While 13

PG&E’s planning process is designed to forecast and minimize equipment 14

overloads, transformer bank, circuit, or component overloads can occur due 15

to: (1) differences between actual growth and forecasted growth; 16

(2) metering device inaccuracies; or (3) weather conditions that exceed 17

PG&E’s design one-in-ten weather event. 18

There are several steps that must be considered when completing load 19

forecasts. Engineers must review available capacity, historical loading, load 20

transfers, and adjustments to future forecasts3 before forecasting future 21

loads. PG&E uses a commercially-available load forecasting program called 22

LoadSEER for this effort. This program consists of two separate forecasting 23

applications: (1) LoadSEER Forecast Integration Tool (FIT) and 24

(2) LoadSEER Geographic Information System (GIS). Each application 25

uses different methodologies to develop a ten-year forecast. 26

Load impacts from existing interconnected small Distribution Generation 27

(DG), Demand Response (DR), solar photovoltaic (PV), and Energy 28

Efficiency (EE) measures are embedded in the annual historic observed 29

peak loads. This historic data, which includes the impacts of existing DERs, 30

is used to determine the level of temperature-normalized historic peak 31

3 Adjustments to future forecasts are based on known-loads and generation or firm

capacity agreements.

(PG&E-2)

6-4

demand in the LoadSEER geospatial forecasts. Incorporating the growth of 1

DERs into the underlying load growth projections ensures they are 2

considered when assessing distribution grid needs and scoping capacity 3

projects. 4

Beginning with the 2017 forecasting cycle, which used the 2016 5

recorded peak update, PG&E disaggregated California Energy Commission 6

(CEC) system-level DER growth forecasts to circuits. The inclusion of these 7

forecasts serves to better model the interaction of customer load growth and 8

DER growth on the distribution system in an era of increasing distributed 9

electric resources. 10

2. Load Forecasting Tool – LoadSEER 11

a. Load Forecast 12

The LoadSEER Program is designed to allow PG&E’s distribution 13

planning engineers (Planning Engineers) to forecast ten years of future 14

non-simultaneous load at the circuit, bank and DPA level, by using 15

two different forecasting methodologies—LoadSEER FIT and 16

LoadSEER GIS. 17

The LoadSEER FIT methodology uses a traditional regression 18

forecast based on historical load peaks for the past 12 years and 19

normalizes the data for both weather and economy. The historical peak 20

loads are also weather-normalized during the regression analysis and 21

the final forecast shows the load normalized to a one-in-two year 22

(50th percentile) and one-in-ten year (90th percentile)4 weather event. 23

Each bank and associated circuit are assigned to a weather station to 24

be used for the weather normalization. 25

The LoadSEER GIS methodology involves a spatial forecasting 26

program that uses proprietary algorithms and satellite imagery to score 27

each acre of PG&E’s service territory for the likelihood of increased 28

load. LoadSEER-GIS replicates urban development processes based 29

on historical land use change, zoning information from government, 30

customer rate class from utilities, and the energy model of consumption 31

patterns. The LoadSEER GIS model also includes 20 years of historical 32

4 Guide for Planning Area Distribution Facilities 050864, p. 4.

(PG&E-2)

6-5

aerial imagery of land use to determine the historical type of expansion 1

that has occurred in an area and facilitate the scoring of each acre. The 2

algorithm used by LoadSEER GIS evaluates and scores each acre 3

based on the likelihood of increased load by customer class 4

(i.e., domestic, commercial, industrial, or agricultural). The program 5

then allocates the CEC’s annual simultaneous distribution system peak 6

load growth projections for each customer class to each parcel and 7

circuit by identifying which circuit is in the closest proximity to the acre. 8

The LoadSEER FIT Program can take the simultaneous peak 9

forecast allocations produced by customer class using the LoadSEER 10

GIS methodology and convert it to a non-simultaneous customer class 11

peak value by utilizing circuit hourly load profiles in LoadSEER FIT. 12

This process has been completed for all circuits within PG&E’s service 13

territory for each of the next ten years. 14

1) Input Data 15

In preparation for the annual forecasting process, PG&E uses 16

three years of customer Advance Metering Infrastructure meter data 17

to develop and update the circuit hourly base load profiles. PG&E 18

then gathers customer class counts, using the current year’s July 19

data point for each circuit and populates them into the program. 20

In addition, the previous 12 months of hourly temperature data and 21

the current year’s summer peak load for each circuit and bank is 22

imported. This data includes the actual peak time and date for 23

meters that have Supervisory Control and Data Acquisition 24

(SCADA) information. All specific data points are applied to the 25

peak load information for each circuit and bank. 26

Other adjustments are entered into the appropriate sections of 27

the LoadSEER Program, including all completed and planned 28

transfers, any new customer load adjustments, and any planned 29

projects that will change the normal- or emergency-capacity of the 30

existing equipment. 31

(PG&E-2)

6-6

2) Load Adjustments 1

Load adjustments are added to the annual bank and circuit 2

peaks to account for the output of the single largest DG connected 3

to the circuit or bank during the time of the circuit or bank’s peak 4

hour. This adjustment is done to ensure that the distribution 5

facilities are adequately sized to serve all customer load, should the 6

largest DG system be unavailable during the time of peak loading. 7

The loss of the largest single DG system is considered to be the 8

only N-1 scenario on the distribution system. The exception to this 9

scenario is if there are multiple hydro-generation units that use the 10

same common water source to generate; the failure of this source 11

could result in all generators being offline. In this situation, the total 12

output of all hydro-generation units connected to the single water 13

source should be used to determine the amount of load adjustment 14

added to the annual peak demand. 15

For photovoltaic (PV) DG systems, only the largest location with 16

a single interconnection point capable of producing an output of 17

500 kilowatts or greater should be considered for an adjustment 18

to the historical peak load. When adding a load adjustment for PV 19

systems of this size, the hour of the circuit peaks and bank peaks5 20

must be compared with the PV output at the time of facility peak to 21

determine the approprate adjustment factor. If SCADA data is not 22

available to determine the hour peak, then the calculated hourly load 23

profile in LoadSEER can be used. If billing or metering data is not 24

available for the generation output, then the nameplate rating should 25

be used to calculate maximum system output using a typical hourly 26

PV output chart. Figure 5-1 below shows the average hourly output 27

for a typical PV system for July in PG&E’s service territory. 28

5 Note that circuit peaks and bank peaks may occur at different hours.

(PG&E-2)

6-7

FIGURE 6-1 PV OUTPUT – MONTH OF JULY

PV HOURLY OUTPUT CURVE

b. Regression Models 1

LoadSEER FIT uses a standard multiple variable regression 2

methodology comparing historical peak load data to historical economic 3

factors and temperature. The coefficient of multiple determinations, 4

commonly known as R-squared, is a statistic value that identifies how 5

well the regression model explains the observed historical data 6

relationship. Generally speaking, an R-squared value of 0 percent 7

suggests that the independent variable(s) poorly support the regression 8

forecast, whereas an R-squared value of 100 percent suggests that all 9

of the variability in the historical data is explained by using the chosen 10

independent variables (e.g., temperature, economic variable). 11

LoadSEER produces an adjusted R-square statistical value to determine 12

the accuracy of multiple variable regression analysis. 13

Month1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

July 0% 0% 0% 0% 0% 10% 27% 47% 67% 82% 93% 98% 98% 92% 79% 62% 40% 17% 3% 0% 0% 0% 0%

Percentage Output by Hour

(PG&E-2)

6-8

1) R-Squared 1

R-squared is equal to SSTO minus SSE, where the error in the 2

response (SSTO) is the total sum, over all observations, of the 3

squared differences of each observation from the overall mean or 4

average of those values, and the error in the regression (SSE) is the 5

sum of squared differences of each observation from the predicted 6

value as expressed by the regression line. This formula is outlined 7

as follows: 8

SSTO = error in the response, or the total sum of squares in the 9

response about the mean 10

SSE = the error in the regression, or sum of squares in the 11

response about the regression line. 12

R-squared = SSTO – SSE 13

Because the SSE is non-negative and cannot exceed the 14

SSTO, R-squared varies between 0 and 1 (i.e., 0 to 100 percent). 15

2) Adjusted R-Squared 16

An adjusted R-squared value “penalizes” (decreases) as one 17

adds more and more predictor variables, demonstrating that too 18

many variables doesn’t necessarily equate to a higher statistical 19

regression model. 20

LoadSEER FIT also produces a “Model Error” score which 21

intends to measure the degree of reliability for a particular model. 22

By removing one of the data points for a given regression model, if 23

there is a significant change in the Adjusted R-squared value, then 24

the accuracy of that model when compared to the historic data is 25

poor, and its “Model Error” score will be high, indicating it is a 26

less-reliable forecast. Ultimately, the goal is to select the regression 27

model with the highest adjusted R-squared and the lowest “Model 28

Error” score. 29

c. Forecast Selection 30

As mentioned earlier, the LoadSEER FIT and LoadSEER GIS 31

provide two types of forecasts. The two forecasts can be helpful to 32

validate the accuracy of each. Basically, if both forecasts are trending 33

(PG&E-2)

6-9

in the same direction, this provides a higher degree of confidence in the 1

accuracy of the forecast. If the two forecasts are trending in opposite 2

directions, this can alert a possible error in forecasting data or significant 3

change in future loading. Conflicting forecasts can reflect various 4

changes in circumstance, such as downturns in economic indicators or 5

an area reaching full buildout following a period of high growth. The 6

LoadSEER FIT Program also has an option that allows a blend of the 7

LoadSEER FIT and LoadSEER GIS forecasts to provide a forecast to 8

evaluate system deficiencies. 9

All bank and circuit load forecasts are calculated and displayed as 10

non-simultaneous peak load values. The LoadSEER Program 11

automatically defaults the forecast to the LoadSEER GIS allocation of 12

the CEC corporate forecast. This ensures that PG&E’s overall 13

distribution system forecast is closely settled to the CEC-level forecast, 14

which has been extensively reviewed and supported by all the 15

Investor-Owned Utilities. 16

Guidelines for blending the LoadSEER FIT regression forecast and 17

LoadSEER GIS corporate forecasts are as follows: 18

1) If the GIS forecast for a given bank or circuit does not include future 19

growth, but the regression forecasts have a reasonable (>.5) 20

adjusted R-squared value and local knowledge of land or load 21

development suggests there should be some amount of future 22

growth, then the two forecasts can be blended, resulting in a small 23

growth rate. 24

2) If the GIS forecast displays a higher growth rate for a given circuit or 25

bank and the regression forecast has a no growth or low growth rate 26

with a reasonable adjusted R-squared value, and local knowledge 27

supports a lower growth rate than the corporate forecast, then the 28

forecasts can be blended, resulting in a lower forecast. 29

3) In all other circumstances the forecast defaults to the CEC corporate 30

forecast. 31

d. DER Forecast 32

LoadSEER also allocates the DER growth scenario forecast using a 33

similar geo-spatial approach as described above for load growth 34

(PG&E-2)

6-10

allocation in Section 2.a. The starting point for each DER scenario is 1

the adopted California Energy Commission’s (CEC) California Energy 2

Demand (CED) forecast. For each of the individual DERs,6 a 3

methodology has been developed to allocate projections consistent with 4

the CED’s PG&E system level projections to each of the approximately 5

3,200 circuits. Demographic variables used for DER allocation include 6

various indicators such as consumption for each customer class, 7

generation by circuit, historical PV adoption by Zip code, s-curve 8

trending model, observed penetration levels, daily peak diversity factors, 9

weather zones, and many other factors specific for each type of DER. 10

An adjustment for each type of DER, including the appropriate 11

hourly profile, is applied at the circuit level. When an adjustment is 12

applied, the DER forecast for PV, DR, EV, and EE are included in the 13

final corporate forecast as an adjustment to the final load growth 14

projection. 15

Currently energy storage (ES) is not disaggregated directly to the 16

circuit level due to the limited amount of ES demographic data available. 17

Instead, the CEC’s Energy Storage discharge forecast at the time of 18

system peak reduces the overall system level load growth forecast. 19

This reduced system-level growth forecast is then disaggregated to 20

circuits based on the GIS load allocation methodology. 21

D. Capability of Facilities 22

As well as creating a load forecast for each distribution transformer bank 23

and circuit in the PG&E distribution system, LoadSEER also includes the normal 24

and emergency capability ratings of these distribution substation assets. The 25

substation asset capabilities are validated as part of the annual load forecasting 26

process. Once the load forecast is complete, Distribution Engineers review the 27

load-carrying capability of the components of the existing distribution system. 28

The most significant type of load growth-related electric distribution capacity 29

expenditures are substation projects (e.g., adding new substations, new 30

6 The individual DERs are residential PV, non-residential PV, energy efficiency, electric

vehicles (EV), and Demand Response (DR).

(PG&E-2)

6-11

substation banks, replacing smaller substation banks with higher capacity units, 1

and adding new circuits, etc.). 2

Distribution substation capability limits are generally defined by the 3

capability of the substation’s transformer banks, the transmission lines supplying 4

the substation, or the distribution lines emanating from the substation. Each 5

substation transformer bank and distribution circuit has a capability rating for 6

both normal and emergency operating conditions. PG&E assigns capability 7

ratings for distribution substation transformers using the equipment 8

manufacturer developed nameplate rating in megavolt-ampere, while assuming 9

a 99 percent power factor to determine the megawatt (MW) rating. 10

E. Distribution Expansion Plans and Costs 11

After the load-carrying capability for a bank or circuit has been determined 12

(as described in Section D) the next task is to determine when forecast annual 13

peak demands (as described in Section C) will exceed the capability of the 14

transformer banks and circuits. 15

PG&E considers both the normal and emergency operating conditions of its 16

distribution system. The following paragraphs briefly summarize the criteria 17

used by distribution planners when preparing and reviewing distribution 18

expansion plans. 19

1. Normal Criteria 20

Area distribution systems must include sufficient substation and circuit 21

capability to supply forecasted peak loads without overloading any PG&E 22

facilities or deviating from normal operating conditions. 23

2. Emergency Criteria 24

Area distribution systems are also planned so that, in the event of the 25

loss of a single distribution facility, the remaining equipment should not be 26

loaded beyond their emergency capabilities and should be returned to the 27

normal capability ratings in 24 hours or less. 28

In this evaluation, PG&E compares the forecasted peak load of a bank 29

or circuit to the amount of emergency capacity at adjacent circuits or 30

substations to determine if sufficient capacity is available to serve the peak 31

load if the bank or circuit is out of service. 32

(PG&E-2)

6-12

Whether considering normal or emergency operating conditions, it is 1

essential to have a working knowledge of facilities to determine what 2

feasible alternatives can be considered to correct projected deficiencies. 3

Before moving ahead with any major change or addition to PG&E’s 4

distribution system, Distribution Engineers perform detailed economic 5

analyses of various feasible alternatives to identify least-cost alternatives 6

that are operationally viable. 7

The distribution expansion plans used for this proceeding were 8

developed as part of PG&E’s annual five-year planning process. For each 9

bank and circuit, Distribution Engineers provided current information on 10

existing capacity, forecasted load growth, and the expected distribution 11

facility installations related to load growth for each year in the five-year 12

planning timeframe. Because construction lead time for some facilities is 13

several years, the five-year planning horizon ensures that facilities are 14

completed on time to meet peak demand. The information developed 15

about expected facility installation and timing provides complete and current 16

expansion costs for major upgrades to PG&E’s distribution system. 17

This information is included in the 2020 GRC Phase I, Exhibit (PG&E-4), 18

Chapter 13 workpapers. This expansion plan and cost information, the load 19

forecast data described in Section C, and division-level accounting data on 20

small projects provide the basis for the calculation of area-specific 21

distribution marginal costs as described in Exhibit (PG&E-2), Chapter 8, 22

“Marginal Distribution Capacity Costs.” The cost information provided is 23

based on the information PG&E included in the 2020 GRC Phase I 24

testimony.7 25

F. Distribution Capacity Planning DER Alternative Analysis 26

PG&E supports the continued and expanded integration of DERs onto 27

PG&E’s distribution grid, while maintaining grid resiliency, safety, reliability, and 28

affordability and honoring customers’ choices for access to clean energy at 29

reasonable rates. 30

7 See Application (A.) 18-12-009, 2020 GRC Phase I, Exhibit (PG&E-4), Ch. 13, “Electric

Distribution Capacity,” and associated workpapers.

(PG&E-2)

6-13

The load growth forecasting process identifies capacity deficiencies and 1

required system expansion projects needed to ensure PG&E’s distribution 2

system is adequate to serve the forecast load. Planning Engineers evaluate 3

potential alternatives for addressing capacity issues, including DER options. As 4

part of its normal analysis of alternatives, PG&E has also used the 5

methodologies outlined in PG&E’s Distribution Resources Plan, Distribution 6

Investment Deferral Planning process to evaluate the cost effectiveness and 7

feasibility of using DERs to reduce peak demand and to allow deferral of the 8

planned capacity investment. 9

Based on the results of the annual load and DER forecast, grid needs 10

identified by the forecast, and the projects identified to meet the grid needs, 11

PG&E issued the Grid Needs Assessment report on June 1, 2018.8 12

Subsequently, PG&E issued the Distribution Deferral Opportunity Report 13

(DDOR) on September 4, 2018. The DDOR includes the list of potential 14

Candidate Deferral Projects that was presented to the Distribution Planning 15

Advisory Group on September 14, 2018. Projects on this list will be considered 16

for Request for Offer (RFO) of DER deferral opportunities, subject to 17

Commission approval. PG&E issued a subsequent DDOR in August 2019. 18

Projects with DER RFOs will remain on the list of PG&E’s planned investments 19

until a contract that results from the RFO has been submitted and approved.9 20

G. Emergency Capacity 2020 GRC Phase I Testimony 21

PG&E believes it is appropriate to have adequate levels of emergency 22

capacity in urban and suburban areas. In Phase I of its 2020 GRC 23

(A.18-12-009), PG&E identified emergency deficiency projects in urban and 24

suburban areas for emergency deficiencies at 10 MW or greater. Using these 25

criteria, PG&E included four transformer emergency deficiency replacement 26

projects in the years between 2020 and 2022.10 27

8 See Decision 18-02-004, Decision on Track 3 Policy Issues, Sub-Track 1 (Growth

Scenarios) and Sub-track 3 (Distribution Investment and Deferral Process), issued on 2/15/18.

9 See Exhibit (PG&E-4), Ch. 13, “Electric Distribution Capacity,” p. 13-9, lines 10-22 and p. 13-10, lines 1-24.

10 See Exhibit (PG&E-4), WP 13-65 to WP 13-66.

(PG&E-2)

6-14

H. Non-Marginality of Distribution O&M and Replacement Costs 1

Generally, the costs of operating, maintaining and replacing distribution 2

equipment, once installed, are independent of usage. Such costs associated 3

with existing distribution equipment are considered fixed costs and excluded 4

from marginal cost calculations. 5

With respect to O&M costs, Commission General Order (GO) 16511 does 6

not permit utilities to reduce the number or frequency of inspections as a 7

function of declining demand. Thus, distribution maintenance costs coming 8

under the purview of GO 165 do not vary as a function of usage. 9

With respect to replacement costs, for example, the life of a distribution pole 10

is generally considered to be approximately 40 years. If a pole is 20 years old, it 11

will require replacement in about 20 years on average. The timing and cost of 12

replacement may be affected by environmental factors specific locations, but are 13

largely unaffected by changes in consumer demand for electricity. Therefore, 14

these costs are properly excluded from marginal cost calculations. The same is 15

true for most other distribution equipment, as long as it is operated within normal 16

operating limits. 17

I. Results and Conclusion 18

This chapter described the planning process for the primary distribution 19

system. The purpose of the planning process is to compare load forecasts to 20

asset capabilities in order to provide sufficient substation and circuit capacity. 21

This ensures that equipment is not overloaded and that service-operating 22

parameters (e.g., voltage limits and adequate reliability levels) are maintained 23

under both normal and emergency operating conditions. 24

11 GO 165 regulates utility inspection and maintenance of certain distribution facilities.

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 7

MARGINAL DISTRIBUTION CAPACITY COSTS

(PG&E-2)

7-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 7

MARGINAL DISTRIBUTION CAPACITY COSTS

TABLE OF CONTENTS

A. Introduction ....................................................................................................... 7-1

B. MDCCs Defined by Voltage Level .................................................................... 7-2

C. MDCCs Defined by Geographic Area (Divisions) ............................................. 7-3

D. Methodologies Implemented in This Filing ....................................................... 7-3

1. Estimating Incremental Capital Addition ..................................................... 7-3

2. Estimating Incremental Peak Demand ....................................................... 7-5

3. Discounted Total Investment Method ......................................................... 7-7

4. NERA Regression Method ......................................................................... 7-9

E. Why Is DTIM a Better Approach Than NERA Regression Method ................. 7-10

F. Background Including Two Additional MDCC Methodologies ......................... 7-11

1. Total Investment Method .......................................................................... 7-13

2. Present Worth Method ............................................................................. 7-14

G. PG&E Recommends DTIM ............................................................................. 7-14

H. RECC and Loading Factors ............................................................................ 7-15

I. Results ........................................................................................................... 7-16

J. Substation MDCC ........................................................................................... 7-17

K. Conclusions .................................................................................................... 7-19

(PG&E-2)

7-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 7 2

MARGINAL DISTRIBUTION CAPACITY COSTS 3

A. Introduction 4

Pacific Gas and Electric Company’s (PG&E) estimated demand-related 5

marginal primary and secondary distribution capacity costs are presented in this 6

chapter. Marginal Distribution Capacity Costs (MDCC) at primary and 7

secondary voltage levels have been estimated using the forecasted peak 8

demand growth and the corresponding investments, i.e., capital additions, based 9

on two methods: (1) Discounted Total Investment Method (DTIM) and 10

(2) Regression Method (RM). The details of these two methodologies and the 11

proposed MDCC estimates are presented in this chapter. PG&E requests the 12

CPUC’s approval of DTIM method, herein, due to the reasons explained in 13

“Section E: Why is DTIM a better approach than [National Economic Research 14

Associates, Inc.] NERA Regression Method,” and “Section G: PG&E 15

recommends DTIM.” 16

In this filing, PG&E has estimated primary and secondary voltage level 17

MDCCs separately for each of its 19 operating divisions. PG&E has also for the 18

first time, separately provided substation MDCC estimates. Additionally, PG&E 19

has implemented an improved method of more accurately estimating 20

incremental peak demand growth in a way that correctly reflects its relationship 21

with corresponding incremental capital investment. This improved method 22

requires that the incremental peak demand be calculated at the distribution 23

circuit (i.e., “feeder”) level, rather than at the division level. Using the circuit-24

level peak demand data to estimate demand growth produces a cost driver 25

which is directly correlated to capital investment because PG&E conducts the 26

distribution investment planning process at the circuit level. Additionally, the 27

increasing adoption by customers of distributed energy resources results in 28

more localized demand growth-related capital investment, and therefore, circuit-29

level peak demand growth is a more accurate cost driver of the localized capital 30

investment. 31

PG&E is requesting that the California Public Utilities Commission (CPUC or 32

Commission) approve PG&E’s marginal distribution capacity cost methodology 33

(PG&E-2)

7-2

and find the resulting MDCC estimates to be reasonable. These MDCC values 1

are used for the following: (1) marginal cost revenue calculations for revenue 2

allocation among customer classes and rate design in this filing; (2) establishing 3

costs where needed for customer-specific contract analysis including analysis of 4

contribution to margin for customers taking service under PG&E’s Economic 5

Development Rate; (3) performing cost-effectiveness analysis of distributed 6

resource evaluation; and (4) use in avoided cost modeling in other Commission 7

proceedings. 8

The remainder of this chapter is organized as follows: 9

Section B – MDCCs Defined by Voltage Level 10

Section C – MDCCs Defined by Geographic Areas (Divisions) 11

Section D – Methodologies Implemented in this Filing 12

Section E – Why is DTIM a Better Approach than NERA Regression Method 13

Section F – Background including Two Additional MDCC Methodologies 14

Section G – PG&E Recommends DTIM 15

Section H – RECC and Loading Factors 16

Section I – Results 17

Section J – Substation MDCC 18

Section K – Conclusions 19

B. MDCCs Defined by Voltage Level 20

PG&E has calculated MDCCs separately for primary and secondary voltage 21

levels. In addition, MDCCs for distribution primary line extensions (referred to as 22

“new business primary”) have been calculated separately. This approach is 23

consistent with PG&E’s 2017 General Rate Case (GRC) Phase II filing. 24

The MDCC estimates for primary voltage level (between 4 kilovolts (kV) and 25

60 kV) are shown as: (1) Primary MDCC, which includes capacity-related 26

distribution plant additions that are generally capacity additions to meet demand 27

growth at existing distribution substations and mainline (primary) distribution 28

feeders; and (2) New Business Primary, MDCC which includes distribution 29

primary line extensions necessary to serve the demand of new customers. 30

Secondary voltage level (less than 4 kV) MDCC estimates are for the 31

capacity growth-related secondary distribution system costs. While secondary 32

distribution costs are comprised of Final Line Transformers (FLT), secondary 33

conductor, service conductor (service drops) and meters, only the secondary 34

(PG&E-2)

7-3

distribution capacity investments made to address demand growth on the 1

existing system are captured in MDCC. Transformer, Service and Meter (TSM) 2

costs for new connections are included in PG&E’s Marginal Customer Access 3

Costs (MCAC) described in Chapter 9 of Exhibit (PG&E-2). 4

C. MDCCs Defined by Geographic Area (Divisions) 5

PG&E continues to differentiate its marginal distribution costs by geographic 6

area. The Commission has indicated that marginal costs should reflect 7

geographic differences where significant.1 Accordingly, PG&E continues to 8

estimate marginal costs by PG&E’s 19 operating divisions to determine the 9

marginal costs presented in this proceeding. 10

D. Methodologies Implemented in This Filing 11

The following four sub-sections present PG&E’s methodologies for 12

calculating Primary, Secondary and New Business Primary MDCC estimates. 13

The first sub-section discusses the details on how the incremental capital 14

addition data are prepared to accurately capture demand growth-related 15

investments. The second sub-section discusses how the incremental demand 16

growth is computed and details PG&E’s improved methodology in this filing. 17

The third sub-section describes the DTIM methodology, and the fourth 18

sub-section describes the NERA Regression Method. As in PG&E’s GRC 2017 19

Phase II filing, PG&E has estimated marginal cost with these two methods in this 20

filing. Two additional known methods, namely Total Investment Method (TIM) 21

and Present Worth Method (PWM) are described in Section F of this chapter, 22

but have not been estimated. 23

1. Estimating Incremental Capital Addition 24

The peak demand growth related capital additions used in estimating 25

the MDCCs are defined in the following three Major Work Categories 26

(MWC):2 27

MWC 06 – Electric Distribution Line Capacity 28

MWC 16 – Electric Distribution Customer Connections 29

1 See Exhibit (PG&E 2), Chapter 1, Section C for discussion. 2 PG&E uses MWCs to group similar types of capital or labor expenses. MWCs are

higher-level aggregations of asset classes and/or Federal Energy Regulatory Commission accounts.

(PG&E-2)

7-4

MWC 46 – Electric Distribution Substation Capacity 1

The types of capital additions defined in these MWCs include both 2

individual projects that are planned for specific substations or feeders, and 3

general “planning orders,” which forecast more generic capital additions at 4

the division or even system level. All capital additions in MWC 06, 16, 5

and 46 are forecast for a 5-year period and are filed in GRC Phase I.3 6

Some capital additions in these MWCs are not strictly related to demand 7

growth and are therefore excluded from the MDCC calculations.4 After 8

excluding non-demand growth-related capital additions, the remaining 9

capital additions are grouped into “projects over $1 million” and “projects 10

under $1 million.” 11

The “projects over $1 million” group consists of MWC 06 and 46. 12

Specific projects that are over $1 million in total can be assigned to a 13

specific circuit and/or Distribution Planning Area (DPA). A few examples 14

include construction of new substation equipment or major upgrades to 15

specific distribution circuits. These projects are exclusively included as part 16

of the Primary MDCC estimate and are calculated at the DPA level5 and 17

then aggregated to division level. 18

Capital additions for projects under $1 million are identified in all 19

three MWCs. These capital additions require some additional processing to 20

be allocated to the Primary, Secondary and New Business Primary MDCC 21

categories. First, forecast capital additions that are not division-specific are 22

weighted and allocated based on each division’s share of forecast system 23

3 Although capital additions are provided in a 5-year forecast format, there are projects

that do not have capital additions for all five years. Particularly, planning orders for smaller generic projects tend of have only two to three years forecast.

4 A few examples of non-demand growth related capital additions include: (1) MWC 06 capital additions related to power factor management or line reinforcement; (2) MWC 16 capital additions for final line TSM assets facilitating new customer connections. (TSM costs related to new connections are calculated in the MCAC model.); and (3) MWC 46 capital additions for restructuring of substation assets due to feeders exceeding 6000 customers or 600 amperes of load.

5 PG&E provides MDCC estimates for primary projects over $1 million at the DPA level because in the Distribution Resource Plan and Integrated Distributed Energy Resources proceedings the Commission and other stakeholders have expressed interest in the development of a DPA specific distribution avoided cost that can be used in the Avoided Cost Calculator (ACC).

(PG&E-2)

7-5

demand growth. Second, capital additions are allocated to primary and 1

secondary voltage using three years of recorded capital additions of projects 2

under $1 million. The allocation of primary and secondary voltage recorded 3

capital additions by division and MWC is applied to the forecast capital 4

additions. 5

2. Estimating Incremental Peak Demand 6

This section explains PG&E’s methodology for calculating the 7

incremental and cumulative peak demand growth for five forecast years, 8

using an illustrative example shown in Table 7-1. For the 2020 GRC 9

Phase II filing, PG&E has calculated incremental demand growth at the 10

distribution circuit level to more accurately capture the load growth which 11

drives demand-related capital additions. Forecast data for Years 1 12

through 5 in this case are for the years 2018 through 2022. The demand 13

forecasts for Year 1 through Year 5 are comprised of (1) ”base load” which 14

is the circuit-level baseline load including any non-specific demand growth; 15

(2) ”adjustments,” which include changes to load due to adoption of 16

DER technology, energy efficiency, demand response and specified 17

demand growth; and (3) ”transfers” which reflect transferring of load 18

between circuits. Details of the load forecasting process are discussed in 19

Chapter 6 of Exhibit (PG&E-2). 20

Additionally, a Year 0 (2017) estimate is also necessary to calculate 21

incremental demand growth for Year 1. The Year 0 estimate of peak 22

demand (referred to as “backcast”) is obtained by subtracting an average 23

annual change in base load forecasts (for Years 1 through 5) from Year 1. 24

The incremental demand growth calculation can thus be performed from the 25

backcast peak demand estimate of year 0 and forecast estimates for 26

Years 1 through 5. 27

In prior GRCs, PG&E has aggregated the circuit-level peak demand 28

forecast to the division level and calculated the year-to-year change in peak 29

demand as “incremental demand growth.” However, at the circuit level, peak 30

demand does not typically exhibit a perfectly positive, linear growth trend. 31

Circuits may exhibit declining peak demand in some forecast years and 32

increasing peak demand in other years. Some circuits may even exhibit 33

declining peak demand across all years. By aggregating annual peak 34

(PG&E-2)

7-6

demands to the division level prior to calculating incremental demand, 1

circuits with declining incremental demand in a given year offset positive 2

incremental demand from other circuits for that year. Given that 3

demand-related capital additions are only invested in circuits with positive 4

incremental demand growth, this calculation dramatically underestimates the 5

demand growth that drives capital additions and ultimately results in 6

artificially high MDCC estimates. 7

While preparing this filing, PG&E noticed a significant change in the 8

distribution circuit-level demand forecast prepared for its 2020 GRC filing, 9

when compared to the demand forecast that was used in the 2017 GRC 10

filing. A large proportion of electric distribution circuits now show changes in 11

forecast peak demand due to changes in the underlying adjustments, 12

described at the beginning of this section, which was not the case for 13

forecasts used in the 2017 GRC filing. By applying the 2017 division-level 14

methodology to the 2020 demand forecast, PG&E calculated extremely low 15

(and in some cases negative) division-level incremental demand growth 16

estimates that were not reasonable. 17

As a result of these extremely low demand growth estimates, PG&E has 18

calculated a “rolling maximum” peak demand at the circuit level. Under this 19

method, the incremental demand growth is based only on the highest peak 20

demands of the entire forecast period. Determining incremental demand in 21

this manner can only be done at the circuit level given that the offsetting 22

effect of varying peak demand trends across different circuits cannot be 23

avoided at any higher level of aggregation.6 PG&E’s rolling maximum peak 24

demand approach reflects the fact that demand-related investments are 25

driven exclusively by absolute maximum demand growth on a circuit.7 This 26

approach more accurately captures the positive demand growth which 27

drives demand-related capital additions and is therefore a better cost driver 28

for use in MDCC estimation. 29

6 After incremental demand has been calculated at the circuit level, they are still

aggregated to the division level in order to produce division-level MDCC estimates. 7 Intermediate years with reduced demand do not offset the capital investments needed

to meet the rolling maximum demand. Likewise, circuits with no positive change in rolling maximum demand require no capital investment.

(PG&E-2)

7-7

Table 7-1 below compares the calculation of incremental load growth 1

between the 2017 GRC Phase II filing methodology and the improved 2

methodology in this filing using an illustrative circuit level peak demand 3

forecast. The table clearly demonstrates how the 2017 GRC methodology 4

results in “negative” incremental peak demand growth which reduces the 5

total cumulative incremental growth for the forecasted years. 6

TABLE 7-1 AN ILLUSTRATIVE EXAMPLE OF

INCREMENTAL AND CUMULATIVE PEAK DEMAND GROWTH CALCULATION IN A DISTRIBUTION CIRCUIT

Line No.

Year 0 (2017)

Year 1 (2018)

Year 2 (2019)

Year 3 (2020)

Year 4 (2021)

Year 5 (2022)

1 Peak Demand (MW) 15 13 17 14 20 18

2 Method Peak Demand Growth

Year 0 Year 1 Year 2 Year 3 Year 4 Year 5

3 2017 GRC Method

Incremental (2) 4 (3) 6 (2)

Cumulative (2) 2 (1) 5 3

4 2020 GRC Method

Incremental – 2 – 3 –

Cumulative – 2 2 5 5

3. Discounted Total Investment Method 7

PG&E has used the DTIM as described by the CPUC in its 1992 gas 8

marginal cost decision: 9

The Discounted Total Investment Method computes a marginal cost by 10 dividing the present value of the planning period’s investment by [the 11 present value of]8 the total load growth. A present value is used in the 12 numerator to give additions further into the future less weight than 13 investments in earlier time periods. The marginal unit cost is then 14 annualized using an RECC factor. (Decision (D.) 92-12-058, p. 33.) 15

8 In the original California description of DTIM, the incremental capacity values in the

denominator were not discounted. The phrase “…the present value of…” is added to correct the description of the DTIM, reflecting later thinking as to the proper calculation of marginal costs using this methodology.

(PG&E-2)

7-8

The DTIM equation can be written as follows: 1

Equation 1: = ∑ (1 + )∑ (1 + ) × ( ) Where: = Incremental capital investment in year = Incremental capacity in year = Annual real discount rate (Based on Weighted Average Cost of Capital) = Number of years in the planning horizon = Real Economic Carrying Charge plus loading factor representing

O&M costs

It is appropriate to discount both the numerator and the denominator in 2

the DTIM equation.9 The discounting of incremental capacity, a 3

non-monetary amount, is proper. The incremental capacity in each year has 4

a value equivalent to a levelized price multiplied by the units of incremental 5

capacity, and the value is greater for having a unit of incremental capacity 6

now versus having a unit of incremental capacity in the future. 7

Mathematically, the discounting of incremental capacity derives from 8

discounting the stream of incremental investments needed to invest in 9

capacity where the discounted incremental investments are equivalent to an 10

application of a levelized cost of capacity to the quantities of incremental 11

capacity. This can be demonstrated by rearranging the equation. 12

Equation 2: (1 + ) × = (1 + ) ×

With this rearrangement of the equation, the term that is being 13

discounted in the summation on the left-hand side of the equation is 14

effectively the annualized price per unit. The sum of the present value of the 15

price in each year multiplied by each year’s quantity of incremental capacity 16

is equal to the present value of the cost for that stream of incremental 17

9 Discounting the denominator prevents underestimating the marginal investment. In

fact, without discounting the denominator, the marginal investment would tend to zero as the planning horizon increased. This is because the numerator would converge to a finite number, while the denominator would grow without a bound.

(PG&E-2)

7-9

capacity investment for the left-hand side of the equation. This present 1

value of investing in incremental capacity at a levelized price in the left-hand 2

side of the equation is equal to the sum of the present value of each year’s 3

incremental investments made for additional capacity in the right-hand side 4

of the equation with application of the Real Economic Carrying Charge 5

(RECC) factor. 6

4. NERA Regression Method 7

In addition to estimating MDCC based on DTIM, PG&E has estimated 8

MDCC using the CPUC adopted RM which was developed by NERA 9

Economic Consulting.10 PG&E performed regression using ten years of 10

historical data along with five years of forecasts, as is used in a typical 11

NERA regression. 12

The NERA model is described in the gas marginal cost decision 13

(D.92-12-058) as follows: 14

The NERA Regression methodology uses a model developed by NERA 15 to obtain a marginal unit capital cost by regressing the cumulative 16 changes in investment with cumulative changes in load. Parties used a 17 combination of historical and forecast period data. The marginal unit 18 cost is then annualized using the RECC factor…. (D.92-12-058, p. 32.) 19

The calculation of marginal cost using the RM consists of estimating the 20

coefficient “m” on the following linear regression equation using an ordinary 21

least squares simple regression approach (Equation 3) is described below. 22

10 The NERA RM MDCC estimates are provided in response to California Public

Advocates’ Master Data Request.

(PG&E-2)

7-10

Equation 3: = × + Where: = Cumulative increase in capital investment through year = Cumulative increase in demand through year = The slope of the line estimated by the linear regression = The zero intercept for the line estimated by the linear regression And: The slope provides an estimate of the marginal impact of for a change in , i.e., the marginal investment per unit of capacity Such that: = × Where: = The Real Economic Carrying Charge plus O&M loading factor

The yearly cumulative investment values “It” are regressed on the yearly 1

cumulative incremental capacity additions “Ct.” The regression’s resulting 2

estimate of the slope coefficient “m” is an estimate of the marginal 3

investment required for a unit of additional capacity. The RECC factor is 4

applied to this marginal investment to yield the annualized marginal cost for 5

investing in an additional unit of capacity. The O&M loading factor is also 6

included to capture the operation and maintenance-related costs associated 7

with the capital investments. 8

E. Why Is DTIM a Better Approach Than NERA Regression Method 9

The NERA regression method relies on a mix of historical data and forecast 10

data, which in PG&E’s view is not a sound approach of estimating MDCC. 11

PG&E has a serious concern regarding the input data used in the NERA RM. 12

While both historical and forecast input are used together, the characteristics of 13

the input forecast data is very different from the input historical data. The 14

forecast capital addition and peak demand data is based on an extreme weather 15

condition described by a one in ten-year probability of occurrence, while the 16

historical data does not represent such an extreme condition. Rather, in the 17

historical data the peak demand growth is based on recorded data for a period 18

of ten years during which such extreme weather conditions do not necessarily 19

occur. Therefore, the historical peak demands, that do not explicitly represent 20

extreme conditions, are not an appropriate driver of historical capital additions 21

which were made on the assumption of extreme conditions. 22

(PG&E-2)

7-11

Additionally, the need to calculate incremental demand growth at the circuit 1

level, as discussed earlier, is another reason the NERA regression method no 2

longer works. Although forecasts of peak demand load at the circuit level can be 3

developed, recorded peak demand is not readily available at the circuit level.11 4

Therefore, the historical incremental peak demand growth may reflect the 5

offsetting effect of “non-maximum” peak demand growth, resulting in artificially 6

high MDCCs, as discussed earlier in this chapter. However, even if historical 7

circuit level data were available for calculating historical incremental demand 8

growth, PG&E would still not recommend the NERA regression method given 9

the fundamental differences between the forecast and historical data that was 10

previously described. 11

Lastly, because the NERA regression uses cumulative incremental demand 12

growth and capital additions, the model is most heavily weighted towards the 13

earliest years of the data. PG&E believes that marginal costs should be 14

forward-looking and therefore believes it is appropriate to use forecast data and 15

not historical data, as is done in the DTIM method. There are additional 16

considerations that make DTIM the best approach of all the alternatives explored 17

by PG&E. A summary of pros and cons of four different methods along with 18

PG&E’s recommendation is presented in Section G, “PG&E Recommends 19

DTIM.” 20

F. Background Including Two Additional MDCC Methodologies 21

This section presents a summary of the methodologies that were historically 22

used by PG&E for estimating MDCC values prior to its introduction of DTIM in its 23

1999 GRC, and discusses two methodologies, namely TIM and PWM. 24

Prior to the 1993 GRC, PG&E estimated Transmission and Distribution 25

(T&D) costs by conducting a regression analysis—relying on ten years of 26

historical and five years of forecast accounting, budget and capacity loading 27

data—resulting in a revenue requirement for the annual average investment. 28

11 PG&E’s SCADA system, which measures recorded circuit-level data, is designed only

to produce data for one circuit at a time and does not compile data for all 3,200 circuits in a readily available manner. Furthermore, SCADA data is only available for more recent years and would not provide the full ten-year horizon needed for NERA regression. Likewise, the customer-level interval meter data used to produce circuit level load profiles for distribution PCAFs described in Chapter 8 of Exhibit (PG&E-2) is not available for the ten-year historical period.

(PG&E-2)

7-12

However, heavy reliance on historical accounting data, representing investments 1

that have already occurred, has limited value if one wants a forward-looking cost 2

estimate, primarily due to the increasing technology adoption over time. By the 3

early 1990s, PG&E found that using the RM with historical data had serious 4

shortcomings, particularly given the large, infrequent nature of investments in 5

each of its T&D planning areas. The NERA RM attempts to smooth the 6

inherently lumpy capital additions and cannot meaningfully reflect the timing of 7

future investments in marginal cost estimates. 8

In contrast, the DTIM and PWM, first adopted by the CPUC in PG&E’s 1993 9

GRC Phase II, do not use historical data. These methods are forward-looking, 10

and use capital addition and demand growth forecasts, rather than historical 11

data. These methods also capture the timing and lumpiness inherent in planned 12

T&D investments. DTIM and the PWM better embody two key principles the 13

CPUC endorsed in PG&E’s 1993 GRC Phase II—that marginal costs should be 14

forward-looking and causally linked to a change in demand, and that they should 15

reflect the timing and magnitude of future investments. 16

In D.92-12-057 (PG&E’s 1993 GRC Phase II proceeding), the Commission 17

adopted timing-sensitive and location-specific electric T&D marginal costs based 18

on the PWM,12 as proposed by PG&E. This was done partially in response to 19

the desire for more accurate marginal costs for use in resource planning. In that 20

decision, the Commission found: 21

It is reasonable that marginal cost components be based on the design and 22 operation of PG&E’s system, accurately signal the cost of providing 23 electrical service, be forward-looking, capture the timing and magnitude of 24 future investments, reflect geographic differences where significant,…and 25 finally, provide consistent signals in the evaluation of supply and demand 26 resources for planning purposes. (D.92-12-057, Finding of Fact 152, 27 mimeo, p. 291). 28

In the same month, the CPUC adopted the RM for gas distribution, but the 29

TIM, described below, for gas transmission and storage in D.92-12-058. 30

12 The PWM calculates the present value of planned investments over a planning horizon

and calculates a second present value for those same investments, but assuming that the investments all will be deferred one year. The difference in these two present values is divided by the average expected capacity growth for the planning horizon resulting in a marginal investment. The marginal investment is then annualized using an RECC factor (or similar Capital Recovery Factor (CRF)) and appropriate marginal cost loadings.

(PG&E-2)

7-13

The Commission rejected the RM13 for gas transmission because of the 1

increased lumpiness of transmission relative to distribution.14 In D.95-12-053 2

(PG&E’s 1995 Biennial Cost Allocation Proceeding (BCAP) decision), 3

the Commission stated: 4

[W]e do see merit in exploring the idea of incorporating the time value of 5 money in the calculation of capital-related marginal costs. We direct PG&E 6 in its next BCAP filing to provide a scenario that incorporates a discounted 7 total investment method to estimate the marginal cost of capital investments. 8 (D.95-12-053, p. 37). 9

PG&E believes that either the DTIM or the PWM are superior to the RM or 10

the TIM. Starting in its 1999 GRC, PG&E introduced the DTIM as its proposed 11

methodology for MDCC calculations in place of the PWM.15 The sub-sections 12

below briefly explain TIM and PWM. Note that DTIM and RM have been 13

described earlier in this chapter. 14

1. Total Investment Method 15

The fundamental premise underlying the TIM is that capacity can be 16

added in very small increments. The first step in the TIM is to divide the 17

cumulative demand growth-related investment for the selected time horizon 18

by the cumulative capacity growth for that same time horizon. 19

The second step is to annualize the marginal investment by multiplying by a 20

13 In D.92-12-058, the CPUC also rejected a replacement cost new-based methodology

for gas transmission marginal cost. 14 The lumpiness of distribution investments depends upon the geographic scale. For a

large utility such as PG&E, distribution investments can be relatively smooth on a systemwide total basis. Regression can work well in such cases. However, the RM approach can mask cost differences that are quite important on a local level, especially for dealing with un-economic bypass or for integrated resource planning that considers: both supply- and demand-side alternatives; cost-effectiveness analysis of Customer Energy Efficiency and Demand Side Management programs and distributed resource evaluation.

15 In PG&E’s 1996 GRC Phase II (D.97-03-017), the CPUC returned to the RM from PG&E’s PWM adopted in its 1993 GRC Phase II (D.92-12-057), citing volatility concerns when using the PWM and concerns with PG&E’s geographically-differentiated marginal T&D costs. PG&E disputes the notion that the RM produces marginal costs that are less volatile than methods that better address the lumpiness and investment timing issue—like the DTIM—that include the time value of money. The Commission did leave the door open to further reconsideration once PG&E had eliminated concerns about ad hoc distribution planning by area. Although PG&E fully addressed those distribution planning concerns in the 1999 GRC, that proceeding was suspended due to the California energy crisis, and the 2003, 2007, 2011, and 2014 GRC Phase II proceedings were all settled.

(PG&E-2)

7-14

RECC factor and applying marginal cost loadings. In the application of the 1

TIM in the past, the time horizon was 15 years, which included ten years of 2

historical data and five years of forecast data. The TIM, as a marginal cost 3

method, is not sensitive to the timing of the forward-looking forecast capacity 4

additions since it includes no discounting for the time value of money. 5

In fact, if all investments in the time horizon are moved to any single year, 6

the result is the same as if the investments were spread across the time 7

horizon. Additionally, the predominance of historical data causes marginal 8

costs estimates to be less forward looking than the estimates produced 9

using DTIM. 10

2. Present Worth Method 11

The PWM is calculated in several steps, as follows: (1) calculate a 12

present value of a stream of investments and subtract from that the present 13

value of that same stream of investments, but deferred by one year; 14

(2) divide the difference in present values by the corresponding capacity 15

growth; (3) apply appropriate marginal cost loaders to account for the 16

operational and maintenance costs; and (4) account for full capital recovery, 17

apply a Capital Recovery Factor—of which the RECC is one. 18

It is important to note that the PWM includes discounting to take into 19

account the time value of money. It can be characterized as producing 20

estimates of marginal costs as the deferral values for streams of 21

investments. For the PWM, the important calculation assumption for 22

producing a deferral value is that the entire stream of demand 23

growth-related investments would be delayed one year if expected growth 24

were not to materialize. Thus, the PWM produces marginal T&D costs that 25

vary over time, and thus can reflect the magnitude and timing of planned 26

investments. 27

G. PG&E Recommends DTIM 28

PG&E has discussed earlier in this chapter the advantages of DTIM and 29

how it provides forward-looking marginal cost estimates. Below is a summary of 30

reasons for DTIM to be the best approach to adopt for PG&E: 31

1) DTIM implemented by PG&E uses 5-year forecast data that best reflects the 32

anticipated level of load growth and subsequent capital additions due to all 33

(PG&E-2)

7-15

underlying factors impacting the load profiles, such as adoption of 1

Distributed Energy Resources, Energy Efficiency, Demand Response, 2

Storage, and Electric Vehicle Charging; 3

2) NERA RM inappropriately uses historical data which does not align with the 4

fundamental assumptions of forecast distribution planning data and cannot 5

be modified to reflect PG&E’s improved circuit-level incremental demand 6

calculations; 7

3) NERA RM’s reliance on cumulative incremental capital additional and peak 8

demand growth means that the results are most heavily weighted to 9

historical data and do not appropriately reflect forecast data; 10

4) NERA RM assumes a linear relationship between capital addition and load 11

growth, which is not always true due to the “lumpy” nature of capital 12

additions, and therefore, does not provide theoretically sound MDCC 13

estimates; 14

5) TIM is not recommended because it does not account for lumpiness of 15

capacity growth-related investments; 16

6) PWM is not recommended because it only indirectly uses the resource 17

planning data to calculate deferral values instead of direct marginal cost 18

values, whereas PG&E’s DTIM directly uses PG&E’s local level resource 19

planning data to calculate marginal costs per unit of capacity per year; and 20

7) DTIM both recognizes the time value with regard to units of capacity and 21

accounts for the “lumpy” nature of capacity-related investments. PWM and 22

TIM are essentially primitive versions of this approach. 23

H. RECC and Loading Factors 24

The calculated MDCCs using the DTIM are annualized using the RECC 25

factor. Other expenses (referenced to as “loaders”) that are related to the 26

addition of capacity to the distribution system are added to the annualized 27

marginal investments. These loaders include Operations and Maintenance 28

(O&M) expenses, Administrative and General expenses for the added O&M, 29

common and general plant expenses, cash working capital expenses, and 30

franchise fees and uncollectibles.16 31

16 See Exhibit (PG&E 2), Chapter 10 for description of the development of these marginal

cost loaders and the RECC financial factor.

(PG&E-2)

7-16

I. Results 1

PG&E’s proposed DTIM-based MDCC estimates for each division, and for 2

the distribution system overall, are shown in Table 7-2. For revenue allocation 3

and rate design, PG&E allocates Primary MDCC using Peak Capacity Allocation 4

Factors (PCAF).17 Secondary and New Business Primary MDCC are allocated 5

using FLT load.18 Therefore, Primary MDCC is shown in dollars per 6

PCAF-kilowatt (kW) per year, while Secondary and New Business Primary are 7

shown in dollars per FLT-kW per year. 8

TABLE 7-2 PROPOSED DTIM BASED MDCC ESTIMATES BY DIVISION AND SYSTEM

(2021 DOLLARS)

Line No. Division

Primary Distribution $/PCAF kW

New Business Primary Distribution

$/FLT kW

Secondary Distribution $/FLT kW

1 Central Coast $42.51 $22.81 $1.66 2 De Anza $44.84 $33.38 $2.20 3 Diablo $69.63 $40.88 $2.68 4 East Bay $40.86 $22.91 $1.84 5 Fresno $48.47 $27.51 $2.01 6 Humboldt $43.54 $31.13 $1.40 7 Kern $51.00 $29.28 $2.20 8 Los Padres $63.38 $33.09 $1.82 9 Mission $43.83 $44.30 $2.23 10 North Bay $52.95 $31.40 $2.06 11 North Valley $63.36 $30.18 $1.73 12 Peninsula $48.18 $29.13 $2.02 13 Sacramento $46.65 $34.94 $2.23 14 San Francisco $48.93 $42.93 $2.36 15 San Jose $46.29 $33.17 $2.45 16 Sierra $43.89 $33.26 $1.88 17 Sonoma $43.15 $52.24 $1.84 18 Stockton $44.06 $27.41 $1.99 19 Yosemite $67.14 $25.65 $1.85 20 System $47.96 $28.10 $1.99

PG&E finds the estimates across 19 divisions to be within a reasonable 9

range. Specifically, the variability across divisions has been significantly 10

reduced compared to the 2017 GRC Phase II estimates, as shown in Table 7-3, 11

which compares the estimates presented in Table 7-2 with the MDCC estimates 12

proposed in 2017 GRC Phase II filing. For example, the 2017 GRC Primary 13

17 See Exhibit (PG&E 2), Chapter 8 for definition of PCAF. 18 See Exhibit (PG&E 2), Chapter 8 for discussion of FLT loads.

(PG&E-2)

7-17

MDCC estimates show a minimum value of $13.63 and maximum value of 1

$121.98, compared to a minimum value of $40.86 and maximum value of $69.63 2

in the 2020 GRC estimates. The 2020 MDCC are in general higher when 3

compared to the 2017 estimates, though not in all cases. 4

TABLE 7-3 COMPARISON WITH 2017 GRC PHASE II ESTIMATES

Line No. Division

Primary Distribution $/PCAF kW

New Business on Primary Distribution

$/FLT kW Secondary Distribution

$/FLT kW

2020 GRC

2017 GRC

2020 GRC

2017 GRC

2020 GRC

2017 GRC

1 Central Coast $42.51 $69.09 $22.81 $14.53 $1.66 $1.04

2 De Anza $44.84 $35.65 $33.38 $19.66 $2.20 $1.01

3 Diablo $69.63 $17.78 $40.88 $23.20 $2.68 $1.56

4 East Bay $40.86 $19.99 $22.91 $18.07 $1.84 $0.88

5 Fresno $48.47 $39.52 $27.51 $15.81 $2.01 $1.36

6 Humboldt $43.54 $73.97 $31.13 $14.20 $1.40 $1.12

7 Kern $51.00 $34.07 $29.28 $16.08 $2.20 $1.23

8 Los Padres $63.38 $56.49 $33.09 $14.41 $1.82 $1.06

9 Mission $43.83 $13.63 $44.30 $16.37 $2.23 $0.97

10 North Bay $52.95 $29.42 $31.40 $14.62 $2.06 $1.75

11 North Valley $63.36 $53.40 $30.18 $19.23 $1.73 $1.26

12 Peninsula $48.18 $31.79 $29.13 $14.02 $2.02 $1.06

13 Sacramento $46.65 $40.91 $34.94 $16.49 $2.23 $1.22

14 San Francisco $48.93 $40.41 $42.93 $19.69 $2.36 $1.52

15 San Jose $46.29 $40.12 $33.17 $17.45 $2.45 $1.16

16 Sierra $43.89 $30.65 $33.26 $20.07 $1.88 $1.25

17 Sonoma $43.15 $121.98 $52.24 $16.65 $1.84 $1.28

18 Stockton $44.06 $33.36 $27.41 $15.13 $1.99 $1.34

19 Yosemite $67.14 $60.18 $25.65 $15.63 $1.85 $1.56

20 System $47.96 $39.43 $28.10 $16.42 $1.99 $1.25

J. Substation MDCC 5

PG&E has used DTIM to estimate Substation MDCC in a way that reflects 6

Substation MDCC as a portion of the total Primary MDCC. In the Primary 7

MDCC calculation, MWC 46 (Electric Distribution Substation Capacity) and 8

MWC 06 (Electric Distribution Line Capacity) capital additions are combined. To 9

(PG&E-2)

7-18

calculate the Substation Primary MDCC, PG&E has used the MWC 46 capital 1

addition only in the numerator, keeping the denominator the same as the 2

denominator used in the calculation of the Primary MDCC.19 Table 7-4 shows 3

the Substation Primary MDCC along with the Circuit Primary and Total Primary 4

MDCC for the purpose of comparison. 5

TABLE 7-4 2020 GRC PHASE II ESTIMATES OF PRIMARY AND SUBSTATION MDCCS

(2021 DOLLARS)

Line No. Division

Primary Distribution $/PCAF kW

Total Primary MDCC

Circuit Primary MDCC

Substation Primary MDCC

1 Central Coast $42.51 $29.49 $13.01

2 De Anza $44.84 $29.49 $15.35

3 Diablo $69.63 $35.52 $34.11

4 East Bay $40.86 $28.35 $12.51

5 Fresno $48.47 $33.40 $15.07

6 Humboldt $43.54 $30.21 $13.33

7 Kern $51.00 $35.21 $15.79

8 Los Padres $63.38 $49.99 $13.39

9 Mission $43.83 $30.41 $13.42

10 North Bay $52.95 $31.79 $21.16

11 North Valley $63.36 $43.55 $19.81

12 Peninsula $48.18 $33.46 $14.72

13 Sacramento $46.65 $33.06 $13.58

14 San Francisco $48.93 $35.40 $13.54

15 San Jose $46.29 $31.25 $15.04

16 Sierra $43.89 $30.22 $13.66

17 Sonoma $43.15 $29.84 $13.30

18 Stockton $44.06 $30.57 $13.49

19 Yosemite $67.14 $48.95 $18.19

20 System $47.96 $33.38 $14.58

19 For marginal cost revenue (MCR) calculation, the substation MDCC estimates are

converted to substation-specific $/PCAF kW unit of measure in order to be applied to a substation-specific PCAF calculations. Details of PG&E’s PCAF calculation are discussed in Chapter 8 of Exhibit (PG&E-2).

(PG&E-2)

7-19

K. Conclusions 1

PG&E performed a detailed study of the input data and implemented DTIM 2

and RM for estimating MDCCs, as it did in 2017 GRC Phase II filing. In this 3

process, however, PG&E made a few important improvements in the way it 4

processes input load growth and capital addition data, compared to the 2017 5

GRC Phase II methodology. These improvements became necessary, primarily 6

due to changes in the demand growth forecast resulting from significant changes 7

in underlying load forecast variables such as Distributed Energy Resources, 8

Energy Efficiency, Demand Response, Storage, and Electric Vehicles. 9

PG&E finds DTIM to be the best choice for reasons described in Section G, 10

“PG&E Recommends DTIM” and requests that the Commission approve the 11

proposed DTIM-based MDCC estimates, including the improved methodology. 12

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 8

DISTRIBUTION CAPACITY COST OF SERVICE

(PG&E-2)

8-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 8

DISTRIBUTION CAPACITY COST OF SERVICE

TABLE OF CONTENTS

A. Introduction ............................................................................................................ 8-1

B. Overview of Methodology ..................................................................................... 8-2

C. Distribution PCAF Methodology ........................................................................... 8-5

D. Distribution FLT Methodology ............................................................................. 8-11

E. Comparison of Distribution Marginal Cost Revenue Per kWh Between NEM and Non-NEM Customers at Distribution System Level .......................... 8-13

F. Comparison of DMCR Per kWh of NEM and Non-NEM Customers Within the Customer Classes......................................................................................... 8-16

G. Primary DMCR Per kWh Comparison During the TOU Periods....................... 8-21

H. Conclusion ........................................................................................................... 8-26

(PG&E-2)

8-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 8 2

DISTRIBUTION CAPACITY COST OF SERVICE 3

A. Introduction 4

In this chapter, Pacific Gas and Electric Company (PG&E) describes its 5

distribution cost of service methodology and results, presented in terms of 6

Marginal Cost Revenue (MCR) per kilowatt-hour (kWh). PG&E’s improved 7

distribution cost of service analysis estimates cost of service for Net Energy 8

Metering (NEM) and Non-NEM customer sub-groups, which leads to more 9

accurate cost of service results overall. PG&E requests that the California 10

Public Utilities Commission (Commission) approve of the methodology and 11

results presented herein. 12

Implementing an improved distribution cost of service framework is critical 13

because of the way PG&E’s customers’ load profiles have been evolving with 14

the adoption of technologies such as solar photovoltaic (PV) and storage that 15

result in flow of electricity not only from utility grid to the customer service point 16

(called “delivered” flow), but also from the customer service point back to the 17

grid (called “received” flow). The following points summarize the impacts in 18

this regard: 19

1) Technology adoptions have led to a change in load shapes including a shift 20

of the peak hours; 21

2) NEM customer load profiles are significantly different than Non-NEM 22

customers due to solar generation from NEM customers during the day; 23

3) Energy “received” from NEM customers can cause peak or near-peak 24

demand which needs consideration for allocation of capacity costs; 25

4) Generation from solar PV causes the net load to decrease significantly 26

during the day, but the hourly load during the system peak hours occurring 27

in the evening remains the same and is not impacted due to the generation 28

from solar PV; and 29

5) For the NEM customers, the average MCR per kWh increases due to 30

relatively higher load during the costlier hours, while load decreases during 31

the significantly less expensive hours of a day due to the adoption of 32

solar PV. 33

(PG&E-2)

8-2

The distribution cost of service methodology including the Peak Capacity 1

Allocation Factor (PCAF) and Final Line Transformer (FLT) methodologies are 2

discussed in detail in the next sections 3

B. Overview of Methodology 4

The metric proposed by PG&E to analyze cost of service and subsequently 5

perform revenue allocation is MCR per kWh. The appropriateness of using 6

MCR per kWh to analyze cost of service has been discussed in Chapter 1, 7

Section A of this exhibit. In order to perform distribution cost of service analysis, 8

distribution capacity MCR is first obtained by multiplying the Marginal 9

Distribution Capacity Cost (MDCC) estimates by the respective cost drivers: 10

(1) PCAF weighted kilowatt (kW) for the primary distribution capacity,1 and 11

(2) FLT demand-related kW, or in short, FLT-kW for secondary and new 12

business primary assets.2 A summary of the distribution cost components and 13

respective cost drivers is shown in Figure 8-1 below. 14

1 In previous GRC filings, primary distribution has included both substation and primary

distribution circuit costs. As a compliance item, PG&E has separately shown substation costs in this filing as described in Chapter 7 of this exhibit. For distribution cost of service analysis, substation MCR and primary circuit MCR have also been calculated separately.

2 PCAF and FLT concepts are presented in detail in Sections C and D in this chapter. In prior General Rate Cases (GRC), PG&E included separate chapters on cost drivers for distribution, namely separate chapters on PCAF and FLT demand. In this application, development of those cost drivers is covered in this chapter.

(PG&E-2)

8-3

FIGURE 8-1 SUMMARY OF DISTRIBUTION CAPACITY COST COMPONENTS

The distribution capacity MCR is then divided by usage (kWh) at the 1

customer class level, separately for NEM and Non-NEM customer sub-groups. 2

The distribution MCR per kWh for primary circuits are also estimated by 3

Time-of-Use (TOU) period since they represent circuit-level coincident peak 4

demand.3 These average MCR per kWh estimates segmented by customer 5

class, NEM versus Non-NEM, and by TOU period provide cost of service-related 6

comparisons and insights. The MCR per kWh estimates are also used in 7

revenue allocation and rate design. 8

In prior GRCs, PG&E had used the PCAF kW and FLT kW demand cost 9

drivers for calculating distribution capacity MCR. PG&E continues to use PCAF 10

and FLT demand metrics as the cost drivers for distribution capacity MCR. 11

However, PG&E has made a few changes to the calculations of PCAF and FLT 12

demand which are described in the following sections. Most importantly, PG&E 13

has revised its distribution capacity cost driver calculations to better reflect the 14

ways customers utilize the distribution grid today. With the increase on the 15

number of technology adoptions customers may adopt, customers are imposing 16

3 Secondary and New Business Primary equipment are designed based on

non-coincident demand at the FLTs, and therefore, the associated MCR per kWh are not suitable for TOU based differentiation.

(PG&E-2)

8-4

costs and providing benefits to the system in ways that were not properly 1

reflected in PG&E’s cost of service evaluation methodologies used in prior GRC 2

Phase II filings. In this chapter, PG&E describes its proposed, improved 3

methodology to determine those costs and benefits at a more granular level, 4

enabling separate quantification of the costs associated with NEM power supply 5

to the grid (i.e., “received” flow) from the cost drivers associated with use of the 6

grid to deliver energy to those same customers (i.e., “delivered” flow). By doing 7

so, cost (a positive MCR per kWh) and benefit (a negative MCR per kWh) are 8

assessed for both “delivered” and “received” direction of flow. 9

In prior GRC Phase II filings, the primary MDCC was allocated at hourly 10

level based on customer coincident demand at division level4 using the 11

distribution PCAF method. In this proceeding, class-level demand contributions 12

to peak hours (i.e., PCAF demand) have been measured at the circuit and 13

substation (rather than division) levels. This more granular analysis is consistent 14

with PG&E’s current distribution investment planning process and is necessary 15

to properly quantify the impact of energy delivered to the grid as well as received 16

by the grid. The Secondary and New Business Primary capacity costs are 17

driven by non-coincident demand of the FLTs, and are, therefore, allocated 18

based on customer demand at their respective FLT using the distribution FLT 19

method. PG&E, in its methodology for FLT demand calculation, has 20

incorporated the direction of the electricity flow, i.e., “delivered” and “received” 21

flow, because FLTs, like the primary distribution circuits and substations, can 22

experience non-coincident demand due to “received” flow back to the grid. 23

Using the hourly level allocated costs, annual unit distribution MCRs are 24

calculated on a per kWh basis for the NEM and Non-NEM customers within 25

each customer class. As mentioned earlier, MCR per kWh for primary are also 26

calculated by TOU period, which are then used for revenue allocation and 27

rate design. 28

The substation marginal costs presented in Chapter 7 of this exhibit are also 29

used in PG&E’s distribution cost of service analysis. Substation MDCC 30

estimates have been multiplied by substation-level PCAF kw to calculate MCR 31

for the equipment that are designed based on load at substation level, rather 32

4 PG&E has 19 divisions in its service territory.

(PG&E-2)

8-5

than at primary circuit level. Finally, the total MCR per kWh for primary 1

distribution have been calculated by summing up the substation and circuit-level 2

MCRs and then dividing the total by respective kWh. 3

C. Distribution PCAF Methodology 4

PG&E has used the PCAF methodology for developing customer class-level 5

contributions to peak period demand for many years. The PCAF methodology 6

provides an hourly percentage allocation (i.e., hourly PCAFs) of primary MDCC 7

over a particular year. In this filing, PG&E has used 2017 recorded, 60-minute 8

interval data to calculate the hourly distribution PCAFs for its customer classes. 9

PG&E has applied winter derating factors at division level to the winter 10

60-minute customer class interval loads as the first step. This is done to 11

account for the fact that the distribution circuits have additional capacity during 12

the winter seasons due to lower ambient temperatures than the summer 13

seasons. In this filing, PG&E introduces two important improvements to its 14

standard PCAF methodology: (1) calculating customer class PCAFs at 15

distribution circuit and substation levels;5 and (2) measuring demand by taking 16

the absolute value of demand caused by “delivered” and “received” flow of 17

power. 18

Performing PCAF calculations at circuit and substation levels instead of 19

division level, as was done in the past, is needed because PG&E conducts its 20

distribution investment planning process at distribution circuit and substation 21

levels. Taking the absolute value of “delivered” and “received” demand is 22

needed because a distribution circuit may experience peak demand not only due 23

to “delivered” load, but also due to “received” load. An illustrative net load profile 24

obtained by taking absolute value of demand in both “delivered” and “received” 25

directions is shown in Figure 8-2. It is possible that the “received” peak 26

becomes greater than the “delivered” peak when a NEM resource generates so 27

much more electricity than the usage that the net “received” flow back to the grid 28

occurring in a time interval (i.e., hour) is greater than the net peak “delivered” 29

electricity occurring in another hour of a year. Additionally, some hours of the 30

“received load” may be large enough to fall within very high demand hours 31

5 For the equipment that are designed based on peak demand at substation level,

substation PCAF is used for cost allocation. For the equipment that are designed based on circuit level peak demand, PCAF at circuit level is used for cost al location.

(PG&E-2)

8-6

caused by the peak occurring during a “delivered” load hour, being above the 1

threshold load6 as defined in the Figure 8-2. In such scenarios, the distribution 2

feeder will experience peak demand hours (i.e., PCAF hours as defined in 3

Figure 8-2) from the “received” flow. Therefore, it is important to take the 4

absolute value of usage in each 60-minute interval of a year first, and then apply 5

the traditional PCAF calculation to the net absolute 60-minute load profile. 6

Please refer to the formula and explanation at the bottom portion of Figure 8-1 7

for the calculation details. 8

The immediate outcome of PCAF methodology is the set of PCAF weights 9

assigned to the hours (known as PCAF hours) that will be subject to capacity 10

cost allocation. The steps of calculating PCAF weights are identical for 11

substation and circuit-level equipment. PCAF weights are calculated by using 12

the formula shown at the bottom of Figure 8-2, and the total sum of the PCAF 13

weights is 100 percent. For each distribution circuit, PCAF weights are 14

multiplied with the respective customer class-level hourly load to obtain hourly 15

PCAF kW for each customer class. Finally, PCAF kW of all customer classes 16

are scaled such that the sum of total PCAF kW for a circuit equals the max 17

demand of that circuit. For each customer class, the hourly PCAF kW are then 18

aggregated first at division level, and then multiplied by the division specific 19

primary circuit MDCC estimate to obtain Primary Distribution Marginal Cost 20

Revenue (Primary DMCR) for the primary circuits. These calculation steps are 21

repeated for PCAF analysis at the substation level to obtain Primary DMCR for 22

substation equipment. Finally, primary circuit DMCR and substation DMCR are 23

added together to obtain the total Primary DMCR. 24

The calculated PCAF kW described in this section can be either positive or 25

negative, as illustrated in Figure 8-3. The concept of whether the customer 26

class load reduces or increases the demand at circuit level is used to determine 27

the sign of PCAF kW. If, during a PCAF hour, a customer class load is in the 28

opposite direction of the flow at the circuit level, then the circuit demand is 29

mitigated by that customer class load. Hence, that customer class PCAF kW 30

should be negative. On the other hand, if during a PCAF hour, a customer class 31

6 80 percent of maximum demand is used as “threshold” below which PCAF weight

becomes zero. (i.e., the hours with demand below the “threshold” do not get any cost allocation.) Please refer to Figure 8-2 for details.

(PG&E-2)

8-7

load is in the same direction of the flow at the circuit level, then the circuit 1

demand is being caused by that customer class load. Hence, that customer 2

class PCAF kW is positive. 3

FIGURE 8-2 ILLUSTRATION OF HANDLING “DELIVERED” AND “RECEIVED” FLOW

IN PCAF COMPUTATION

(PG&E-2)

8-8

FIGURE 8-3 EXPLANATION OF POSITIVE AND NEGATIVE PCAF-KW

Table 8-1 shows summarized circuit level PCAF results by month and by 1

hour of day for PG&E’s Non-NEM customers who have only “delivered” load. 2

The upper portion of the table (Table 8-1, Part 1) shows PCAF kW, and the 3

lower portion (Table 8-1, Part 2) shows the kWh. Based on the PCAF kW 4

estimates shown, PG&E finds that approximately 91 percent of the PCAF kW 5

have occurred in June through September, i.e., the summer months. In 6

Table 8-2, summary PCAF results for the NEM customers are shown in the 7

same format as the Table 8-1, for “delivered” load only. The important 8

difference to note is that for NEM customers, peak usage kWh occurs in later 9

evening hours in comparison to Non-NEM customers. This difference is caused 10

by NEM generation during the day hours. 11

(PG&E-2)

8-9

TABLE 8-1 DISTRIBUTION SYSTEM LEVEL PCAF USING 2017 RECORDED CIRCUIT DATA,

NON-NEM CUSTOMERS

_______________ Note: Updated (May 01, 2020).

(PG&E-2)

8-10

TABLE 8-2 DISTRIBUTION SYSTEM LEVEL PCAF USING 2017 RECORDED CIRCUIT DATA,

NEM CUSTOMERS, DELIVERED LOAD

_______________ Note: Updated (May 01, 2020).

Table 8-3 shows the summary PCAF results of NEM customers for their 1

“received” flow, indicating that the PCAF kW at this time are relatively small 2

occurring mostly in summer months during the hours of solar generation. 3

However, it is possible that PCAF kW due to “received” flow becomes bigger 4

over time due to increasing NEM adoption. 5

(PG&E-2)

8-11

TABLE 8-3 DISTRIBUTION SYSTEM LEVEL PCAF USING 2017 RECORDED CIRCUIT DATA,

NEM CUSTOMERS, RECEIVED LOAD

_______________ Note: Updates (May 01, 2020).

D. Distribution FLT Methodology 1

Non-coincident demand at the FLT is the cost driver for the new business 2

primary and secondary MDCCs. PG&E has, therefore, used FLT non-coincident 3

demand-based allocation of costs associated with secondary and new business 4

primary assets. As mentioned earlier in this chapter, PG&E has implemented 5

improvements to the FLT non-coincident demand methodology used in PG&E’s 6

2017 GRC Phase II filing. 7

First, PG&E has implemented non-coincident demand calculation by directly 8

using the load at the FLTs. In 2017 GRC Phase II filing, non-coincident demand 9

was calculated based on a sample of customer load and then diversity factors 10

(PG&E-2)

8-12

were applied for the residential and small light and power customers because 1

these customers share FLTs. However, this is no longer necessary. In this 2

filing, PG&E calculates FLT demand directly by using the load profile of FLTs. 3

Additionally, PG&E uses the full customer data instead of class kW sample data 4

that was used in its 2017 GRC Phase II filing. The other important 5

methodological change incorporated in this filing is that the direction of the 6

electricity flow has been considered: “delivered” and “received” electricity have 7

been treated in a way that demand from “received” electricity is also captured in 8

the calculations. This is particularly important for capturing the stress caused by 9

NEM customer’s “received” loads on PG&E’s secondary and new business 10

systems. In other words, for this GRC filing, a FLT’s non-coincident demand has 11

been measured by the largest 60-minute demand in a year, considering both 12

“delivered” and “received” flow directions. PG&E accomplishes this by taking 13

the absolute value of the 60-minute load in an FLT for the entire year and finding 14

the 60-minute interval that has the highest load which is called the 15

non-coincident peak demand hour. PG&E then finds the customer load at that 16

non-coincident peak demand interval. 17

The illustration in Figure 8-3 for positive and negative PCAF kW also applies 18

to the FLT kW. When an FLT kW is due to the “delivered” flow, the customer will 19

have a positive FLT kW if that customer also has a “delivered” load. Or, the 20

customer will have a negative FLT kW if the customer has a “received” load 21

(causing a reduction to the “delivered” peak). Alternatively, when an FLT kW is 22

due to the “received” flow, the customer will have a positive FLT kW if that 23

customer also has a “received” load. Or, the customer will have a negative FLT 24

kW if the customer has a “delivered” load (causing a reduction to the “received” 25

peak). It is important to note that “benefit” FLT kW can only occur for 26

transformers serving multiple customers, since that is the only case where a 27

customer has the ability to use energy in the opposite direction of the 28

transformer. For transformers serving single customers, (such as large 29

commercial customers) the FLT kW will always be a cost, regardless of which 30

direction of flow causes the peak load. 31

(PG&E-2)

8-13

There are situations when peak non-coincident demand in an FLT occurs at 1

multiple 60-minute intervals.7 Such multiple peak demand-related FLT demand 2

calculations for a customer are performed by taking an average over all the 3

peak intervals. 4

E. Comparison of Distribution Marginal Cost Revenue Per kWh Between NEM 5

and Non-NEM Customers at Distribution System Level 6

Table 8-4A presents the cost of service results for all customers, broken 7

down to show the Non-NEM and NEM sub-groups. Table 8-4A also shows the 8

costs and benefits for the “delivered” and the “received” flows, as well as for the 9

net flow. DMCR are shown separately for the primary MDCC and new business 10

primary and secondary MDCC. Total usage kWh is shown and finally the Total 11

DMCR per kWh is shown to provide perspectives on how the cost of service 12

estimates compare between the Non-NEM and NEM customer sub-groups. 13

New Business Primary DMCR and Secondary DMCR are combined since the 14

denominator is the same, i.e., FLT non-coincident demand for these 15

two categories. 16

The Non-NEM customers have “delivered” load, and no “received” load. 17

However, there are small “benefit” dollar amounts in comparison to the “cost” 18

dollar amounts for the primary and secondary voltage levels, caused by 19

consumption during the over-supply situations, which mostly occur during the 20

summer and winter off-peak periods.8 The net DMCR per kWh for the 21

Non-NEM customers is 2.572 cents per kWh. 22

The NEM customers have both “delivered” and “received” load-related 23

DMCRs. It is important to note that the DMCR “benefit” is relatively small in 24

comparison to the “cost” due to the “received” load resulting from NEM 25

generation that occurred during the hours that have minimal PCAF9 and FLT 26

kW. NEM “received” load occurs during the daytime, when the solar PV 27

generation takes place, while the higher weight PCAF and FLT hours occur 28

mostly during the late afternoons and evenings, approximately from 5 p.m. until 29

7 PG&E has observed multiple peak hours in the data caused by its Streetlight

customers. 8 TOU period specific analyses have been presented in Section G. 9 See Table 8-3 for PCAF and Table 8-6 for FLT for the NEM “received” flow.

(PG&E-2)

8-14

10 p.m., as can be seen from the “delivered” load PCAF and FLT results 1

(Tables 8-1 and 8-2).2

8-15

(PG&E-2) TA

BLE

8-4A

DI

STRI

BUTI

ON

COST

OF

SERV

ICE

COM

PARI

SON

BETW

EEN

NON-

NEM

AND

NEM

CUS

TOM

ERS,

201

7 RE

CORD

ED

Line

N

o.

Com

pone

nt

Flow

D

irect

ion

Form

ula

Non

-NE

M

NE

M

Cos

t B

enef

it N

et

Cos

t B

enef

it N

et

1 P

rimar

y D

MC

R

Del

iver

ed

$9

22,9

05,1

61

-$4,

670

$922

,900

,491

$8

4,56

6,34

2 -$

1,17

3 $8

4,56

5,16

9

2 R

ecei

ved

$149

,006

-$

6,45

1,86

0 -$

6,30

2,85

4

3 N

et

a +

b $9

22,9

05,1

61

-$4,

670

$922

,900

,491

$8

4,71

5,34

8 -$

6,45

3,03

3 $7

8,26

2,31

6

4 S

econ

dary

an

d N

ew

Bus

ines

s P

rimar

y

DM

CR

Del

iver

ed

$9

33,9

73,5

66

-$45

,108

$9

33,9

28,4

58

$87,

644,

685

-$17

0,54

3 $8

7,47

4,14

3

5 R

ecei

ved

$11,

721,

190

-$30

6,10

0 $1

1,41

5,09

0

6 N

et

d +

e $9

33,9

73,5

66

-$45

,108

$9

33,9

28,4

58

$99,

365,

875

-$47

6,64

3 $9

8,88

9,23

3

7 To

tal D

MC

R D

eliv

ered

a

+ d

$1,8

56,8

78,7

27

-$49

,778

$1

,856

,828

,949

$1

72,2

11,0

28

-$17

1,71

6 $1

72,0

39,3

12

8 R

ecei

ved

b +

e

$11,

870,

196

-$6,

757,

960

$5,1

12,2

36

9 N

et

c +

f $1

,856

,878

,727

-$

49,7

78

$1,8

56,8

28,9

49

$184

,081

,224

-$

6,92

9,67

6 $1

77,1

51,5

48

10

kWh

Del

iver

ed

72

,181

,698

,956

5,

639,

747,

925

11

Rec

eive

d

1,

838,

025,

810

12

Net

j -

k

72,1

81,6

98,9

56

3,80

1,72

2,11

5

13

Tota

l DM

CR

Per

kW

h D

eliv

ered

g

/ j

$0.0

2573

$0

.000

00

$0.0

2572

$0

.030

54

-$0.

0000

3 $0

.030

50

14

Rec

eive

d h

/ k

$0

.006

46

-$0.

0036

8 $0

.002

78

15

Net

i /

abs

(l)

$0.0

2573

$0

.000

00

$0.0

2572

$0

.048

42

-$0.

0018

2 $0

.046

60

____

____

____

N

ote

Cal

cula

tion

step

s sh

own

in th

e lig

ht g

reen

sha

ded

colu

mn

are

sam

e fo

r the

Tab

les

8-4

thro

ugh

8-8.

1

(PG&E-2)

8-16

In this analysis, NEM customers had approximately 5.64 billion kWh 1

“delivered” load and 1.84 billion kWh “received” load, resulting in 3.80 billion 2

kWh of net load in 2017. It is important to note that the net DMCR for NEM 3

customers is 4.66 cents per kWh, which is significantly higher than the DMCR 4

for “delivered” kWh of 3.050 cents per kWh. Although both “delivered” and 5

“received” flows resulted in “cost” DMCR, the fact that net flow is smaller than 6

the “delivered” flow (i.e., the net flow is significantly less than the absolute sum 7

of “delivered” and “received” kWh) caused the net DMCR per kWh to be 8

significantly higher than the “delivered” DMCR per kWh. 9

F. Comparison of DMCR Per kWh of NEM and Non-NEM Customers Within 10

the Customer Classes 11

Table 8-4B provides comparison of the total marginal DMCR per kWh 12

across four customer classes: Residential, Commercial, Agricultural and Large 13

Commercial and Industrial. The estimate shown in this table shows that the 14

difference between Non-NEM and NEM net DMCR per kWh is the highest for 15

the residential class, while that difference decreases for the rest of the classes. 16

This is driven by the load shape differences across these customer classes. 17

Residential class load shapes peak during the same hours as the system peak 18

(approximately 5 p.m. to 10 p.m.), while the load shapes of other classes show 19

different behavior. This causes the difference to be the greatest for the 20

Residential class. 21

Tables 8-5 through 8-8 present the detail by customer class, in a format 22

similar to the Table 8-4A. Table 8-5 provides cost of service analysis results for 23

residential customers, which shows a similar pattern as the cost of service 24

analysis results for all customers shown in Table 8-4A. An interesting 25

phenomenon can be seen in the net cost of “delivered” load. The Residential 26

NEM customers “delivered” load profiles are more expensive (4.096 cents per 27

kWh) than that of the Residential Non-NEM customers (3.202 cents per kWh). 28

The late evening peaking load profiles of residential customers cause this to 29

happen. In other words, the relatively greater consumption during the peak 30

evening hours compared to the day-time hours (when NEM generation takes 31

place) causes a higher cost for “delivered” load for residential NEM customers 32

than Non-NEM customers. 33

(PG&E-2)

8-17

Residential Non-NEM customers have a net total DMCR of 3.2020 cents per 1

kWh, while the Residential NEM customers have a net total DMCR of 2

8.865 cents per kWh. In summary, NEM customers have a significantly higher 3

MCR per kWh (approximately 5.663 cents per kWh higher) on a net basis. This 4

is also shown in Table 8-4B. 5

TABLE 8-4B NET AND DELIVERED MARGINAL DISTRIBUTION COST OF SERVICE COMPARISON

BETWEEN RESIDENTIAL NON-NEM AND NEM CUSTOMERS

_______________ Note Non-NEM delivered is same as Non-NEM net since Non-NEM does not have received load.

The values in this column can be compared with NEM delivered and NEM net separately to gain insight on the change in costs for the delivered load, and net load due to NEM.

Total DMCR/kWh by Customer Class based on 2017 Recorded DataNon-NEM NEM NEM Delivered/ Delivered Net

Net

Residential $0.03202 $0.04096 $0.08865

Commercial $0.02073 $0.02196 $0.02942

Agricultural $0.03792 $0.03011 $0.04653

Large Commerical & Industrial $0.01593 $0.01647 $0.01682

(PG&E-2)

8-18

TABLE 8-5 DISTRIBUTION COST OF SERVICE COMPARISON BETWEEN RESIDENTIAL NON-NEM AND

NEM CUSTOMERS, 2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $445,800,799 -$344 $445,800,456 $51,950,625 $0 $51,950,625

Received $0 -$4,532,018 -$4,532,018

Net $445,800,799 -$344 $445,800,456 $51,950,625 -$4,532,018 $47,418,607

2 Secondary and New Business Primary DMCR

Delivered $446,138,278 -$23,288 $446,114,989 $54,653,129 -$135,235 $54,517,893

Received $3,660,320 -$267,711 $3,392,608

Net $446,138,278 -$23,288 $446,114,989 $58,313,449 -$402,947 $57,910,502

3 Total DMCR

Delivered $891,939,077 -$23,632 $891,915,445 $106,603,754 -$135,235 $106,468,519

Received $3,660,320 -$4,799,730 -$1,139,410

Net $891,939,077 -$23,632 $891,915,445 $110,264,074 -$4,934,965 $105,329,109

4 kWh Delivered 27,853,656,478 2,599,496,305

Received 1,411,388,928

Net 27,853,656,478 1,188,107,378

5 Total DMCR Per kWh

Delivered $0.03202 $0.00000 $0.03202 $0.04101 -$0.00005 $0.04096

Received $0.00259 -$0.00340 -$0.00081

Net $0.03202 $0.00000 $0.03202 $0.09281 -$0.00415 $0.08865 ___________

Note Please refer to Table 8-4A, column shaded light green, for calculation steps.

Table 8-6 provides the results of cost of service analysis for the Non-NEM 1

and NEM customers within the Commercial customer class. The results are in 2

the same direction as seen for all customers in Table 8-4A, and for the 3

residential customers in Table 8-5. However, the NEM customers, although 4

having higher total net DMCR of 2.942 cents per kWh, than the Non-NEM 5

customers’ total net DMCR of 2.073 cents per kWh, the difference is less in 6

comparison to the residential customers. The divergence of DMCR per kWh 7

between NEM and Non-NEM customers is greatest in the Residential customer 8

class. Table 8-7 shows the cost of service details for the Agricultural class 9

customers Non-NEM and NEM sub-groups. The difference in net DMCR/kWh 10

between Non-NEM and NEM is small; NEM shows a slightly higher cost. This is 11

due to the load shape of the Agricultural customers. These load shapes typically 12

show that NEM adoption causes reduction in load of relatively average cost 13

hours in a day (rather than low cost hours seen in residential load shape), 14

(PG&E-2)

8-19

thereby being more effective in comparison to other classes, such as the 1

Residential class, in offsetting the overall cost. Also, Agricultural class 2

Non-NEM load shape show that their peak load hours occur approximately 3

during 8 a.m. through 1 p.m. Consequently, the Agricultural class NEM cost 4

does not become significantly higher than the Non-NEM counterpart. Table 8-8 5

shows the DMCR per kWh estimates for the Large Commercial and Industrial 6

customers. The difference in DMCR/kWh for the Non-NEM and NEM customer 7

sub-groups is small, indicating that the load shapes of these two sub-groups are 8

quite similar. 9

TABLE 8-6 DISTRIBUTION COST OF SERVICE COMPARISON BETWEEN COMMERCIAL NON-NEM AND

NEM CUSTOMERS, 2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $315,085,271 -$1,101 $315,084,170 $19,838,297 -$1,173 $19,837,124

Received $101,977 -$1,485,653 -$1,383,676

Net $315,085,271 -$1,101 $315,084,170 $19,940,274 -$1,486,826 $18,453,448

2 Secondary and New Business Primary DMCR

Delivered $322,319,222 -$16,957 $322,302,266 $21,865,890 -$18,597 $21,847,293

Received $6,651,394 -$34,782 $6,616,612

Net $322,319,222 -$16,957 $322,302,266 $28,517,284 -$53,379 $28,463,905

3 Total DMCR

Delivered $637,404,493 -$18,058 $637,386,435 $41,704,187 -$19,770 $41,684,417

Received $6,753,371 -$1,520,435 $5,232,936

Net $637,404,493 -$18,058 $637,386,435 $48,457,558 -$1,540,205 $46,917,352

4 kWh Delivered 30,745,922,560 1,898,466,341

Received 303,480,963

Net 30,745,922,560 1,594,985,377

5 Total DMCR Per kWh

Delivered $0.02073 $0.00000 $0.02073 $0.02197 -$0.00001 $0.02196

Received $0.02225 -$0.00501 $0.01724

Net $0.02073 $0.00000 $0.02073 $0.03038 -$0.00097 $0.02942 _______________ Note Please refer to Table 8-4A, column shaded light green, for calculation steps.

(PG&E-2)

8-20

TABLE 8-7 DISTRIBUTION COST OF SERVICE COMPARISON BETWEEN AGRICULTURAL NON-NEM AND

NEM CUSTOMERS, 2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $83,809,349 -$3,225 $83,806,124 $5,493,163 $0 $5,493,163

Received $47,029 -$389,304 -$342,275

Net $83,809,349 -$3,225 $83,806,124 $5,540,192 -$389,304 $5,150,888

2 Secondary and New Business Primary DMCR

Delivered $107,905,678 -$4,863 $107,900,815 $5,743,255 -$16,710 $5,726,545

Received $1,301,675 -$3,168 $1,298,507

Net $107,905,678 -$4,863 $107,900,815 $7,044,930 -$19,878 $7,025,052

3 Total DMCR

Delivered $191,715,027 -$8,088 $191,706,939 $11,236,418 -$16,710 $11,219,708

Received $1,348,704 -$392,472 $956,232

Net $191,715,027 -$8,088 $191,706,939 $12,585,122 -$409,182 $12,175,940

4 kWh Delivered 5,055,773,386 372,588,919

Received 110,898,987

Net 5,055,773,386 261,689,932

5 Total DMCR Per kWh

Delivered $0.03792 $0.00000 $0.03792 $0.03016 -$0.00004 $0.03011

Received $0.01216 -$0.00354 $0.00862

Net $0.03792 $0.00000 $0.03792 $0.04809 -$0.00156 $0.04653 _______________ Note Please refer to Table 8-4A, column shaded light green, for calculation steps.

(PG&E-2)

8-21

TABLE 8-8 DISTRIBUTION COST OF SERVICE COMPARISON BETWEEN LARGE COMMERICAL AND

INDUSTRIAL NON-NEM AND NEM CUSTOMERS, 2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $78,209,742 $0 $78,209,742 $7,284,257 $0 $7,284,257

Received $0 -$44,884 -$44,884

Net $78,209,742 $0 $78,209,742 $7,284,257 -$44,884 $7,239,373

2 Secondary and New Business Primary DMCR

Delivered $57,610,388 $0 $57,610,388 $5,382,411 $0 $5,382,411

Received $107,802 -$439 $107,363

Net $57,610,388 $0 $57,610,388 $5,490,213 -$439 $5,489,774

3 Total DMCR

Delivered $135,820,130 $0 $135,820,130 $12,666,669 $0 $12,666,669

Received $107,802 -$45,323 $62,478

Net $135,820,130 $0 $135,820,130 $12,774,470 -$45,323 $12,729,147

4 kWh Delivered 8,526,346,532 769,196,360

Received 12,256,932

Net 8,526,346,532 756,939,428

5 Total DMCR Per kWh

Delivered $0.01593 $0.00000 $0.01593 $0.01647 $0.00000 $0.01647

Received $0.00880 -$0.00370 $0.00510

Net $0.01593 $0.00000 $0.01593 $0.01688 -$0.00006 $0.01682 _______________ Note See page 8-17 for insight on the numbers.

G. Primary DMCR Per kWh Comparison During the TOU Periods 1

PG&E has studied the cost of service by TOU period,10 and the results of 2

the analysis are presented in this section. For this purpose, only Primary DMCR 3

per kWh has been analyzed by TOU period, because secondary FLT-based 4

DMCR is related to the annual non-coincident demand at the Final Line 5

Transformers. The new connection related capacity costs captured under “New 6

Business Primary” category are the costs reported in Major Work Category 7

(MWC) 16 only, which represent those connection related capacity costs that are 8

customer specific and are designed based on the annual peak demand at the 9

10 TOU periods are defined for summer and winter seasons. Summer months are from

June through September, and the rest of the months of the year are winter months. Peak period for both Summer and Winter is from 4 p.m. to 9 p.m., partial-summer peak period is from 2 p.m. to 4 p.m., and 9 p.m. to 11 p.m. Winter Super off-peak period occurs in March through May, from 9 a.m. through 2 p.m. The rest of the periods are considered off-peak periods.

(PG&E-2)

8-22

respective FLTs. Circuit or substation costs related to new connections, if there 1

are any, are captured under MWC 06 and are included in calculating Primary 2

DMCR per kWh. Therefore, the Primary DMCR per kWh should be TOU 3

differentiable, while the New Business Primary DMCR/kWh should not be TOU 4

differentiated. 5

Table 8-9 shows a comparison between Non-NEM and NEM customers. 6

For the Non-NEM customers, there is no “received” load, and therefore, the 7

estimates are the same for “delivered” and “net” flows. For the “delivered” flow, 8

NEM customers have significantly higher MCR per kWh during the summer peak 9

period in comparison to the Non-NEM customers, while it is lower for the rest of 10

the TOU periods. This makes sense because the peak period occurs in the late 11

evening when solar generation is much less available compared to the day-time. 12

Net DMCR/kWh is, however, higher for the NEM customers for all summer TOU 13

periods. 14

TABLE 8-9 COMPARISON OF PRIMARY DMCR PER KWH BETWEEN NON-NEM AND NEM CUSTOMERS

BASED ON 2017 RECORDED DATA

________________

Note: Table 8-10 shows the primary MCR, kWh and the MCR per kWh estimates from the 2017 recorded data for the summer peak period. NEM load profiles are more expensive during the summer peak period, than the Non-NEM load profile, which is evident in the results shown in this table.

Primary DMCR/kWh by TOU Period based on 2017 Recorded DataNon-NEM NEM NEM Delivered/ Delivered Net

NetSummer Peak $0.07780 $0.09911 $0.10920Summer Part-peak $0.03124 $0.02679 $0.03824Summer Off-peak $0.00778 $0.00644 $0.00908Winter Peak $0.00360 $0.00305 $0.00329Winter Off-peak $0.00121 $0.00094 $0.00119Winter Super Off-peak $0.00183 $0.00179 $0.00093

(PG&E-2)

8-23

TABLE 8-10 DISTRIBUTION COST OF SERVICE COMPARISON DURING SUMMER PEAK PERIOD,

2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $554,952,878 – $554,952,878 $61,341,822 – $61,341,822

Received – $(2,518,719) (2,518,719)

Net $554,952,878 – $554,952,878 $61,341,822 $(2,518,719) $58,823,103

2 kWh Delivered 7,133,447,743 618,921,868

Received 80,241,705

Net 7,133,447,743 538,680,164

3 Primary DMCR Per kWh

Delivered $0.07780 – $0.07780 $0.09911 – $0.09911

Received – (0.03139) (0.03139)

Net $0.07780 – $0.07780 $0.11387 $(0.00468) $0.10920

Table 8-11 below shows the primary MCR, kWh and the MCR per kWh 1

estimates based on 2017 recorded data for the summer part-peak period. For 2

the summer part-peak period, NEM load profiles are more expensive for the net 3

flow. 4

TABLE 8-11 DISTRIBUTION COST OF SERVICE COMPARISON DURING SUMMER PART-PEAK PERIOD,

2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $166,022,252 $(22) $166,022,230 $10,444,444 $(147) $10,444,297

Received 12,523 (2,145,384) (2,132,861)

Net $166,022,252 $(22) $166,022,230 $10,456,967 $(2,145,531) $8,311,436

2 kWh Delivered 5,313,578,380 389,882,842

Received 172,561,862

Net 5,313,578,380 217,320,979

3 Primary DMCR Per kWh

Delivered $0.03124 – $0.03124 $0.02679 – $0.02679

Received 0.00007 $(0.01243) (0.01236)

Net $0.03124 – $0.03124 $0.04812 $(0.00987) $0.03824

Table 8-12 shows the primary MCR, kWh and MCR per kWh estimates from 5

the 2017 recorded data for the summer off-peak period. For summer off-peak 6

period, the primary PCAF kW and the estimated costs for Non-NEM and NEM 7

(PG&E-2)

8-24

customers are relatively small. NEM customer load profiles continue to be more 1

expensive for the net flow. 2

TABLE 8-12 DISTRIBUTION COST OF SERVICE COMPARISON DURING SUMMER OFF-PEAK PERIOD,

2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $121,162,826 $(1,474) $121,161,353 $7,423,805 $(703) $7,423,102

Received 74,990 (1,436,162) (1,361,172)

Net $121,162,826 $(1,474) $121,161,353 $7,498,795 $(1,436,865) $6,061,930

2 kWh Delivered 15,571,195,330 1,152,226,963

Received 484,304,272

Net 15,571,195,330 667,922,690

3 Primary DMCR Per kWh

Delivered $0.00778 – $0.00778 $0.00644 – $0.00644

Received 0.00015 $(0.00297) (0.00281)

Net $0.00778 – $0.00778 $0.01123 $(0.00215) $0.00908

Tables 8-13, 8-14, and 8-15 show the winter TOU period results for the 3

Non-NEM and NEM customers. PCAF is mostly concentrated in the summer 4

season. Therefore, we see relatively small costs during these TOU periods. 5

NEM load profiles are found to be slightly less costly during the winter TOU 6

periods. The winter daily load shapes are more flat in comparison to the 7

summer load shapes. As a result, generation from NEM customers is more 8

effective in reducing the overall cost.11 9

11 A comparison between Table 8-10 (summer peak period results) and Table 8-13 (winter

peak period results) show that the net MCR for NEM customers account for 9.6 percent of the combined net MCR (NEM and Non-NEM) in the summer peak period, while that proportion is 6.9 percent for the winter peak period.

(PG&E-2)

8-25

TABLE 8-13 DISTRIBUTION COST OF SERVICE COMPARISON DURING WINTER PEAK TOU PERIOD,

2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $37,662,257 – $37,662,257 $2,824,237 – $2,824,237

Received – $(45,833) (45,833)

Net $37,662,257 – $37,662,257 $2,824,237 $(45,833) $2,778,404

2 kWh Delivered 10,465,577,216 926,984,820

Received 81,851,170

Net 10,465,577,216 845,133,650

3 Primary DMCR Per kWh

Delivered $0.00360 – $0.00360 $0.00305 – $0.00305

Received – $(0.00056) (0.00056)

Net $0.00360 – $0.00360 $0.00334 $(0.00005) $0.00329

TABLE 8-14 DISTRIBUTION COST OF SERVICE COMPARISON DURING WINTER OFF-PEAK PERIOD,

2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $36,008,973 $(3,174) $36,005,799 $2,271,077 $(61) $2,271,016

Received 40,826 (230,337) (189,511)

Net $36,008,973 $(3,174) $36,005,799 $2,311,903 $(230,398) $2,081,505

2 kWh Delivered 29,822,624,131 2,405,747,256

Received 652,203,608

Net 29,822,624,131 1,753,543,648

3 Primary DMCR Per kWh

Delivered $0.00121 – $0.00121 $0.00094 – $0.00094

Received 0.00006 $(0.00035) (0.00029)

Net $0.00121 – $0.00121 $0.00132 $(0.00013) $(0.00119)

(PG&E-2)

8-26

TABLE 8-15 DISTRIBUTION COST OF SERVICE COMPARISON DURING WINTER SUPER OFF-PEAK

PERIOD, 2017 RECORDED

Line No.

Non-NEM NEM

Cost Benefit Net Cost Benefit Net

1 Primary DMCR

Delivered $7,095,975 – $7,095,975 $260,957 $(263) $260,695

Received 20,667 (75,424) (54,757)

Net $7,095,975 – $7,095,975 $281,624 $(75,686) $205,938

2 kWh Delivered 3,875,276,155 145,984,175

Received 366,863,192

Net 3,875,276,155 (220,879,017)

3 Primary DMCR Per kWh

Delivered $0.00183 – $0.00183 $0.00179 – $0.00179

Received 0.00006 $(0.00021) (0.00015)

Net $0.00183 – $0.00183 $0.00128 $(0.00034) $0.00093

H. Conclusion 1

PG&E is seeking the Commission’s approval of the distribution cost of 2

service methodologies presented in this chapter, including the PCAF and 3

FLT cost, as well as the results and corresponding estimates included in the 4

Marginal Cost table (MC table)12 used for revenue allocation. Key features of 5

the methodology are: (1) treating “delivered” and “received” flow of electricity 6

separately in cost analysis; (2) coincident demand driven cost driver (PCAF) 7

applied at distribution circuit and substation levels; and (3) non-coincident 8

demand cost driver FLT. These key features, along with other changes in cost 9

allocation as described in this chapter, have resulted in a more granular and 10

accurate cost allocation. PG&E believes that the methodology presented is 11

well-suited for analyzing the impacts of technology adoption, such as NEM 12

and storage. 13

12 MC table is presented in Attachment A of Chapter 1.

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 9

MARGINAL CUSTOMER ACCESS COSTS

(PG&E-2)

9-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 9

MARGINAL CUSTOMER ACCESS COSTS

TABLE OF CONTENTS

A. Introduction ............................................................................................................ 9-1

B. Methodology to Estimate Marginal Connection Equipment Cost ....................... 9-4

1. RECC Method to Estimate Marginal Connection Equipment Cost .............. 9-4

2. Underlying Connection Equipment Cost Data .............................................. 9-5

3. Rule 16 Service Extension Tariffs Is the Basis for Determining Connection Equipment Cost .......................................................................... 9-7

4. Additional Model Improvements .................................................................... 9-9

C. Superiority of RECC Method in Estimating Connection Equipment Costs ...... 9-10

1. Historical Use of the RECC Method and Recent Adoption by CPUC for Gas Distribution Revenue Allocation ..................................................... 9-11

2. Results Under the RECC Method Are Consistent With Marginal Cost Theory and Formulation ............................................................................... 9-12

3. The New Connection Rate Used in the NCO Method Lacks Stability; Results Can Be Highly Volatile When Number of Customers and New Connections Are Small................................................................................. 9-13

4. NCO Method Understates Marginal Cost, Often Severely ......................... 9-14

5. Under NCO, Revenue Allocation Is Dominated by Marginal Distribution Capacity Cost, Inappropriately Lessening the Influence of Marginal Customer Access Cost ................................................................. 9-15

D. Introduction to Revenue Cycle Services Marginal Costs .................................. 9-17

E. Historical Background on RCS Marginal Costs Methodology .......................... 9-17

F. Revenue Cycle Services Model.......................................................................... 9-19

1. Methodology ................................................................................................. 9-19

2. An Example of Allocation of RCS Costs ..................................................... 9-20

3. Changes in Financial Costs Model .............................................................. 9-23

4. Correction in Account Set-Up Calculation................................................... 9-24

5. Exclusion of Meter Replacement Costs ...................................................... 9-25

(PG&E-2) PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 9 MARGINAL CUSTOMER ACCESS COSTS

TABLE OF CONTENTS

(CONTINUED)

9-ii

6. Streetlight Changes ...................................................................................... 9-26

G. 2020 GRC RCS Results Segmented by NEM and Non-NEM Customers ....... 9-26

1. Details of Residential Customers’ RCS Marginal Costs............................. 9-27

a. Meter Services ....................................................................................... 9-28

2. Billing and Payments .................................................................................... 9-29

H. Conclusion ........................................................................................................... 9-30

(PG&E-2)

9-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 9 2

MARGINAL CUSTOMER ACCESS COSTS 3

A. Introduction 4

This chapter presents Pacific Gas and Electric Company’s (PG&E or the 5

Utility) proposed methodology and results for its Marginal Customer Access 6

Costs (MCAC) which includes Marginal Connection Equipment Costs and 7

Marginal Revenue Cycle Services (RCS) costs. 8

The California Public Utilities Commission (CPUC or Commission) has 9

historically distinguished between two types of electric distribution costs: 10

1) demand-related costs, which vary with the level of electricity demand 11

(in kilowatts (kW)) (discussed in Chapter 7); and 12

2) customer access-related costs (discussed in this chapter), which vary with 13

the number of customer connections. 14

In this chapter, PG&E seeks to continue the practice of developing MCAC 15

based on the sum of two previously-adopted components:1 16

1) the Marginal Connection Equipment Costs associated with the equipment 17

necessary for connecting customers to the grid, which include Transformer, 18

Service Drop, and Meter (TSM)—collectively, the “TSM”, or “connection 19

equipment”; and 20

2) the Marginal RCS Costs associated with serving customers once connected 21

to the grid on an ongoing basis.2 22

However, compared to PG&E’s previous filings, this filing provides additional 23

granularity in the MCAC estimates. PG&E, for the first time, is separately 24

breaking-out RCS component of MCAC estimates, and therefore, the total 25

1 PG&E first proposed the distinction of two components of MCAC in its 1993 General

Rate Case (GRC), and this distinction was adopted by the CPUC in Decision (D.) 92-12-057.

2 Marginal RCS costs include the costs of meter reading, meter services, account setup, billing and payment processing, and credit and collections.

(PG&E-2)

9-2

MCAC estimate for customers with and without Net Energy Metering (NEM).3 1

However, PG&E also shows the combined MCAC estimate for all customers 2

(NEM and Non-NEM) within each customer class, for reference and comparison 3

with estimates filled in prior GRCs. 4

PG&E is requesting that the Commission find reasonable PG&E’s estimates 5

of MCAC as well as the underlying methodologies and associated costs used to 6

prepare its results, including the methodology and costs used for separately 7

identifying NEM and Non-NEM MCACs. 8

For connection equipment costs (i.e., TSM costs), PG&E requests that the 9

Commission find the Real Economic Carrying Charge (RECC) Method4 10

appropriate for estimating the marginal connection equipment cost component of 11

MCAC and not the New Customer Only (NCO)5 Method, because the NCO 12

Method incorrectly calculates the marginal costs of connection equipment as 13

well as results in volatile marginal estimates for relatively small customer 14

classes. Additionally, PG&E has recently demonstrated in its gas allocation 15

proceedings how important it is to use an RECC method-based estimate in lieu 16

of an NCO-based estimate to achieve equitable revenue allocation across 17

customer classes, which was adopted by CPUC (D.19-10-036). The merits of 18

using the RECC Method are presented in detail in Section C of this chapter.6 19

For the RCS costs, PG&E requests that the Commission find PG&E’s 20

activity-based costing methodology presented in this chapter reasonable. 21

Table 9-1 presents a table of PG&E’s customer classes for which MCAC 22

values are estimated. PG&E has kept the same customer classes as were 23

3 PG&E has not separated Marginal Connection Equipment Cost for NEM and Non-NEM

because connection equipment (meter and service) that fall under Rule 16 are same for all new connections irrespective of being either NEM or Non-NEM. Also, the sample of NEM customers in the January 2016 through June 2018 data is very small (0.3 percent of total new customers).

4 Adopted in D.89-12-057 for PG&E’s electric marginal costs. (Sometimes the RECC Method has been referred to as the “Rental Method;” this filing will use the term RECC.)

5 The NCO Method, which has often also been called the One-Time Hook-Up Cost method, was first adopted for use by PG&E in D.92-12-057 for electric marginal costs and in D.95-12-053 for gas marginal costs.

6 Although PG&E is proposing adoption of the RECC Method, PG&E also includes, some detail on the currently-adopted NCO methodology, in Attachment A of this chapter, along with the estimated results under the NCO Method to allow parties to compare PG&E’s proposed RECC results with the results based on the NCO Method.

(PG&E-2)

9-3

defined in its 2017 GRC Phase II filing, except that the Streetlight class definition 1

has been modified. The Streetlights customer class, previously presented as a 2

single customer class, has now been presented in four separate schedules: 3

LS1, LS2, LS3 and OL1. Note that PG&E made a few changes in its 2017 GRC 4

Phase II filing relative to its prior GRC Phase II filings (however, the MCAC 5

estimates were not litigated in that proceeding).7 6

Table 9-1, below, presents PG&E’s estimates of MCAC values by customer 7

class in dollars per customer-year, using PG&E’s proposed RECC Method for 8

estimating the marginal costs of the connection equipment (TSM). 9

TABLE 9-1 PROPOSED TEST YEAR 2021 MCAC ESTIMATES

(DOLLARS PER CUSTOMER-YEAR)

Line No. Customer Class

Sub-Class or Rate Schedule

MCAC Values (RECC Method-Based)

Combined(a) NEM Non-NEM

1 Residential $138.25 $207.79 $134.45 2 Agricultural Ag A $837.76 $1,286.08 $825.59 3 Ag B – Small $2,665.58 $3,309.54 $2,621.68 4 Ag B – Large $2,698.27 $3,364.13 $2,678.09 5 Small Commercial Single Phase $414.40 $806.64 $410.62 6 Poly Phase $1,454.39 $1,972.38 $1,442.85 7 Medium Commercial A-10-S/E-19VS $3,689.15 $4,230.56 $3,679.87 8 A-10-P/E-19VP $4,281.21 $4,822.61 $4,271.92 9 Large L&P E-19-S $6,676.24 $6,814.13 $6,656.44

10 E-19-P $5,658.96 $5,796.85 $5,639.15 11 E-19-T $5,658.96 $5,796.85 $5,639.15 12 E-20-S $7,117.25 $7,255.14 $7,097.45 13 E-20-P $5,658.96 $5,796.85 $5,639.15 14 E-20-T $5,658.96 $5,796.85 $5,639.15 15 Streetlights LS1 $44.74 – $44.74 16 LS2 $25.67 – $25.67 17 LS3 $65.66 – $65.66 18 OL1 $14.76 – $14.76 19 Traf f ic Control $474.81 – $474.81

_______________ (a) Note that Marginal Connection Equipment Cost component of MCAC is same for both

NEM and Non-NEM, as PG&E uses the combined result of adding the NEM and Non-NEM MCAC values for reference and comparison with estimates filed in prior GRCs. Marginal RCS costs for the combined scenario are calculated as average costs obtained from NEM and Non-NEM combined data, without distinguishing whether an observation is NEM or not.

7 Changes made were: (a) breaking out Small L&P into single phase and polyphase; and

(b) combining E-19V with Medium L&P.

(PG&E-2)

9-4

The remainder of this chapter is organized as follows: 1

Section B – Methodology to Estimate Marginal Connection Equipment Cost 2

Section C – Superiority of RECC Method in Estimating Connection 3

Equipment Costs 4

Section D – Introduction to Revenue Cycle Services Marginal Costs 5

Section E – Historical Background on RCS Marginal Costs Methodology 6

Section F – Revenue Cycle Services Model 7

Section G – 2020 GRC RCS Results Segmented by NEM and Non-NEM 8

Customers 9

Section H – Conclusion 10

B. Methodology to Estimate Marginal Connection Equipment Cost 11

This section discusses PG&E’s proposed RECC Method and the MCAC 12

estimates based on RECC Method, the underlying data and Rule 16. Rule 16 is 13

the basis for determining connection equipment costs. 14

1. RECC Method to Estimate Marginal Connection Equipment Cost 15

PG&E’s RECC method-based calculation of the Marginal Connection 16

Equipment Cost is accomplished in two steps. First, at the margin, the 17

per-customer access equipment (i.e., TSM) costs for a connection are 18

determined for each customer class using the actual new connection 19

contract cost data. Second, the per-customer access equipment costs are 20

converted to an annualized real carrying cost by using an RECC factor and 21

other appropriate loaders to include expense costs associated with the TSM 22

equipment capital costs. 23

Marginal Connection Equipment Cost = Per-Customer TSM Cost x (RECC + Loaders)

The development of the RECC factor and the Loaders are discussed in 24

Chapter 10, “Marginal Cost Loaders and Financial Factors,” of this exhibit. 25

Marginal Connection Equipment Costs of all customer classes are 26

presented in Table 9-2. 27

(PG&E-2)

9-5

TABLE 9-2 PROPOSED TEST YEAR 2021 MARGINAL CONNECTION EQUIPMENT COST ESTIMATES

(RECC METHOD) (DOLLARS PER CUSTOMER-YEAR)

Line No. Customer Class

Sub-Class or Rate Schedule

Marginal Connection Equipment

Cost

1 Residential $103.11 2 Agricultural Ag A $743.00 3 Ag B – Small $2,476.63 4 Ag B – Large $2,476.63 5 Small Commercial Single Phase $369.43 6 Poly Phase $1,393.58 7 Medium Commercial A-10-S/E-19VS $3,532.12 8 A-10-P/E-19VP $4,124.17 9 Large L&P E-19-S $5,310.04

10 E-19-P $4,292.76 11 E-19-T $4,292.76 12 E-20-S $5,751.05 13 E-20-P $4,292.76 14 E-20-T $4,292.76 15 Streetlights LS1 $31.40 16 LS2 $17.85 17 LS3 $52.58 18 OL1 – 19 Traf f ic Control $443.25

2. Underlying Connection Equipment Cost Data 1

PG&E’s 2020 GRC Phase II Marginal Connection Equipment Cost 2

estimates are based on an analysis of January 2016 – June 2018 3

(30 months) customer connection cost data obtained from PG&E’s 4

Customer Contract Billing System (CCBS).8 The CCBS data are 5

supplemented with new connection contract-specific transformer data from 6

PG&E’s Fast Flow Estimating tool, certain other new connection data from 7

PG&E’s Main Line Extension System, and rate schedule information from 8

the Customer Care and Billing (CC&B) system. 9

The use of actual job cost estimate as a data source for this approach to 10

compute marginal costs of connecting new customers differs from previously 11

adopted methods in a significant way. This approach uses costs that are 12

based on actual field-produced job cost estimates obtained from customer 13

8 The CCBS application applies the tariff and accounting rules to the project costs to

determine how much the customer pays and how much PG&E pays for the construction work related to new customer connections.

(PG&E-2)

9-6

contracts in PG&E’s CCBS application rather than a limited number of 1

estimated “typical customer connection” costs.9 Using actual field-estimated 2

job costs in place of “typical customer connection” costs represents a 3

significant improvement in the accuracy of the methodology and should be 4

adopted. 5

In this 2020 GRC Phase II showing on customer marginal costs for 6

new connections, PG&E is using data representing more than 104,000 new 7

service connections. Non-residential marginal customer connection costs 8

are based on contract data for more than 37,000 new service connections. 9

Residential marginal customer connection costs are based on contract data 10

for more than 66,000 new connections. 11

Similar to what was done in PG&E’s 2003, 2007, 2011, 2014 and 12

2017 GRC Phase II MCAC calculations, the 2020 GRC Phase II marginal 13

customer connection costs presented here are the average (i.e., mean) net 14

PG&E connection cost for each customer class, that is, exclusive of 15

applicant-borne costs as determined by Rule 16.10,11 PG&E proposes to 16

use an average of the net new connection cost because it recognizes the full 17

range of data, including the valid data on costs at the low- and high-ends of 18

the cost spectrum. 19

The following process is used to estimate new connection costs of 20

residential and non-residential customer classes: 21

a. Determine the mean cost of a residential connection (where “total 22

service cost” as referenced here and in the remainder of this section is 23

inclusive of transformer and meter costs): 24

9 “Typical job costs” were developed for each of the various customer classes using

engineering estimates produced from PG&E’s former COMPRESS estimating tool, not from actual, f ield-estimated job costs. This method was used in PG&E’s 1990, 1993 and 1996 GRC Phase II proceedings. See the 1996 GRC Phase II, Exhibit 302, p. 7-8.

10 PG&E continues to see a wide dispersion of new connection costs for the non-residential customer classes, with all observed values found to be valid. Therefore, all such values were used in PG&E’s mean new connection cost calculation.

11 Background on Rule 16 limits to PG&E’s new connection equipment cost responsibili ty is provided in Section D, below.

(PG&E-2)

9-7

For each residential contract, determine PG&E’s share of the new 1

connection costs, which is limited to the per-meter allowance 2

granted;12 and 3

For the mean residential new connection cost, PG&E obtains the 4

breakout of the transformer, service and meter costs. 5

b. Determine the mean meter and service cost of non-residential new 6

connections: 7

For each non-residential contract, determine PG&E’s share of the 8

new connection costs, which is limited to the allowance granted and 9

the value of the 50 percent discount given if the 50 percent 10

non-refundable discount option is selected in favor of the ten-year 11

refundable payment option; and 12

After matching Service Point and Premise ID to rate schedule data 13

from CC&B, calculate mean meter and mean total service costs by 14

non-residential customer class. 15

Finally, for each class’s mean new connection cost, PG&E determines 16

the breakout for the TSM costs. 17

3. Rule 16 Service Extension Tariffs Is the Basis for Determining 18

Connection Equipment Cost 19

Consistent with prior filings, PG&E proposes to capture new connection 20

cost-sharing in this 2020 GRC proposal. This cost-sharing is defined by the 21

service extension tariff Rule 16 and quantified with analyses of actual CCBS 22

contract data. Capturing this cost-sharing ensures that customer new 23

connection cost results only capture the marginal cost incurred by PG&E. 24

When a customer or developer applies for electric service at a new location, 25

the following costs are incurred in most cases: 26

a) The cost of planning, designing and engineering service extensions 27

using PG&E’s design standards for design, materials and construction; 28

b) The cost to install primary or secondary service conductors (service 29

drops) to connect to individual customer service locations (and, for 30

12 For residential new connection costs analyses, the allowance is $1,918 for all new

connection contracts between July 1, 2009 and January 1, 2018 and $2,154 thereafter (Advice Letter 3438-E-A).

(PG&E-2)

9-8

overhead service conductors, the cost of poles that may be needed to 1

support the service conductors); and 2

c) The cost to install a meter for each new customer. 3

These three items are the major cost items that fall within the purview of 4

the Commission’s service extension tariff (Rule 16). When a new customer 5

or developer applies for electric service at a new premise, Rule 16 governs 6

which costs of connecting new customers are borne by the Utility (and 7

therefore included in rates), and which costs are directly borne by the 8

applicants (customers or developers) for new service (and therefore not 9

included in rates). Further, Rule 16 governs which costs are subject to 10

allowance. Allowances granted to customers or developers are effectively 11

costs borne by the Utility and are included in utility marginal costs. 12

Prior to 1995, the line extension rules provided allowances that were 13

sufficiently generous that the Utility, and ultimately ratepayers, paid most of 14

the cost of connecting new customers.13 This meant that the line extension 15

rules had no practical effect on MCAC. This situation began to change in 16

the mid-1990s due to a series of decisions in the Commission’s Line 17

Extension Rulemaking (R.92-03-050). Beginning in 1995, Rules 15 and 16 18

were modified so that allowances had to be “revenue justified.” A “flat,” 19

per-meter residential allowance was adopted for Electric Rule 15 (and 20

referenced by Rule 16). In 1998, transformers, meters and services were, 21

for the first time, included in the customer’s costs, subject to allowance. 22

With these changes, developers and customers are now often required to 23

absorb a portion of new customer connection costs. 24

PG&E proposes to use the full refundable cost14 of the service 25

extension (the cost of the customer access equipment) for the purpose of 26

estimating the new connection component of the MCAC. This amount 27

13 In general, an applicant for new service was required to make a deposit to cover the

“refundable cost” of connection. Deposits were usually fully refunded, over a period of up to ten years.

14 Rule 16 specifies that certain items are normally the cost responsibility of the developer or customer. Such items are excluded from service costs subject to allowance and from the refundable costs. Generally, applicants for new service have the option to install such facilities at their own expense, or to make a nonrefundable payment to PG&E to cover the cost of utility installation. In either case, the cost of such facilities is properly excluded from the marginal costs used for ratemaking.

(PG&E-2)

9-9

excludes the non-refundable costs that are considered exclusively the 1

customer’s cost responsibility. PG&E’s cost responsibility of the full 2

refundable cost comprises the allowance given to the customer. The 3

remainder of the full refundable costs after deducting the allowance, if any, 4

is considered the customer’s responsibility. 5

In addition to the PG&E share of the full refundable cost (the allowance) 6

and in the case of the non-residential customers only, PG&E could bear 7

some of the customer’s share of the full refundable costs if the customer 8

selects the 50 percent discount option in favor of the standard refundable 9

option. 10

In summary, Rule 16 should govern the calculation of MCAC because it 11

determines which customer connection costs are the responsibility of the 12

Utility enabling applicant-borne costs to be identified and excluded from 13

marginal costs used for ratemaking. PG&E continues to propose that these 14

costs be set to the Utility’s share of the installation costs under Electric 15

Rule 16, “Service Extensions.” PG&E’s costs are, therefore, net of customer 16

contributions. 17

4. Additional Model Improvements 18

After a careful review of customer contracts, which outline the new 19

connection costs for Rule 15 and Rule 16, PG&E has discovered incorrect 20

implementation of certain variables in CCBS data used in the calculation of 21

TSM costs in its 2017 GRC Phase II filing. The issues caused by the 22

misunderstanding of the data are the following: 23

For non-residential customers: analysis in 2017 GRC Phase II filing had 24

mistakenly included, in some cases, Rule 15 costs in addition to Rule 16 25

costs and had underestimated Rule 16 costs in other cases in the 26

TSM calculation. 27

For residential customers: for the residential customers that had mixed 28

contracts (residential and non-residential combined), residential portion 29

was assumed to have chosen 50 percent discount option when only 30

non-residential customers should have been considered to have chosen 31

the 50 percent discount option. As a result, such residential customers 32

were being assigned an over-estimated cost. 33

(PG&E-2)

9-10

The MCAC model has been corrected to fix these problems described 1

above. 2

C. Superiority of RECC Method in Estimating Connection Equipment Costs 3

This section discusses the appropriateness of the RECC Method, which was 4

adopted prior to the NCO methodology to calculate Marginal Connection 5

Equipment Costs, and compares it with the NCO Method to demonstrate its 6

superiority to the NCO Method. The RECC Method is presented in Section B, 7

while the NCO methodology is presented in Attachment A of this chapter. There 8

are several issues with using the NCO Method to estimate Marginal Connection 9

Equipment Costs, while PG&E finds that the RECC Method is well grounded 10

with economic theory and formulation of marginal costs estimation and is 11

superior to the NCO Method for use in revenue allocation and rate design. A 12

synoptic comparison between the RECC Method and the NCO Method is 13

presented in Table 9-3, as well as further explanation are provided subsequently 14

in this section. 15

TABLE 9-3 A SYNOPTIC COMPARISON OF RECC METHOD WITH THE NCO METHOD FOR CALCULATING

MARGINAL CONNECTION EQUIPMENT COSTS

Line No. Criterion RECC Method NCO Method 1 Adoption by CPUC

for California IOUs Was adopted for PG&E prior to 1993. In a recent development, the Commission has approved the RECC Method for PG&E’s allocation of gas distribution revenue requirement across gas customer classes (D.19-10-036). The RECC Method was adopted for development of SDG&E’s electric marginal costs (D.88-12-085). The RECC has been in use by SDG&E ever since. Although the RECC Method was never adopted for SCE, SCE has proposed the RECC Method, along with PG&E in its 2018 RDW Phase III f iling as well as its last GRC Phase II.

Last adopted in 1996 for PG&E. The NCO Method has never been adopted for development of SDG&E’s electric marginal costs. Although SCE never proposed the NCO Method, it was adopted for SCE in D.96-04-050. Currently, SCE uses an averaged value of the RECC estimate and the NCO estimate that was adopted as a non-precedential Settlement in D.18-11-027.

(PG&E-2)

9-11

TABLE 9-3 A SYNOPTIC COMPARISON OF RECC METHOD WITH THE NCO METHOD FOR CALCULATING

MARGINAL CONNECTION EQUIPMENT COSTS (CONTINUED)

Line No. Criterion RECC Method NCO Method 2 Marginal Cost Theory

and Formulation and Correctness of Estimation

Consistent with economic theory; marginal cost definition and calculation. As a result, the RECC Method provides a true marginal cost estimate by calculating incremental cost of an additional unit of connection equipment.

Not consistent with economic theory of marginal cost def inition and calculation. In fact, the NCO Method does not calculate true marginal cost; it merely calculates an average incremental revenue requirement per customer.

3 Stability of Estimate Provides a stable estimate irrespective of size of a customer class and the number of new connections during the time window of recorded data (30 months for PG&E) used for connection equipment cost estimation.

The NCO estimates can be highly volatile if the class size is small, and number of new connections is small.

4 Understating Marginal Cost Estimate

The RECC Method provides a reasonable marginal cost estimate since it calculates the costs associated with each additional unit of connection equipment. It does not use the total number of customers in the denominator, like the NCO Method does.

The NCO severely understates the marginal cost estimate by using total number of customers in a class in the denominator, thereby calculating average incremental revenue requirement responsibility per customer for the entire class. The estimate is far lower than what marginal cost should be.

5 Problem with Marginal Cost-Based Revenue Allocation

The RECC Method-based marginal cost estimate, being consistent with marginal cost theory, provides a reasonable allocation of revenue across customer classes. Hence, preserves appropriate customer access-related price signal in the revenue allocation process.

Because the NCO provides a severely understated MCAC value, this causes the revenue allocation to be inf luenced predominantly by relatively large Marginal Distribution Capacity Cost (MDCC) Revenue, with MCAC Revenue being relatively small.

1. Historical Use of the RECC Method and Recent Adoption by CPUC for 1

Gas Distribution Revenue Allocation 2

The RECC Method was the adopted method for development of PG&E’s 3

electric marginal cost prior to PG&E’s 1993 GRC Phase II when the CPUC 4

adopted the NCO Method for PG&E (in D.92-12-057, and last adopted in 5

PG&E’s 1996 GRC Phase II, in D.97-03-017). Most recently, Commission 6

(PG&E-2)

9-12

has approved using RECC Method based MCAC estimates for PG&E’s 1

allocation of gas distribution revenue requirement across gas customer 2

classes (D.19-10-036). In addition, the RECC Method was adopted to be 3

used in SDG&E’s electric marginal cost development in D.88-12-085, and 4

the RECC has been in use by SDG&E ever since. Although the NCO was 5

adopted for SCE in D.96-04-050, SCE proposed the RECC Method for its 6

2018 RDW Phase III and last GRC Phase II filings. 7

2. Results Under the RECC Method Are Consistent With Marginal Cost 8

Theory and Formulation 9

Marginal Connection Equipment Costs calculated using the RECC 10

Method are consistent with the definition of marginal costs and economic 11

theory. In economics, marginal cost is the change in the total cost that 12

arises when the quantity produced is incremented by one unit. Therefore, 13

the marginal cost (MC) of a product is measured by dividing the change in 14

the total cost (∆TC) by the change in quantity (∆Q) that causes the ∆TC.15 15

It is important to note that the denominator here is the delta 16

(i.e., incremental) quantity, and not the quantity itself. The methodology that 17

one chooses to calculate marginal cost must be based on this 18

understanding. Since under the RECC Method, the capital portion of the 19

Marginal Connection Equipment Cost is calculated by dividing the total 20

connection equipment capital costs (akin to ∆TC) of all newly added 21

customers with the number of newly added customers in that class (akin 22

to ∆Q), the RECC Method is consistent with the definition of marginal cost 23

as well as how marginal cost should be calculated. 24

Furthermore, the NCO Method does not calculate a true marginal cost. 25

To begin with, both the RECC Method and the NCO Method use the class-26

average TSM capital cost as the basis for their respective calculations. 27

15 Ref: Ch. 7, p. 166, “Principles of Economics 2e” by Steven A. Greenlaw, University of

Mary Washington, David Shapiro, Pennsylvania State University.

= ∆∆

(PG&E-2)

9-13

However, where these methodologies differ is in their mechanisms for 1

recovering the TSM capital cost. 2

As explained in Attachment A of this chapter, the NCO Method 3

calculates the Present Value of Revenue Requirement (PVRR) of the 4

equipment, including lifetime operations and maintenance (O&M) cost, and 5

multiplies the resulting value by the class connection rate (i.e., the ratio of 6

new customers to existing customers). Effectively, the NCO Method 7

calculates the total lifetime cost for all newly connected customers and 8

spreads the cost to all existing customers. The NCO Method does not 9

calculate the marginal cost of a single additional unit. Rather, it calculates 10

the incremental revenue responsibility of the class on a per-customer basis. 11

The NCO Method can be expressed mathematically as shown below, where 12

∆TC is the total lifetime cost of all new connections (PVRR plus lifetime 13

O&M), ∆Q is the number of new customers, and TQ is the total number of 14

existing customers: 15

One can see from this equation that the NCO Method does not follow 16

the marginal cost formulation described earlier in this sub-section. Although 17

PVRR and lifetime O&M is a reasonable method to determine the total 18

lifetime marginal cost of a TSM hookup, multiplying this by the class 19

connection rate (∆Q/TQ portion in the equation above) effectively changes 20

the denominator to represent the total number of customers which does not 21

result in a unit marginal cost estimate, but rather estimates the marginal per-22

customer cost responsibility of the entire class. 23

3. The New Connection Rate Used in the NCO Method Lacks Stability; 24

Results Can Be Highly Volatile When Number of Customers and New 25

Connections Are Small 26

The NCO Method can provide highly volatile estimates of Marginal 27

Connection Equipment Cost for customer classes with a small number of 28

new connections resulting from the use of the new connection rate which is 29

the ratio of new number of connections to the existing number of 30

connections. The sensitivity is greater for some customer classes compared 31

= ∆∆ * ∆

(PG&E-2)

9-14

to others, and with greater sensitivity to the new connection rate16 comes 1

greater uncertainty in the accuracy of the estimate of the class connection 2

equipment cost component of MCAC. Classes with relatively small numbers 3

of average annual billed customers tend to be more sensitive to the 4

connection rate. Inaccuracies in the estimated connection cost component 5

of MCAC drive the inaccuracies in Marginal Cost Revenue (MCR) class 6

responsibilities. Holding all else equal, it is the changes in the classes’ 7

MCAC relative to one another, that shifts associated marginal customer cost 8

responsibility between classes. 9

4. NCO Method Understates Marginal Cost, Often Severely 10

NCO Method provides an estimate of incremental revenue requirement 11

of a customer class due to a new connection; it does not calculate 12

incremental cost of a new connection, i.e., marginal cost as explained earlier 13

in this section. In other words, the denominator of this calculation is the total 14

number of customers in a class, instead of the number of new connections. 15

The RECC Method calculates marginal cost correctly since it uses the 16

number of new customers in the denominator. Because the NCO Method 17

uses the total number of customers in the denominator, which is often 18

significantly greater than the number of new connections, it ends up 19

providing an estimate that is far below what the estimate should be if an 20

economic theory-based method of computation, such as the RECC Method, 21

were used. A comparison of the NCO and RECC estimates is shown in 22

Figure 9-1. Consequently, the NCO Method estimate results in wrong 23

revenue allocation causing a subdued, inadequate price signal for the 24

connection equipment—an issue that will discussed in detail in Sub-25

Section 5 below. 26

16 The classes particularly sensitive to the NCO Method’s new connection rate are

ML&P – Primary, E-19 Secondary, Primary and Transmission, and E-20 Secondary, Primary, and Transmission. Adding a single new customer connection in the calculation of the new connection rates for these classes, increases the NCO estimate by 14 percent to 41 percent.

(PG&E-2)

9-15

FIGURE 9-1 MARGINAL CONNECTION EQUIPMENT COST ESTIMATES BY CUSTOMER CLASS

5. Under NCO, Revenue Allocation Is Dominated by Marginal Distribution 1

Capacity Cost, Inappropriately Lessening the Influence of Marginal 2

Customer Access Cost 3

To understand how NCO method-based estimate reduces the customer 4

access cost-related price signal, it is important to review how distribution 5

revenue allocation is performed. Distribution revenue allocation is 6

performed by first adding the MDCC revenue with the MCAC revenue. This 7

sum is called marginal distribution cost revenue. 8

Marginal Distribution Cost Revenue = MDCC Revenue + MCAC Revenue

The proportions of distribution MCR across PG&E’s customer classes 9

are used as the distribution percentage revenue allocators. 10

Distribution Percentage Revenue Allocator of Customer Class “i” = (Marginal Distribution Cost Revenue of Customer Class “i”) ÷ (Total Marginal Distribution Cost Revenue of All Customer Classes)

(PG&E-2)

9-16

Formulation of the Distribution Percentage Revenue Allocator 1

mentioned above shows that if the MCAC revenue is unreasonably small in 2

comparison to the MDCC revenue, then the distribution revenue allocation 3

will be driven predominantly by MDCC revenue and customer access 4

related price signal will reduce significantly. Correct estimates of MDCC as 5

well as MCAC are very important to ensure a fair revenue allocation. Since 6

the NCO Method does not provide true marginal cost estimates that are also 7

unreasonably low, it causes an unfair revenue allocation by letting MDCC 8

drive the revenue allocation with a subdued influence of MCAC. 9

Figure 9-2 below shows that when the RECC-Method based estimate is 10

used, the proportion of customer access-related MCR is 44 percent of the 11

total distribution MCR, while the percentage of customer access revenue 12

reduces to 22 percent when the NCO Method is used, which is a substantial 13

and inappropriate reduction caused by the NCO Method. PG&E has also 14

investigated the revenue allocation impact of using the NCO Method instead 15

of the RECC Method and found that certain customer classes, specifically 16

the residential class and the agricultural class, will be subject to higher 17

revenue allocation if the NCO Method is used, as shown in Figure 9-3. 18

PG&E strongly recommends the use of the RECC Method to achieve 19

equitable revenue allocation across all customer classes. 20

FIGURE 9-2 COMPARISON OF CAPACITY VERSUS CUSTOMER ACCESS REVENUE PROPORTIONS

UNDER RECC METHOD AND NCO METHOD

(PG&E-2)

9-17

FIGURE 9-3 COMPARISON OF RECC METHOD BASED REVENUE ALLOCATOR WITH NCO METHOD

BASED REVENUE ALLOCATOR

D. Introduction to Revenue Cycle Services Marginal Costs 1

This section presents the methodology and resulting estimates of RCS 2

marginal costs, which is another component of the MCAC model. The RCS 3

methodology used in this filing is the same as the one PG&E introduced in its 4

2017 GRC Phase II. That RCS methodology included significant improvements 5

over the models used in PG&E’s previous GRC Phase II filings. These 6

improvements included updating the cost driver data from 2003 to 2014, using 7

actual 2014 financial cost data for all RCS activities instead of the loaded labor 8

rate as a proxy for the RCS costs, and improving segmentation and accuracy by 9

being able to calculate marginal RCS costs at the rate schedule level. For the 10

2020 GRC Phase II, PG&E has introduced several changes to the RCS model. 11

The two main changes include: (1) basing the RCS cost estimates on the 12

average of three years of data (2015-2017), instead of one year; and13

(2) implementing separate cost estimates for the NEM and Non-NEM customer 14

groups within each customer class. The other changes and corrections made to 15

the RCS model will be discussed in Section F. 16

E. Historical Background on RCS Marginal Costs Methodology 17

Following the Electric Industry Restructuring (EIR) in the 1990s and the 18

unbundling of billing and meter services, PG&E developed activity-based costing 19

(PG&E-2)

9-18

models (RCS models) to accurately estimate the rate credits for when these 1

services were competitively provided by Electric Service Providers. Such RCS 2

models provided a very detailed analysis of the individual activities that 3

comprised a service, such as meter reading, and enabled PG&E to separate the 4

fixed cost from the marginal cost of providing that service. In addition, these 5

models allowed PG&E to analyze the savings it expected when its distribution 6

customers had services performed by entities other than PG&E. Before the 7

advent of EIR, PG&E did not have the necessary data to perform a separate 8

marginal cost analysis for these ongoing costs. Instead, PG&E estimated its 9

average cost per customer for ongoing customer services and used this average 10

cost as a proxy for the marginal cost. 11

PG&E developed its first RCS models for use in the RCS proceeding 12

Application (A.) 97-11-004 to estimate avoided costs when customers switched 13

to alternate RCS providers. In the RCS Decision (D.98-09-070), the CPUC 14

ordered that credits from these models be based on short-run avoided costs. In 15

that decision, the CPUC also ordered PG&E to compute long-run avoided costs 16

“or a reasonable proxy.”17 17

In its 2003 GRC Phase II filing, PG&E adopted the long-run approach for 18

determining the marginal ongoing cost component of MCACs. In addition, 19

PG&E updated the activities in the RCS models it had used in the RCS 20

proceeding A.97-11-004 to reflect changes in billing and payment processing 21

services and updated the labor rates used in the prior models. PG&E also 22

proposed to use a survey methodology to estimate the portion of marginal 23

ongoing customer access costs not covered by the RCS models. These costs 24

are referred to as Other Account 903 costs since they are recorded in FERC 25

Account 903. Other Account 903 costs include, for example, the costs of local 26

office and neighborhood payment center transactions. 27

In its 2014 GRC Phase II filing, PG&E significantly revised the meter 28

services and meter reading components of the RCS models due to the 29

implementation of PG&E’s SmartMeter™ Program, which resulted in a major 30

change in technology and associated costs. PG&E also updated its Other 31

Account 903 costs based on a survey of subject matter experts. 32

17 See D.98-09-070, mimeo, p. 29, Ordering Paragraph 7.

(PG&E-2)

9-19

In its 2017 GRC Phase II, PG&E introduced a new RCS model which was a 1

significant improvement over the models used in PG&E’s previous GRC Phase II 2

filings. These improvements included using longer-run marginal ongoing 3

customer access costs, updating the cost driver data from 2003 to 2014, using 4

actual 2014 financial cost data for all RCS activities instead of using the loaded 5

labor rate as a proxy for the RCS costs, and improving segmentation and 6

accuracy by being able to calculate marginal RCS costs at the rate schedule 7

level. The new RCS model produced numbers that are more granular than 8

previous RCS models, resulting in more cost-based rates. 9

Because all of PG&E’s GRC Phase II proceedings for the past 15 years 10

have been settled, there has been no express adoption of an RCS model for the 11

development of marginal costs. 12

F. Revenue Cycle Services Model 13

1. Methodology 14

The RCS model is a top-down, activities-based model that produces 15

results by rate schedule, which are then aggregated into results by customer 16

class. The activities in the RCS model reflect PG&E’s current billing and 17

payments, credit and collections, customer inquiry, meter services and 18

meter reading operations. PG&E’s current operations and organizational 19

structure are assumed to remain in place for the 2020 test year. 20

Figure 9-4 below lists the type of activities included in the RCS model. 21

Each specific activity includes an annual cost and a cost driver for the 22

activity. A cost driver value is a measure of the share of each rate schedule 23

in an activity. Furthermore, for the 2020 GRC, customers included in each 24

rate schedule are segmented into NEM customers versus Non-NEM 25

customers. This additional segmentation allows us to distinguish between 26

the activities of NEM and Non-NEM customers within a rate schedule. The 27

activities in the RCS model are sufficiently refined such that, as the cost 28

driver value increases, the annual cost for that activity increases in direct 29

proportion. The allocation of the annual costs of each activity is based on 30

how much each rate schedule’s NEM and Non-NEM customers contribute to 31

that activity. These allocated costs are then divided by the number of 32

(PG&E-2)

9-20

annual NEM and Non-NEM customers included in the rate schedule to 1

arrive at the annual cost per customer. 2

FIGURE 9-4 TYPES OF ACTIVITIES INCLUDED IN RCS MODEL

2. An Example of Allocation of RCS Costs 3

The following example demonstrates how 2015 electronic bill delivery 4

costs are calculated for customers on PG&E’s E-1 rate schedule, the 5

standard electric rate schedule for residential customers. As Figure 9-4 6

shows, the five major RCS activities are (1) account set-up; (2) billing and 7

payments; (3) credit and collections; (4) meter services; and (5) meter 8

reading. 9

Costs related to the major RCS activity types are grouped into several 10

Service, Subservices, and Subservices Type categories. Each Service, 11

Subservice, and Subservice Type category coincides with a unique cost 12

driver. For instance, Table 9-4 shows the categories related to electronic bill 13

delivery costs. Service is always one of the major types of RCS activities. 14

In this example, the Service is “Billing and Payments,” and the Subservice is 15

“Bill Delivery” and the Type is “Electronic.” Other Subservices under “Billing 16

and Payments” include “Bill Processing” and “Payment Processing.” A 17

Subservice Type under Bill Delivery also includes “Printed Bills.” The cost 18

driver for electronic bill delivery costs is the “Number of Electronic Bills 19

Delivered.” 20

(PG&E-2)

9-21

Table 9-5 illustrates the method to allocate the correct amount of costs 1

to E-1 customers. The calculation steps are labeled A through J and are 2

shown in the left-most column of Table 9-5. Using these steps, the 3

calculation details are shown in the right-most column of Table 9-5. In 2015, 4

the total cost of delivery electronic bills was $1,640,172 and the total number 5

of electronic bills delivered was a little over 34 million units. The number of 6

bills delivered to E-1 Non-NEM customers and E-1 NEM customers was 7

15,460,606 and 449,940 units, respectively. Therefore, 45.2 percent of the 8

electronic bills were delivered to the E-1 Non-NEM customers while 9

1.3 percent where delivered to E-1 NEM customers. Therefore, E-1 10

Non-NEM customers are allocated 45.2 percent of the total costs, $740,480. 11

Similarly, E-1 NEM customers have been allocated $21,550. To get the cost 12

per customer, the total costs allocated to E-1 Non-NEM and E-1 NEM 13

customers are divided by the number of customers in the Non-NEM and 14

NEM sub-group, respectively. As shown in rows I and J of Table 9-5, the 15

final per customer costs for E-1 Non-NEM and NEM customers are 16

$0.17 and $0.19, respectively. 17

TABLE 9-4 ELECTRONIC BILL DELIVERY COST IN 2015

Line No. Service Subservice Type Cost Driver Cost

1 Billing and Payments

Bill Delivery Electronic Bills Number of Electronic Bills Delivered

$1,640,172

(PG&E-2)

9-22

TABLE 9-5 YEARLY ELECTRONIC BILL DELIVERY COSTS FOR E-1 CUSTOMERS IN 2015

Line No. Steps Calculation Description

Calculation Values Formula

1 A Total Costs of Electronic Bill Delivery in 2015 $1,640,172

2 B Total Numbers of Bills Sent Electronically in 2015: 34,245,423

3 C Numbers of Bills Sent to Schedule E-1 Non-NEM Customers 15,460,606

4 D Numbers of Bills Sent to Schedule E-1 NEM Customers 449,940

5 E Electronic Bill Delivery Cost Allocated to E-1 Non-NEM Customers $740,480 (C/B) x A 6 F Electronic Bill Delivery Cost Allocated to E-1 NEM Customers $21,550 (D/B) x A 7 G Total Number of E-1 Non-NEM Customers 4,452,888

8 H Total Number of E-1 NEM Customers 112,407

9 I Electronic Bill Delivery Cost Allocated per E-1 Non-NEM Customers $0.17 E/G 10 J Electronic Bill Delivery Cost Allocated per E-1 NEM Customers $0.19 F/H

To find the total RCS costs that are allocated to E-1 customers, the 1

steps illustrated in Table 9-5 are repeated for all the other cost components 2

related to billing and payments. The method of calculation is the same for 3

the other major RCS services, such as account-up, credit and collections, 4

customer inquiry, meter reading, and meter services. 5

Table 9-6 shows a summary of unit costs of all RCS service categories 6

based on the 2015 recorded data. The upper portion of this table has 7

estimates for the E-1 Non-NEM customers and the lower portion has 8

estimates for the E-1 NEM customers. For example, in 2015, the account 9

set-up cost for E-1 Non-NEM customers was $4.63 million, and the number 10

of total accounts was 4.45 million. Therefore, the account set-up costs per 11

E-1 Non-NEM customers per year was $1.04. The total unit costs for all of 12

the services is $34.03 for E-1 Non-NEM customers and $167.18 for E-1 13

NEM customers. 14

(PG&E-2)

9-23

TABLE 9-6 2015 TOTAL RCS COSTS FOR E1 CUSTOMERS

Line No. Metrics

Account Setup

Billing and Payments

Credit and Collections

Meter Reading

Meter Services

2015 Total RCS Cost

1 E1 Non-NEM Customers 2 2015 Total RCS

Costs $4,626,437 $71,682,573 $9,017,729 $24,046,453 $42,143,015 $151,516,208

3 Total Number of Customers

4,452,888 4,452,888 4,452,888 4,452,888 4,452,888

4 2015 RCS Cost Per Customer

$1.04 $16.10 $2.03 $5.40 $9.46 $34.03

5 NEM Customers 6 2015 Total RCS

Costs $135,869 $6,338,961 $313,347 $1,321,792 $10,682,301 $18,792,270

7 Total Number of Customers

112,407 112,407 112,407 112,407 112,407

8 2015 RCS Cost Per Customer

$1.21 $56.39 $2.79 $11.76 $95.03 $167.18

3. Changes in Financial Costs Model 1

There are three categories of RCS model input data: financial costs, 2

cost drivers, and customer data. Customer data comes from CC&B and the 3

cost drivers data are provided by the PG&E departments responsible for the 4

underlying services. The financial cost data is housed in the SAP (System, 5

Application, and Products) data warehouse. The financial data is based on 6

a financial cost model that assigns and allocates costs for internal 7

management reporting. Moreover, this financial cost data can be 8

segmented by Major Work Categories (MWC), which are used to collect 9

costs associated with a complete, distinct, and ongoing work process, often 10

specific to a single Line of Business. 11

Beginning in 2016, a new financial costs model was implemented that 12

changed the labor rate from “fully loaded” to “labor only” which excludes all 13

labor benefits and payroll taxes. In addition to aligning PG&E’s cost model 14

with other utilities, the new cost model creates greater accountability for 15

costs and allows for more effective resource decisions. 16

Because of the changes in the cost model that occurred at the beginning 17

of 2016, the financial data used for the 2020 GRC includes data from both 18

the old cost model from 2015, and the new model, from 2016-2017. To 19

(PG&E-2)

9-24

reconcile these three years of data, conversion factors were created to 1

convert the 2016-2017 RCS costs back into costs based on the “fully 2

loaded” labor rate model. 3

PG&E used the Cost Model Analysis Recast Tool, a tool developed by 4

the Financial Cost Model team, to convert 2016 and 2017 cost data backed 5

to the “fully loaded” labor rate model. The Recast tool allows the user to see 6

the percentage cost difference between the old “fully loaded” labor rate 7

model and the new “labor only” labor rate model by each of the MWCs. 8

PG&E used the Recast tool to retrieve the percentages cost differences 9

between the old and the new cost models for each of the MWCs that are 10

included RCS financial costs data. These percentage differences were used 11

to create conversion factors that were applied to 2016-2017 financial cost 12

data so that these costs could be converted to the “fully loaded” labor rate 13

costs. Table 9-7 lists the MWCs related to RCS activities along with each 14

MWC’s conversion factor. 15

TABLE 9-7 MAJOR WORK CATEGORIES RELATED TO RCS ACTIVITIES

Line No. MWC MWC Description

2016-2017 Conversion Factors

1 25 Install New Electric Meters 1.148 2 74 Install New Gas Meters 1.100 3 AR Read & Investigate Meters 1.670 4 DD Provide Field Service 2.255 5 DK Manage Customer Inquiries 1.489 6 EY Change/Maint Used Elec Meter 1.773 7 FK Retain & Grow Customers 1.243 8 HY Change/Maint Used Gas Meters 2.052 9 IG Manage Var Bal Acct Processes 1.223

10 IS Bill Customers 1.163 11 IT Manage Credit 1.472 12 IU Collect Revenue 1.580 13 IV Provide Account Services 1.492 14 JV Maintain IT Apps & Infra 1.220

4. Correction in Account Set-Up Calculation 16

When writing the testimony for the 2017 GRC Phase II Application and 17

the 2018 Rate Design Window (RDW) Phase III, A.17-12-011, PG&E 18

calculated the account set-up costs per customer by dividing the account 19

set-up costs by the number of new customers. Since the account set-up 20

(PG&E-2)

9-25

costs are one-time costs for new customers, PG&E believed account set-up 1

costs should be divided by new accounts. However, PG&E calculated the 2

cost per customer costs for all the other RCS activities such as billing and 3

payment, credit and collection, meter reading, and meter services by 4

dividing the total costs by the number of ongoing customers. 5

In the 2018 RDW Phase III prepared testimony on behalf of The Utility 6

Reform Network (TURN), it was argued that PG&E made a conceptual error 7

by dividing total account set-up costs by total new customers. TURN 8

asserted that PG&E’s methodology overstated costs and the account set-up 9

costs should be divided by total number of customers instead of number of 10

new customers.18 11

After conducting a detailed comparison of PG&E’s and TURN’s 12

methods, PG&E was persuaded by TURN’s recommendation. PG&E now 13

agrees with TURN that dividing the account set-up costs by new customers 14

leads to an overstatement of the total RCS costs per customer.19 15

Therefore, for the 2020 GRC Phase II, PG&E has calculated the total 16

RCS costs per customer by dividing all sub-unit costs by total number 17

of customers. 18

5. Exclusion of Meter Replacement Costs 19

In the 2017 GRC Phase II and the 2018 RDW Phase III, meter 20

replacement costs were included in the RCS meter services costs category. 21

In TURN’s testimony, it states the meter replacement costs should be 22

removed from the RCS costs since the RECC Method assumes that the 23

meter is replaced indefinitely and that TURN’s NCO Method with 24

replacement includes the replacements as part of the capital costs.20 In 25

PG&E’s RDW 2018 rebuttal,21 PG&E concludes that meter replacement 26

costs should not be included in the RCS cost for the RECC or NCO 27

Methods. TURN’s assumptions are correct about the RECC Method, which 28

18 See 2018 RDW Phase III, Exhibit TURN-016, A.17-12-011, p. 24. 19 See Exhibit (PGE-19), RDW 2018 Rebuttal Testimony Fixed Charge Proposal in

Phase III. A.17-12-001, pp. 2B-15 and 2B-16. 20 See 2018 RDW Phase III, Exhibit TURN-016, A.17-12-011, p. 25. 21 See Exhibit (PG&E-19), RDW 2018 Rebuttal Testimony Fixed Charge Proposal in

Phase III, A.17-12-001, pp. 2B-16 and 2B-17.

(PG&E-2)

9-26

uses the RECC factor that assumes perpetual replacement of assets. 1

PG&E’s proposal to exclude connection equipment replacement costs from 2

the NCO would represent a return to the methodology the CPUC adopted in 3

D.92-12-057, which excluded connection equipment replacement costs from 4

the MCAC calculation. Therefore, in this proceeding, PG&E proposes to 5

exclude meter replacement costs from marginal RCS cost estimation. 6

6. Streetlight Changes 7

In the 2017 GRC Phase II, PG&E included rate Schedule OL-1 (Outdoor 8

Area Lighting Service) in the Medium Light and Power customer class. 9

However, to be more aligned with Rate Design models, Schedule OL-1 has 10

been moved to the Streetlight customer class. Therefore, the 2020 GRC 11

RCS results for the Streetlight customer class includes costs for four rate 12

schedules: LS-1, LS-2, LS-3 and OL-1. Beginning with the 2020 GRC, the 13

Meter Reading and Meter Services costs will no longer be allocated to 14

Schedules LS-1, LS-2 or OL-1, as service on these rate schedules does not 15

require a meter. Since only metered street lights are billed under the LS-3 16

rate schedule, LS-3 is the only Streetlight rate schedule allocated any Meter 17

Reading and Meter Services costs. 18

G. 2020 GRC RCS Results Segmented by NEM and Non-NEM Customers 19

Table 9-8 shows the average annual RCS costs per customer for the years 20

2015-2017 for all the major class levels in 2021 dollars. The RCS costs shown 21

in Table 9-8 are broken out by NEM versus Non-NEM customers. Table 9-8 22

shows that the RCS costs per customer is greater for NEM customers in 23

comparison to the Non-NEM customers across most of the customer levels. 24

These cost differences are greater for non-residential customers than for 25

residential customers. The cost differences between NEM and Non-NEM 26

customers are mainly due to: 27

Both residential and non-residential NEM customers having more complex 28

billing issues and longer work times when performing other meter services 29

such as turn-on/turn-offs, meter maintenance and meter testing; and 30

Non-residential NEM customers having longer call times when calling into 31

Call Centers and a larger percentage of manual meter reads. 32

(PG&E-2)

9-27

The next section will clarify some of the above points by looking in detail at the 1

underlying drivers and activities that cause differences in costs for residential 2

NEM customers as compared with Non-NEM customers. 3

TABLE 9-8 AVERAGE ANNUAL RCS COSTS PER CUSTOMER ($2021): 2015 – 2017

Line No. Customer Class

Account Setup

Billing and Payments

Credit and Collections

Meter Reading

Meter Services

Total Yearly Costs

1 A-10 Medium L & P $2.88 $55.99 $8.12 $19.58 $68.74 $155.32 2 Non-NEM $2.86 $51.15 $8.07 $18.95 $65.11 $146.14 3 NEM $4.38 $338.04 $11.03 $56.52 $280.81 $690.78 4 Residential $1.07 $15.43 $2.09 $4.85 $11.30 $34.76 5 Non-NEM $1.08 $14.36 $2.10 $4.72 $8.74 $31.00 6 NEM $0.82 $35.08 $2.04 $7.26 $58.32 $103.53 7 Small L&P 3P $1.46 $18.77 $5.73 $8.42 $25.77 $60.14 8 Non-NEM $1.43 $11.86 $5.70 $7.82 $21.92 $48.73 9 NEM $2.80 $328.64 $7.33 $35.21 $198.47 $572.46 10 Small L&P 1P $1.02 $13.46 $2.41 $9.13 $18.46 $44.48 11 Non-NEM $1.02 $10.50 $2.40 $9.00 $17.82 $40.74 12 NEM $1.60 $320.75 $3.64 $22.17 $84.26 $432.42 13 Agricultural A $0.86 $19.49 $3.79 $13.72 $55.85 $93.72 14 Non-NEM $0.85 $10.90 $3.76 $13.24 $52.93 $81.69 15 NEM $1.38 $336.13 $4.78 $31.36 $163.46 $537.12 16 Agricultural B – Large $3.13 $37.83 $7.08 $37.70 $133.47 $219.21 17 Non-NEM $3.08 $27.33 $7.01 $36.00 $125.83 $199.25 18 NEM $4.64 $384.34 $9.31 $93.98 $385.50 $877.77 19 Agricultural B – Small $1.67 $34.43 $4.39 $30.03 $116.36 $186.88 20 Non-NEM $1.37 $12.52 $4.14 $26.76 $98.68 $143.46 21 NEM $6.08 $355.85 $8.11 $77.98 $375.75 $823.78 22 E-19/E-20 $23.78 $544.32 $54.92 $156.86 $571.34 $1,351.22 23 Non-NEM $24.96 $554.59 $57.76 $154.66 $539.66 $1,331.63 24 NEM $15.55 $472.81 $35.18 $172.15 $791.89 $1,487.59 25 Streetlights $0.86 $9.01 $1.81 $0.27 $0.38 $12.33 26 Traf f ic Control $0.51 $5.71 $1.21 $9.26 $14.54 $31.22

1. Details of Residential Customers’ RCS Marginal Costs 4

Table 9-8 shows Residential RCS costs per customer broken down by 5

the major RCS categories, account setup, billing and payments, credit and 6

collections, meter services and meter reading. The category with the 7

biggest differences in cost is Meter Services. A NEM customer’s meter 8

(PG&E-2)

9-28

services cost is more than six times the cost of Non-NEM customer. The 1

categories with the next two biggest cost differences are Billing and 2

Payments and Meter Reading. The category with the smallest difference is 3

Credit and Collections. 4

a. Meter Services 5

The types of meter services provided to residential customers are 6

listed in Table 9-9. Each type of meter service activity is described in 7

terms of a Service, a Subservice, and a Type. Each Service, 8

Subservice, and Type grouping has its own costs, which represents a 9

portion of the total meter services costs. For instance, as Table 9-9 10

shows, the costs related to Other Meter Services represents $14 of the 11

total $58 of a NEM customer’s meter services costs. Each Service, 12

Subservice, and Type grouping also has its own cost driver. The 13

two meter services that are driving the costs differences between NEM 14

and Non-NEM customers are turn-ons/turn-offs not related to non-15

payments and Other Meter Services. The cost driver for turn-ons/ 16

turn-offs is the labor time. The cost driver for Other Meter Services is 17

time spent performing these services, which includes a variety of field 18

meter services to ensure accurate billing, such as field test, information 19

verification, maintenance replacement, and meter removal. 20

It takes an average of 0.80 minutes to complete a turn-ons/turn-offs 21

service for a Non-NEM customer versus 13.1 minutes for a 22

NEM customer. Therefore, meter services costs are higher for NEM 23

customers. Likewise, it takes 1.2 minutes on average to complete a job 24

related to Other Meter Services for a Non-NEM versus 7.7 minutes for a 25

NEM customer.22 26

22 The cost driver data from 2015 was used to calculate the cost driver statistics shown in

Tables 9-10 and 9-11.

(PG&E-2)

9-29

TABLE 9-9 TYPE OF RESIDENTIAL CLASS METER SERVICES

Line No. Service Subservice Type Non-NEM NEM 1

Customer Inquiry Call Center

Meter Services $2.18 $2.52

2 Local Office $0.12 $0.04 3

Meter Services Field

Turn-ons/Turn-offs, Not SONP/RLNP $3.60 $41.28

4 Other Meter Services $2.65 $14.36 5 SONP and RLNP $0.18 $0.12

6 Grand Total $8.74 $58.32

TABLE 9-10 METER SERVICES COST DRIVER STATISTICS

Line No. Type of Meter Services Cost Drive Percentages Non-NEM NEM

1 Turn-ons/Turn-offs, Not SONP/RLNP

Average Time Spent Performing Service (minutes) 0.80 13.07

2 Other Meter Services Average Time Spent Performing Service (minutes) 1.16 7.72

2. Billing and Payments 1

There are several service activities that are included in the major Billing 2

and Payments category. However, as shown in Table 9-11, it is the cost of 3

servicing Complex Billing that drives the differences in NEM and Non-NEM 4

costs. As previously stated, Complex Billing is a billing service provided by 5

the CC&B Department that is responsible for billing customers that cannot 6

be billed in the standard way through the CC&B system due to the sheer 7

size and/or complexity of their billing contracts. Only 1 percent of the total 8

Complex Billing customers are Non-NEM Residential customers, while 9

55 percent of the total Complex Billing customers are NEM Residential 10

customers. 11

(PG&E-2)

9-30

TABLE 9-11 RESIDENTIAL CLASS BILLING AND PAYMENT SERVICES

Line No. Service Subservice Type Non-NEM NEM 1

Billing and Payments

Bill Delivery Electronic Bills $0.21 $0.25 2 Printed Bills $3.52 $3.12 3 Bill Processing Records and Billing Accuracy $2.00 $1.90 4

Complex Billing

General $0.03 $14.86 5 NEM $0.00 $7.36 6 Demand Response, Interval, and

Special Projects $0.05 $0.01

7

Customer Inquiry Assistance

Electronic Payment $0.28 $0.28 8 General $0.20 $0.16 9 Kiosk $0.02 $0.01

10 Neighborhood Payment Center $0.02 $0.00 11 US Mail $0.17 $0.12 12

Payment Processing

Electronic Payment $0.51 $0.54 13 Kiosk $0.08 $0.04 14 Local Office $1.93 $0.77 15 Neighborhood Payment Center $0.34 $0.11 16 US Mail $0.33 $0.24 17 Customer

Inquiry Call Center Billing and Payments $4.41 $5.24

18 Local Office Billing and Payments $0.26 $0.08 19 Grand Total $14.36 $35.08 20 Service Cost Driver Statistics Non-NEM NEM

21 Complex Billing Percentage of Complex Billing Issues 1.29% 54.77%

Although this section only investigated the drivers responsible for the 1

cost differences between residential NEM and Non-NEM customers, it can 2

also be shown that these are some of the same drivers responsible for the 3

cost differences between NEM and Non-NEM non-residential customers. 4

H. Conclusion 5

In this chapter, PG&E has presented its proposed 2020 GRC Phase II 6

methodologies and the resulting estimates for Marginal Connection Equipment 7

costs and the Revenue Cycle Services costs, the two components of the 8

Marginal Access Customer Cost model. In order to ensure more accurate 9

estimates, PG&E has corrected some errors in both the New Connection and 10

RCS methodologies. In addition, PG&E proposes the RECC Method over the 11

NCO Method for estimating the Marginal Connection Equipment costs and 12

describes the flaws in the NCO Method. The RCS costs are now based on the 13

(PG&E-2)

9-31

average of data from years 2015-2017 instead of one year of data. Also, certain 1

cost adjustments have been made due to changes in the financial costs model. 2

For each of the class levels, separate marginal RCS costs have been calculated 3

for NEM and Non-NEM customers. As a result, PG&E is now able to show that 4

there are significant marginal cost differences between NEM and Non-NEM 5

customers across all customer classes. 6

PG&E respectfully requests that the Commission approve its proposed, 7

improved and more granular MCAC estimates along with the methodologies 8

presented in this chapter. PG&E also requests that the Commission approve 9

the underlying conceptual methodology for calculating NEM versus Non-NEM 10

MCACs, so that that methodology is available for use in future proceedings. 11

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 9

ATTACHMENT A

NEW CUSTOMER ONLY METHOD

(PG&E-2)

9-AtchA-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 9

ATTACHMENT A NEW CUSTOMER ONLY METHOD

TABLE OF CONTENTS

A. Lifetime O&M for New Connections ..................................................................... 2

B. Determining New Connection Rate Used in NCO Method .................................. 4

(PG&E-2)

9-AtchA-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 9 2

ATTACHMENT A 3

NEW CUSTOMER ONLY METHOD 4

Although Pacific Gas and Electric Company (PG&E or the Company) is 5

requesting that the California Public Utilities Commission adopt the RECC Method 6

for PG&E, PG&E is also presenting information here on what its Marginal Customer 7

Access Cost (MCAC) estimates would have been under the previously-adopted New 8

Customer Only (NCO) method. PG&E presents this data purely for informational 9

purposes, to allow the parties to compare updated NCO costs to those in PG&E’s 10

last General Rate Case (GRC) Phase II showing. PG&E’s proposal presented in 11

Chapter 9 is for adoption of the RECC Method. 12

There are four steps for the NCO MCAC calculation of the marginal connection 13

equipment cost component, with a final step to arrive at the total MCAC estimate: 14

First, the per-customer access equipment costs for a new connection are 15

determined for each customer class. These class-level new connection costs are 16

identical to those used for the RECC Method MCAC estimates. 17

Second, the per-new customer access equipment costs are converted to a 18

Present Value of Revenue Requirement (PVRR) using a PVRR factor and other 19

appropriate loaders and financial factors, resulting in the marginal cost responsibility 20

per new customer in each class.1 21

Third, factors for lifetime operations and maintenance (O&M), stated as 22

percentages of the respective equipment capital cost, are applied to the respective 23

connection equipment. The results of this third step are values representing the 24

present value of O&M and other appropriate loaders for maintaining transformers 25

and services over their lives.2 26

Fourth, for each class, the sum of marginal cost responsibility for new 27

connection equipment and corresponding lifetime O&M for transformers and 28

1 See Exhibit (PG&E-2), Chapter 10 for discussion about the development of the PVRR

factor and loaders. 2 See Section D below for additional background on lifetime O&M.

(PG&E-2)

9-AtchA-2

services3 is converted to the per-customer class cost responsibility for new 1

connection equipment and corresponding lifetime O&M by multiplying the per-new 2

customer marginal cost responsibility by the class’s new connection rate. The 3

customer new connection rate is simply the ratio of the number of new customer 4

connections in any given year to the average number of customers4 for the 5

preceding year. The result of this fourth step effectively is to produce a weighted 6

average marginal connection equipment cost, including O&M and other appropriate 7

loaders, per customer in the class using the new connection rate as a weighting 8

factor applied to the sum of marginal connection equipment costs and corresponding 9

lifetime O&M for new customers and assuming $0 marginal connection equipment 10

costs for existing customers. 11

Finally, for the total MCAC estimate using the NCO method for the marginal 12

connection equipment cost component, for each class, the ongoing customer access 13

(RCS) costs estimated for the class is added to the class per-customer cost 14

responsibility for new connection equipment and corresponding O&M. This total by 15

class of new connection equipment and corresponding O&M and ongoing cost 16

responsibility is the NCO method’s total MCAC result by class as shown in 17

Table 9A-1. 18

MCACNCO = (Customer New Connection Rate NCO Marginal Connection Equipment Cost)

+ Marginal RCS Cost

A. Lifetime O&M for New Connections 19

Under the NCO method, the cost imposed on the system for access 20

equipment associated with new customers at the margin should be only 21

the costs of the access equipment—final line transformers5 as applicable, 22

secondary as applicable, services and meters—installed to connect the new 23

3 A lifetime O&M factor, with other appropriate loadings, is not applied to meters; the

ongoing cost for maintaining and servicing meters is captured in the marginal ongoing Revenue Cycle Services (RCS) costs.

4 The number of “customers” is the number of “billed accounts;” the text continues to refer to the number of “customers” instead of “billed accounts” to be consistent with past terminology.

5 As distinct from distribution transformers, which are included in demand-related marginal distribution capacity costs.

(PG&E-2)

9-AtchA-3

customers to PG&E’s distribution system and any related lifetime maintenance 1

and other costs directly related to the added access equipment.6 Once a new 2

connection is established, it imposes on the system the obligation to maintain 3

that equipment going forward. For existing connections, the number of 4

connections having customer turnover and temporary vacancies does not affect 5

the obligation to maintain the access equipment at the margin.7 It is the addition 6

of new access equipment for new connections that imposes an incremental 7

obligation. To capture these incremental maintenance cost obligations, PG&E 8

applies of lifetime O&M adders to final line transformers and services.8 (This 9

lifetime O&M methodology is not applied to meters since meter maintenance 10

costs are captured in the RCS models.) 11

PG&E first used the application of lifetime O&M cost adders in the 12

Company’s 2003 GRC and again in its 2007, 2011, 2014 and 2017 GRC 13

6 This Lifetime O&M methodology is applicable only to the NCO methodology where

the connection equipment costs under the NCO methodology for new customers are captured as the present values of revenue requirements for the new connection equipment capital investments. Therefore, the O&M costs and other relevant loadings are similarly captured as the present value of those costs for the lifetimes of the respective connection equipment. Under the RECC Method, these O&M costs and other relevant loadings are captured as part of the annualized RECC Method connection equipment costs via loadings to the RECC factors that are applied to the connection equipment capital investments.

7 Replacement costs of customer access equipment were not included in the marginal costs adopted in PG&E’s 1993 GRC Phase II Decision (D.) 92-12-057 but were included in the marginal costs adopted for limited purposes in PG&E’s 1996 GRC Phase II D.97-03-017. PG&E proposed inclusion of replacement costs in its 1999, 2003, 2007 and, initially, 2011 GRC Phase II showings. In PG&E’s 2014 and 2017 GRC Phase II showings, PG&E excluded replacement costs from its NCO-based marginal cost estimate of hooking-up a new customers and continues to do so for this GRC Phase II filing.

8 The term “lifetime O&M adder” is used for simplicity. This adder also includes Administrative and General, common and general plant costs, and other capital-related costs arising from the added transformers and service for new connections over the equipment lives. Incorporating O&M costs by way of lifetime O&M adders differs from the prior 2003 and 2007 proposals. In those proposals, PG&E allocated an estimate of marginal O&M to access equipment for all customer connections, existing and new. Here, PG&E only uses O&M costs for new connections. The number of new connections affects O&M costs. Customer turnover and temporary vacancies has little impact on O&M for existing connections: the number of connections having customer turnover and temporary vacancies does not change, at the margin, the O&M costs for those connections. It is new connections that, at the margin, add O&M costs.

(PG&E-2)

9-AtchA-4

proceedings.9 The application of lifetime O&M adders to access equipment 1

installed for new connections captures the present value of the ongoing 2

maintenance cost obligations that those new connections impose on the system. 3

PG&E uses lifetime O&M adders for two of three primary access equipment 4

asset types: transformers and services. The lifetime O&M adder for each of 5

these equipment asset types is calculated as follows: 6

1) The O&M expense is calculated for a unit of capital for each year of the 7

asset’s life, based on the relation of annual O&M to capital charge using 8

loaders and financial factors; 9

2) Each year of the expenses per unit of capital in (1) is discounted to the 10

present by application of a discount factor and summed to total present 11

value of expenses per unit of capital; 12

3) Additional capital costs are captured through a revenue requirement scalar 13

per unit of capital, where the revenue requirement scalar represents the 14

levelized annual revenue requirement factor for a unit of invested capital 15

plus associated return and taxes; and 16

4) The present value of expenses per unit of capital in (2) is added to the 17

additional capital costs per unit of capital in (3) resulting in the lifetime 18

O&M adder per unit of capital, expressed as a percentage.10 19

The mean new access equipment costs for transformers and services are 20

multiplied by the respective lifetime O&M adders for each customer class. The 21

results are the present value of the annual revenue requirement for lifetime 22

O&M costs imposed by each new connection in each class for these two types 23

of access equipment. 24

B. Determining New Connection Rate Used in NCO Method 25

For the determination of the number of new connections, PG&E continues to 26

retain an important revision from PG&E’s 2011 GRC Phase II. Starting with 27

PG&E’s 2011 GRC Phase II Application and continuing in its 2014 and 2017 28

showings as well as in this application, PG&E revised its method for determining 29

9 In PG&E’s 2003 and 2007 GRCs, proposed lifetime O&M adders were applicable to

certain primary and secondary distribution equipment that was outside of the customer access equipment boundary at the final line transformer used for the purpose of defining MCAC and Marginal Distribution Capacity Cost in this GRC Phase II filing.

10 See the workpapers supporting this chapter for details of the calculation.

(PG&E-2)

9-AtchA-5

the number of new customers (actual new customer connection count), which is 1

the numerator in the customer new connection rate. In GRCs prior to 2011, the 2

number of new customer connections had been estimated as the difference in 3

the year-end customer counts in two successive years as a proxy for new 4

connections.11 Using the difference in customers as a proxy for new customer 5

connections can be problematic, especially for customer classes that show net 6

declines in the number of customers. This proxy fails to capture the true rate of 7

new connections occurring because the proxy is the net of new connections and 8

disconnections. Even with net declining customers in a class due to 9

disconnections, new connections do occur, and the class needs to cover its 10

cost of those new connections by recognizing new connections in isolation, 11

rather than using new connections net of disconnections. For this 2020 GRC, 12

PG&E uses new connection forecasts by customer class to calculate new 13

connection rates instead of the proxy calculation using net changes in number of 14

customers. PG&E believes this improved method provides more accurate 15

estimates of customer new connection rates than did the pre-2011 proxy 16

method. 17

11 D.92-12-057, which first adopted the NCO method for PG&E, called for use of

three years of recorded customer data to estimate the customer new connection rate.

(PG&E-2)

9-AtchA-6

TABLE 9A-1 SUMMARY OF TEST YEAR 2021 MCAC ESTIMATES (NCO METHOD)

(DOLLARS PER CUSTOMER-YEAR)

Line No. Customer Class

Sub-Class or Rate Schedule

MCAC Values (NCO Method-Based)

Combined(a) NEM Non-NEM

1 Residential $53.58 $123.12 $49.78 2 Agricultural Ag A $262.88 $711.20 $250.71 3 Ag B – Small $1,577.98 $2,221.94 $1,534.08 4 Ag B – Large $2,199.86 $2,865.73 $2,179.69 5 Small Commercial Single Phase $166.56 $558.79 $162.77 6 Poly Phase $558.19 $1,076.19 $546.65 7 Medium Commercial A-10-S/E-19VS $502.22 $1,043.62 $492.94 8 A-10-P/E-19VP $1,263.40 $1,804.80 $1,254.12 9 Large L&P E-19-S $1,782.05 $1,919.93 $1,762.24 10 E-19-P $2,047.20 $2,185.08 $2,027.39 11 E-19-T $2,047.20 $2,185.08 $2,027.39 12 E-20-S $2,362.71 $2,500.60 $2,342.91 13 E-20-P $2,047.20 $2,185.08 $2,027.39 14 E-20-T $2,047.20 $2,185.08 $2,027.39 15 Streetlights LS1 $37.00 - $37.00 16 LS2 $16.90 - $16.90 17 LS3 $36.01 - $36.01 18 OL1 $14.76 - $14.76 19 Traffic Control $155.52 - $155.52

_______________

(a) Note that Marginal Connection Equipment Cost component of MCAC is same for both NEM and Non-NEM. Marginal RCS costs for the combined scenario are calculated as average costs obtained from NEM and Non-NEM combined data, without distinguishing if an observation is NEM or not.

(PG&E-2)

9-AtchA-7

TABLE 9A-2 TEST YEAR 2021 MARGINAL CONNECTION EQUIPMENT COST ESTIMATES (NCO METHOD)

(DOLLARS PER CUSTOMER-YEAR)

Line No. Customer Class

Sub-Class or Rate Schedule

Marginal Connection Equipment

Cost

1 Residential $18.44 2 Agricultural Ag A $168.12 3 Ag B – Small $1,389.02 4 Ag B – Large $1,978.22 5 Small Commercial Single Phase $121.58 6 Poly Phase $497.38 7 Medium Commercial A-10-S/E-19VS $345.18 8 A-10-P/E-19VP $1,106.36 9 Large L&P E-19-S $415.84 10 E-19-P $680.99 11 E-19-T $680.99 12 E-20-S $996.51 13 E-20-P $680.99 14 E-20-T $680.99 15 Streetlights LS1 $23.66 16 LS2 $9.08 17 LS3 $22.92 18 OL1 $0.00 19 Traffic Control $123.96

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 10

MARGINAL COST LOADERS AND FINANCIAL FACTORS

(PG&E-2)

10-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 10

MARGINAL COST LOADERS AND FINANCIAL FACTORS

TABLE OF CONTENTS

A. Introduction ..................................................................................................... 10-1

B. Methods .......................................................................................................... 10-1

1. Financial Factors ...................................................................................... 10-2

a. Real Economic Carrying Charge ....................................................... 10-2

b. Present Value of Revenue Requirement Factor ................................ 10-3

2. Loading Factors ....................................................................................... 10-3

a. Operations and Maintenance Loading Factor .................................... 10-3

b. A&G Expenses Loading Factor ......................................................... 10-4

c. General Plant Loading Factor ............................................................ 10-5

d. M&S Loading Factor .......................................................................... 10-5

e. Cash Working Capital Loading Factor ............................................... 10-6

f. Franchise Fees and Uncollectible Expenses Loading Factor ............ 10-6

g. Loaders Model Changes .................................................................... 10-7

C. Results ........................................................................................................... 10-7

D. Conclusion ...................................................................................................... 10-8

(PG&E-2)

10-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 10 2

MARGINAL COST LOADERS AND FINANCIAL FACTORS 3

A. Introduction 4

Pacific Gas and Electric Company (PG&E or the Company) incurs capital 5

investments and expenses to meet customer and demand growth as the number 6

of customers and kilowatt demand grows over time. Growth-related capital 7

investments and Operations and Maintenance (O&M) expenses are the major 8

components of PG&E’s marginal costs. PG&E also incurs secondary expenses 9

that are driven by capital investments and O&M expenses. This chapter 10

discusses the loading factors and financial factors that are used to include 11

expenses and financial costs in PG&E’s marginal cost calculations. 12

PG&E requests that the California Public Utilities Commission (CPUC or 13

Commission) find reasonable PG&E’s proposed financial factors and loading 14

factors, and the methodologies for calculating them. These must be included in 15

PG&E’s marginal cost calculations in order to obtain an accurate determination 16

of marginal cost. The results from PG&E’s financial factors model and loaders 17

model are summarized in Table 10-1 and Table 10-2, respectively, at the end of 18

this chapter. 19

The remainder of this chapter is organized as follows: 20

Section B – Methods 21

Section C – Results 22

Section D – Conclusion 23

B. Methods 24

Marginal transmission, distribution and customer access costs have two 25

components: capital and expense. 26

Calculation of the capital portions of the marginal costs requires use of 27

financial factors. The financial factors used by PG&E in its marginal costs 28

calculations are: (1) Present Value of Revenue Requirement (PVRR) factors; 29

and (2) Real Economic Carrying Charge (RECC) financial factors. For each 30

dollar of incremental capital in transmission and distribution capacity or customer 31

(PG&E-2)

10-2

connection equipment,1 PG&E incurs incremental O&M expense and 1

Administrative and General (A&G) expense. These incremental O&M and A&G 2

expenses in turn require incremental general plant, Materials and Supplies 3

(M&S), and Cash Working Capital (CWC). 4

Incremental O&M and A&G expenses, as well as the other incremental 5

expenses to support the O&M and A&G, are captured in the marginal cost 6

loading factors. Franchise Fees and Uncollectibles (FF&U) loading factor is 7

taken directly from PG&E’s 2020 General Rate Case (GRC) Phase I Application 8

(Application (A.) 18-12-009) and is applied to all incremental expense loadings. 9

The methodologies for each of the financial factors and loading factors 10

proposed by PG&E for use in this proceeding are described below. 11

1. Financial Factors 12

a. Real Economic Carrying Charge 13

The RECC factor represents a first-year levelized revenue 14

requirement in constant dollars over the service life of an investment 15

adjusted for inflation and discounted at PG&E’s current after-tax 16

weighted average cost of capital. Multiplying a marginal capital 17

investment by an RECC factor results in the first-year revenue 18

requirement for that investment. Using PG&E’s financial factors model, 19

PG&E calculates RECC factors for a variety of specific capital asset 20

classes. Additionally, results for selected asset classes are weighted to 21

develop composite RECC factors for major asset categories, such as 22

electric distribution, common plant, and general plant. These RECC 23

factors are used in PG&E’s Marginal Transmission Capacity Cost 24

(MTCC) calculations, Marginal Distribution Capacity Cost (MDCC) 25

calculations, and in the RECC-based calculations for the marginal 26

connection equipment cost component of Marginal Customer Access 27

Costs (MCAC). 28

1 In contrast to growth in demand for transmission, distribution and customer access,

which require PG&E-owned capital investments, generation demand growth is assumed to be met from an external marketplace. Thus, it is assumed that PG&E will not incur marginal A&G costs or other loaders for generation, beyond levels included in the market price of generation capacity and energy.

(PG&E-2)

10-3

The results of PG&E’s RECC calculations are summarized in 1

Table 10-1. 2

b. Present Value of Revenue Requirement Factor 3

The PVRR factor, when applied to a capital investment, yields a 4

value that represents the present value of annual revenue requirements 5

for the capital-related charges (depreciation expense less net salvage 6

cost), income taxes, and property insurance discounted at PG&E’s 7

current cost of capital to earn the rate of return over the service life of a 8

plant addition. PG&E calculates PVRR factors for the same asset 9

classes and weighted composites of major asset classes as calculated 10

for RECC factors. Using PG&E’s financial factors model, PVRR factors 11

are calculated as the PVRR for an assumed level of up-front capital 12

investment for each asset class that is divided by that up-front capital 13

investment amount, resulting in unitless PVRR factors. These PVRR 14

factors are used in the New Customer Only (NCO) calculation, one of 15

two methods, the CPUC has adopted for calculating the marginal 16

connection equipment cost component of MCAC, PG&E believes 17

that the RECC is the more accurate and appropriate of these two 18

methods, and therefore recommends the RECC Method over the NCO 19

method, but nonetheless includes the results of the NCO method in 20

Attachment A to Chapter 9, for purely informational purposes, to allow 21

the parties to compare updated NCO-based costs to those presented in 22

PG&E’s last GRC Phase II showing. 23

The results of PG&E’s PVRR factor calculations are summarized in 24

Table 10-1. 25

2. Loading Factors 26

a. Operations and Maintenance Loading Factor 27

When PG&E adds to its transmission and distribution facilities to 28

meet forecast increases in demand, PG&E must incur additional O&M 29

expense to operate and maintain the expanded lines and substation 30

facilities. Likewise, when PG&E connects additional customers, 31

PG&E must incur additional O&M expense to operate and maintain the 32

customer connection equipment (final line transformers, secondary 33

(PG&E-2)

10-4

conductors, services, and meters). Accordingly, PG&E calculates O&M 1

loading factors for MTCC and MDCC calculations, and for the 2

connection equipment component in the MCAC calculations. Annual 3

O&M expenses were estimated by relating average annual O&M from 4

2015 to 2017 to average annual plant balances for 2015 to 2017. 5

For marginal costs, PG&E calculates O&M loading factors 6

separately for different types of assets and differentiates O&M between 7

overhead and underground capital equipment where appropriate. 8

Expenses that are: (1) not related to transmission equipment, 9

distribution equipment, services, and meters; (2) not marginal; or 10

(3) directly paid by customers are removed from these calculations. 11

For example, vegetation management (i.e., tree-trimming) expenses that 12

are not related to service conductors are removed from the overhead 13

line maintenance expenses. The data for these calculations are from 14

PG&E’s FERC Form 1 filings for 2015-2017 and from PG&E’s 15

2020 GRC Phase I (A.18-12-009). 16

PG&E’s proposed loading factors for all distribution and 17

transmission O&M categories are summarized in Table 10-2. 18

b. A&G Expenses Loading Factor 19

The A&G loading factor reflects the increase in labor-related A&G 20

expenses and payroll taxes associated with increased O&M expenses. 21

This loading factor accounts for the costs of payroll taxes, pensions and 22

benefits, workers compensation, and liability insurance. Specifically, the 23

A&G loading factor is the ratio of 2020 Test Year marginal A&G 24

expenses and payroll taxes from PG&E’s 2020 GRC Phase I to 25

2020 Test Year total O&M expenses (including plant, customer services, 26

and customer accounts O&M) from PG&E’s 2020 GRC, Phase I. PG&E 27

proposes an A&G loading factor equal to 31.63 percent for distribution 28

and 17.85 percent for transmission, as shown in Table 10-2, below. 29

(PG&E-2)

10-5

c. General Plant Loading Factor 1

The General Plant2 Loading Factor (GPLF) accounts for the 2

additional common and general plant required to support incremental 3

O&M and marginal A&G expense. Common and general plant includes 4

such items as structures and improvements, office furniture and 5

equipment, computers and software, vehicles, telecommunications 6

equipment and data systems, and other equipment. 7

The GPLF reflects the ratio of the incremental annual carrying costs 8

for general and common plant in service assigned to distribution O&M 9

expenses and marginal A&G expenses. RECC factors for common and 10

general plant asset categories, as discussed above, are applied to the 11

respective plant in service balances to derive weighted total annualized 12

cost of common and general plant. PG&E proposes a GPLF loader 13

equal to 6.03 percent for distribution and 14.59 percent for transmission, 14

as shown in Table 10-2. 15

d. M&S Loading Factor 16

The M&S loading factor accounts for the cost to keep M&S in stock 17

for use in daily field operations as well as prepayments for insurance.3 18

M&S as well as prepayments are included in this calculation as they 19

both relate to O&M and marginal A&G expenses. Prepayments are 20

removed from working cash to develop the CWC loading factor, 21

discussed below, to prevent double counting because prepayments are 22

included here. 23

This M&S loading factor is the ratio of M&S and prepayments 24

expenses to distribution O&M and marginal A&G expenses that is 25

2 In accordance with past marginal cost practice, and for simplicity, the term “General

Plant” is used here to refer to both common plant and general plant. Common plant describes plant that is used for both gas and electric service, such as communications equipment, buildings, vehicles, etc. General Plant can be specific for one service such as gas or electric. However, it cannot be (or is not) specific to the various functions within a service (e.g., distribution, transmission, or generation). For example, some equipment may be identified as used for electric service, but not identified with a particular function. Because it was not identified as associated with a particular function, that equipment was assigned to electric general plant.

3 In accordance with past marginal cost practice, and for simplicity, the term “Materials and Supplies” is used here to refer to M&S, CWC and prepayments for insurance.

(PG&E-2)

10-6

annualized to a levelized revenue requirement. This approach, first 1

proposed in PG&E’s 2014 GRC Phase II, is a refinement of PG&E’s 2

prior methodology by calculating the loading factor based on distribution 3

O&M and marginal A&G expenses rather than total rate base less 4

working capital. This refinement results in a loading factor that is 5

calculated on a basis that is consistent with the amounts to which it is 6

applied. PG&E calculated the M&S loading factor using data obtained 7

from PG&E’s 2020 GRC Phase I testimony and workpapers. PG&E 8

proposes an M&S loading factor of 0.88 percent, as shown in 9

Table 10-2, below. 10

e. Cash Working Capital Loading Factor 11

The CWC loading factor accounts for the requirement that PG&E 12

must keep funds available to meet the costs of daily operations. 13

These funds, referred to as CWC, provide funds required for day-to-day 14

operations as well as funds used to pay operating expenses in advance 15

of receiving customer payments. 16

The CWC loading factor reflects the ratio of working cash related to 17

electric distribution (with prepayments for insurance removed) to electric 18

distribution O&M and marginal A&G expenses that is then annualized to 19

a levelized annual revenue requirement for these costs. This approach, 20

first proposed in PG&E’s 2014 GRC Phase II, is a refinement of PG&E’s 21

prior methodology by calculating the loading factor on the basis of 22

distribution O&M and marginal A&G expenses rather than total rate 23

base less working capital. This refinement results in a loading factor 24

that is calculated on a basis that is consistent with the amounts to which 25

it is applied. PG&E calculated this CWC loading factor using data 26

obtained from PG&E’s 2020 GRC Phase I testimony and workpapers. 27

PG&E proposes an annualized CWC loading factor of 3.11 percent, 28

as shown in Table 10-2, below. 29

f. Franchise Fees and Uncollectible Expenses Loading Factor 30

PG&E must pay franchise fees to cities and counties for the right to 31

install and maintain equipment on public streets, roads, and 32

rights-of-ways. PG&E also incurs uncollectible expenses due to some 33

(PG&E-2)

10-7

customers defaulting on their credit (i.e., not paying for energy they 1

received). Both of these expenses are related to the total expenses 2

estimated to be incurred by the Company and are recovered from 3

customers through rates. 4

PG&E proposed an FF&U factor in the Company’s 2020 GRC 5

Phase I (A.18-12-009), applicable to all revenue requirement expenses. 6

In that Application, PG&E calculated an FF&U factor of1.0111. This 7

factor is applied to marginal distribution expenses related to investment 8

in distribution plant for the MDCC calculations and to expenses related 9

to the investment in connection equipment and to ongoing customer 10

access expenses in the MCAC calculations. 11

g. Loaders Model Changes 12

O&M loading factors have been calculated separately for each year 13

of data and those estimates were used to compute average as was 14

done for the 2017 GRC Phase II filing. 15

C. Results 16

The results from PG&E’s financial factors model and loaders model are 17

summarized in Table 10-1 and Table 10-2 below: 18

TABLE 10-1 SUMMARY OF RESULTS OF FINANCIAL FACTORS MODEL

Line No. Category PVRR RECC

1 Transformers Composite 1.32179 7.34%

2 Services Composite 1.31324 6.17%

3 Meters 1.19298 8.50%

4 Primary Composite 1.33026 6.36%

5 Secondary Composite 1.33139 6.53%

6 Customer Composite 1.31760 6.78%

8 Common Plant Composite 1.17857 12.00%

9 Electric General Plant Composite 1.14638 8.95%

10 Electric Transmission Composite 1.30759 6.56%

(PG&E-2)

10-8

TABLE 10-2 SUMMARY OF LOADERS BY CATEGORY

Line No. Loader Category Value

1 Distribution O&M Substation Plant 2.13%

2 Distribution O&M Meter Plant 1.27%

3 Distribution O&M Overhead Primary Line Plant 8.18%

4 Distribution O&M Overhead Secondary Line Plant 3.14%

5 Distribution O&M Underground Primary Line Plant 1.10%

6 Distribution O&M Underground Secondary Line Plant 1.10%

7 Distribution O&M Service Drop Plant 0.48%

8 Distribution O&M Overhead Service Drop Plant 0.43%

9 Distribution O&M Underground Service Drop Plant 0.64%

10 Distribution O&M Final Line Transformer Plant 0.08%

11 Distribution O&M Aboveground Final Line Transformers Plant 0.05%

12 Distribution O&M Underground Final Line Transformers Plant 0.18%

13 Transmission O&M Transmission Plant 2.98%

14 Transmission A&G Transmission O&M Expenses 17.85%

15 Transmission GPLF Transmission O&M + A&G Expenses 14.59%

16 Distribution A&G Distribution O&M expenses 31.63%

17 Distribution GPLF Distribution O&M + A&G expenses 6.03%

18 M&S Distribution O&M + A&G expenses 0.88%

19 CWC Distribution O&M + A&G expenses 3.11%

20 FF&U All Expenses 1.0111

D. Conclusion 1

Marginal cost loading factors and financial factors must be included in the 2

marginal cost calculations to obtain an accurate determination of marginal cost. 3

PG&E requests that Commission approve the financial factors and loading 4

factors presented above, as well as the methodologies for calculating them, 5

for inclusion in the appropriate marginal costs. 6

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 11

TIME-OF-USE PERIOD ASSESSMENT AND ANALYSIS

(PG&E-2)

11-i

PACIFIC GAS AND ELECTRIC COMPANY CHAPTER 11

TIME-OF-USE PERIOD ASSESSMENT AND ANALYSIS

TABLE OF CONTENTS

A. Introduction ..................................................................................................... 11-1

B. Marginal Distribution Costs, TOU Information in FERC Filings, and Status of DER Valuation Methodologies .................................................................... 11-4

1. Marginal Distribution Cost in 2025 ........................................................... 11-4

2. TOU Information Contained in FERC Transmission Filings ..................... 11-5

3. Distribution Resource Plan and Integrated Distributed Energy Resources Valuation Methodologies ........................................................ 11-5

C. Revisiting the Summer Period: Should June Still Be Designated a Summer Month? ........................................................................................... 11-10

D. Revisiting TOU Periods: PG&E’s Dead Band Tolerance Criteria ................ 11-13

1. Definition of PG&E’s Dead Band Tolerance Criteria .............................. 11-13

2. Calculation of Goodness of Separation Metrics for Peak TOU Periods . 11-14

3. Calculation of GOS Metrics for Super Off-Peak TOU Periods ............... 11-19

E. Conclusion .................................................................................................... 11-23

(PG&E-2)

11-1

PACIFIC GAS AND ELECTRIC COMPANY 1

CHAPTER 11 2

TIME-OF-USE PERIOD ASSESSMENT AND ANALYSIS 3

A. Introduction 4

Two California Public Utilities Commission (CPUC or Commission) decisions 5

require Pacific Gas and Electric Company (PG&E) to provide data on marginal 6

distribution costs (MDC) that contribute to total peak-hour marginal cost1 and 7

assess the appropriateness of the Time-of-Use (TOU) periods and seasons it 8

currently uses in rates. This chapter provides the required data and 9

assessments and presents how PG&E proposes to proceed based on 10

its analysis. 11

First, Decision (D.) 17-01-006 directs PG&E and the other investor-owned 12

utilities (IOU) to provide three types of data: (1) “marginal distribution costs that 13

contribute to total peak-hour marginal cost;” (2) TOU information included in IOU 14

transmission filings at the Federal Energy Regulatory Commission (FERC) or 15

adopted in FERC transmission rate proceedings; and (3) information on the 16

status of Distributed Energy Resource (DER) valuation methodologies being 17

developed in Rulemaking (R.) 14-08-013 and 14-10-003 or successor 18

proceedings.2 PG&E describes the required data and information in Section B. 19

Second, D.18-08-013 directs PG&E to “refresh its data appearing in 20

Chapter 12 of PG&E-9 for its next GRC Phase II Application and describe why 21

June should or should not be included in its summer season in that 22

Application.”3 PG&E describes the methodology and data from Chapter 12 of its 23

2017 General Rate Case (GRC) Phase II (Application (A.) 16-06-013), and the 24

results from refreshing these data based on modeling proposed in this 25

proceeding, in Section C, below. 26

1 PG&E considers that only Primary distribution costs are time-differentiated. Thus, only

Primary marginal distribution costs “contribute to total peak-hour marginal cost.” In this chapter, MDC therefore refers to what are called Primary Marginal Distribution Capacity Costs (MDCC) in Chapter 8.

2 See Ordering Paragraph (OP) 3. 3 D.17-01-006, Id.

(PG&E-2)

11-2

Finally, D.17-01-006 directs PG&E and the other IOUs to submit Tier 2 1

advice letters (AL) setting forth their proposals for determining when a change in 2

the time pattern of electricity costs would be sufficiently large (exceeding a 3

“Dead Band Tolerance”) to allow a proposal to revise TOU periods more 4

frequently than every two GRC cycles, along with a mechanism for 5

implementation.4,5 D.17-01-006 requires a base TOU period analysis to be 6

provided in each GRC, even if the IOU does not propose a change in Base TOU 7

periods.6 PG&E describes its Dead Band Tolerance analysis methodology and 8

results in Section D, below. 9

Because the last two directives listed above require PG&E to consider only 10

marginal generation costs (MGC) in its evaluation of TOU periods and seasons,7 11

PG&E considers both seasonal and TOU changes initially using only MGC data. 12

However, to facilitate consideration of the contribution of MDCs to time-varying 13

marginal costs, and seasonal and TOU definitions, PG&E also provides the 14

same analyses using a combination of MGCs and MDCs. 15

The results of the analysis show that the Dead Band Tolerance range for the 16

peak period has been exceeded based on MGCs; however, PG&E is not 17

proposing a change in Base TOU periods at this time. This is consistent with 18

PG&E’s objective in this proceeding as described in PG&E’s policy chapter, 19

Exhibit (PG&E-1), Chapter 1, to minimize rate design changes at this time 20

(e.g., levels of customer charges, TOU and demand charge relationships). 21

4 Also, on February 16, 2017, the CPUC issued D.17-02-017, titled “Order Correcting

Errors in Decision 17-01-006.” 5 On March 30, 2017, PG&E submitted AL 5037-E, which described PG&E’s original

Dead Band Tolerance proposal, and a proposed mechanism for implementation. On November 29, 2018, the Commission issued Resolution E-4948, approving with modifications the Dead Band Tolerance proposals of PG&E and the other IOUs and directing the IOUs to modify their proposals via supplemental compliance ALs within 30 days of the effective date of the order. PG&E then issued Supplemental AL E-4948-E-A on December 28, 2018, which became effective as of January 2, 2019.

6 See D.17-01-006, Appendix 1, p. 2, Section 6. 7 For the question of whether June should or should not be included in the summer

season, the “data appearing in Chapter 12 of PG&E-9” are based on MGCs, and thus PG&E’s refresh of those data should also consider only MGCs. Also, the second “general principle” adopted by the Commission in D.17-01-006, states “Base TOU periods should be based on utility-specific marginal costs, rather than on a statewide load assessment. This marginal cost analysis should use MGC, consisting of marginal energy costs and marginal generation capacity costs.” (D.17-01-006, p. 12.)

(PG&E-2)

11-3

Minimizing rate design changes will provide a reasonable degree of stability in 1

rates for the 2020 GRC Phase II cycle, needed due to the significant customer 2

transitions to new rates and new, later TOU periods in all customer sectors that 3

the CPUC has already approved and which are still being rolled out to 4

customers during the period 2020-2022. Per D.18-05-011 and D.19-07-004, 5

PG&E will begin transitioning eligible Residential customers in waves to the 6

default TOU rate with a 4 p.m. to 9 p.m. peak period starting in October 2020 7

and finishing in or about early 2022. Per D.18-08-013 and PG&E 8

Advice Letter 5785-E, approved April 20, 2020, PG&E will be transitioning all 9

eligible Commercial customers to rates with a 4 p.m.- 9 p.m. peak period in 10

March 2021, and all eligible Agricultural customers to rates with a 5 p.m.-8 p.m. 11

peak period.8 12

Rate structure stability is needed to provide time for customers to adapt to 13

their new TOU rate structures, avoid customer confusion that would result if 14

TOU periods were to change soon after this ongoing rollout-out process, and 15

increase customer understanding and acceptance of rate transitions. Note that 16

if PG&E were to propose changes to TOU periods in this proceeding, the 17

changes would be expected to be in place sometime in 2023, just one to two 18

years after customers would have become subject to new TOU periods. This 19

does not seem to allow enough time for such customers to have adapted to 20

those TOU periods, which will be widely marketed, and PG&E believes would be 21

too soon to force them to shift their business systems yet again to accommodate 22

yet another change in TOU period hours. 23

8 The new Commercial rates will also have a Super-Off-Peak period of 9 a.m.-2 p.m. in

March through May, and Commercial and Agricultural summer / winter season definitions are changing from a six month summer / six month winter, to a four month summer / eight month winter.

(PG&E-2)

11-4

B. Marginal Distribution Costs, TOU Information in FERC Filings, and Status 1

of DER Valuation Methodologies 2

1. Marginal Distribution Cost in 2025 3

In accordance with OP 3 of D.17-01-006, PG&E forecasted Primary 4

MDCs as of the TOU Target Year of 2025.9 While both historical and 5

forecast information on distribution marginal costs can be developed at a 6

Division level, D.17-01-006 concludes that 7

[g]eographically-differentiated TOU time periods within an IOU’s service 8 territory are not required or encouraged at this time.10 9

PG&E therefore developed a 2025 forecast of MDCs from available 10

forecasts of aggregate (service territory-wide) loads, incorporating forecasts 11

of aggregate DER levels. MDCs were calculated by multiplying the annual 12

Primary MDCC of $47.96 per kilowatt-year set forth in Chapter 7 of this 13

exhibit by Peak Capacity Allocation Factors, as described in Chapter 8 of 14

this exhibit.11 Table 11-1 presents the average MDCs12 by TOU period, 15

while Figure 11-1 shows average MDCs by hour ending (HE) and month. 16

In Figure 11-1, the current summer Peak period is outlined in red; the 17

summer Partial-Peak period is outlined in orange; and the Super-Off-Peak 18

(SOP) period is outlined in green. Data in Table 11-1 and Figure 11-1 show 19

that the vast majority of MDCs occur during the summer Peak period, with 20

lesser but still significant costs in HE 22 (i.e., the 60-minute period between 21

9 p.m. and 10 p.m.) in summer, the Partial Peak in September, and during 22

the winter Peak in October. MDCs in all other month-hour combinations are 23

less than one cent per kilowatt-hour. 24

9 In accordance with D.17-01-006 (specifically, General Principle #4), TOU periods must

be evaluated using marginal costs forecasted as of at least three years after the new rates would go into effect. Assuming that a Final Decision in this Application is issued in or about mid-2021, the earliest that rates based on this Application could be implemented using new definitions of season or TOU period would likely be mid-2022, due to the necessary structural Information Technology programming as well as customer education that would be required to prepare customers for new seasons and/or TOU periods. This implies that the forecast year to be used in evaluating seasons and TOU periods (which PG&E refers to as the TOU Target Year) is 2025.

10 See Appendix 1, p. 1. 11 Details of the calculations are available in Workpapers. 12 These data are not weighted by hourly-average load.

(PG&E-2)

11-5

TABLE 11-1 AVERAGE MARGINAL PRIMARY DISTRIBUTION COSTS

BY TIME-OF-USE PERIOD AS OF 2025 (CENTS PER KWH)

Line No. TOU Period

Marginal Distribution Cost

1 Summer Peak 8.47 2 Summer Partial-Peak 1.23 3 Summer Off-Peak 0.07 4 Winter Peak 0.61 5 Winter Off-Peak 0.04 6 Spring SOP 0.03

FIGURE 11-1 AVERAGE MARGINAL DISTRIBUTION COSTS BY MONTH AND HOUR ENDING AS OF 2025

(CENTS PER KWH)

2. TOU Information Contained in FERC Transmission Filings 1

In D.17-01-006, the Commission required that the IOUs report on TOU 2

information included in IOU transmission filings at the FERC or adopted in 3

FERC transmission rate proceedings. PG&E has not included TOU 4

information in any FERC filings to date, nor has the FERC adopted any 5

transmission rates for PG&E that include TOU information. 6

3. Distribution Resource Plan and Integrated Distributed Energy 7

Resources Valuation Methodologies 8

In D.17-01-006, the Commission required that the IOUs include 9

information on the status of the Distribution Resource Plan (DRP) and 10

Integrated Distributed Energy Resources (IDER) valuation methodologies 11

Month 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 241 0 0 0 0 0 0 0 0.03 0.03 0.01 0.01 0.01 0.00 0.01 0.01 0.02 0.09 0.35 0.51 0.52 0.36 0.16 0.01 0.002 0 0 0 0 0 0 0 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.13 0.21 0.12 0.03 0.00 0.003 0 0 0 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.06 0.06 0.01 0.00 0.004 0 0 0 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.005 0 0 0 0 0 0 0 0.00 0.01 0.05 0.07 0.09 0.10 0.11 0.11 0.10 0.20 0.36 0.26 0.16 0.28 0.18 0.02 0.006 0 0 0 0 0 0 0 0.00 0.01 0.03 0.04 0.06 0.04 0.07 0.22 0.51 1.59 6.16 12.01 13.03 9.50 5.03 0.86 0.027 0 0 0 0 0 0 0 0.00 0.01 0.04 0.05 0.04 0.03 0.02 0.02 0.04 0.16 1.71 10.00 11.98 7.21 2.21 0.13 0.008 0 0 0 0 0 0 0 0.00 0.07 0.16 0.16 0.16 0.15 0.13 0.14 0.35 2.07 7.46 13.93 12.82 8.55 2.98 0.37 0.019 0.006 0 0 0 0 0 0 0.08 0.26 0.38 0.40 0.45 0.52 0.64 1.02 2.24 6.11 12.22 13.23 11.72 8.05 2.97 0.58 0.02

10 0 0 0 0 0 0 0 0.23 0.25 0.19 0.25 0.31 0.37 0.43 0.50 0.91 2.71 4.66 5.04 4.89 2.28 0.49 0.04 0.0011 0 0 0 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.08 0.08 0.04 0.01 0.00 0.0012 0 0 0 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.23 0.26 0.19 0.08 0.00 0.00

(PG&E-2)

11-6

and relationship of these methodologies to the data presented by the IOU.13 1

In this section, PG&E provides an update from these proceedings, which 2

initially proceeded on parallel paths but are now moving closer to a universal 3

cost-effectiveness framework. 4

In October 2014, the Commission opened IDER R.14-10-003 to 5

consider the development and adoption of a regulatory framework to provide 6

policy consistency for the direction and review of demand-side resource 7

programs (the “IDER Proceeding”). One of the cornerstones of the IDER 8

Proceeding is the development of technology-neutral cost-effectiveness 9

methods and protocols including standardization of the Avoided Cost 10

Calculator (ACC) across DER proceedings and development of a Societal 11

Cost Test (SCT) for determining cost-effectiveness of demand-side 12

resources. 13

On September 28, 2017, the Commission issued D.17-09-026 (the 14

“Decision”) adopting the Locational Net Benefit Analysis (LNBA) 15

methodology from the Track 1 decision of the DRP proceeding 16

(R.14-08-013). That Track 1 decision had found the LNBA methodology 17

developed by the LNBA working group to be useful for calculating the value 18

of avoided costs provided by DERs for specific distribution deferral projects 19

that the IOUs were considering for competitive solicitation. In addition, 20

PG&E and the other large IOUs were directed to use LNBA for the Public 21

Tool and Heat Map and the Distribution Investment Deferral Framework that 22

was being considered at the time (and was ultimately adopted in 23

D.18-02-004). The September 2017 Decision further directed that the LNBA 24

tool incorporate additional value streams, including avoided distribution 25

capacity costs beyond the ten-year planning cycle, asset life extension 26

avoided costs, and contributions from smart inverters. The Decision also 27

requires the LNBA to consider DER integration costs to inform other 28

Commission proceedings (e.g., net energy metering). 29

Concurrent with the development of the two consensus use cases 30

adopted in D.17-09-026, in a Ruling dated June 7, 2017, the Assigned 31

Commissioner directed continued discussions on long-term refinements 32

13 See D.17-01-006, p. 28; see also OP 3.

(PG&E-2)

11-7

pertaining to LNBA, including the appropriate avoided local generation 1

capacity costs and avoided local transmission costs. Working group 2

meetings continued until December 2017, and the IOUs filed a final working 3

group report on long-term LNBA refinements on January 9, 2018. 4

The scope of the DRP proceeding (R.14-08-013), in accordance with 5

Public Utilities Code (Pub. Util. Code) Section 769, includes determining 6

how to calculate the value of avoided transmission and distribution (T&D) 7

costs for DERs procured through Commission mandated programs, such as 8

energy efficiency or net energy metering, including 9

[e]valuat[ing] locational benefits and costs of distributed resources 10 located on the distribution system. This evaluation shall be based on 11 reductions or increases in local generation capacity needs, avoided or 12 increased investments in distribution infrastructure, safety benefits, 13 reliability benefits, and any other savings the distributed resources 14 provide to the electrical grid or costs to ratepayers of the electrical 15 corporation.”14 16

To resolve this issue, Administrative Law Judge (ALJ) Mason issued a 17

Ruling on June 5, 2019, seeking comments on an Energy Division (ED) staff 18

white paper proposing a new methodology for calculating the value that 19

results from DERs deferring T&D investments. That Ruling and staff white 20

paper have begun a stakeholder process for updating the avoided T&D cost 21

methodology for use in the ACC. On June 21, 2019, the parties, including 22

the IOUs, Solar Energy Industries Association, and The Utility Reform 23

Network, filed opening comments. On July 18, 2019, ED staff hosted a 24

workshop to review and gather input from parties on the staff proposal and 25

allow parties to present methodologies for calculating a value that results 26

from DERs deferring transmission investments. Reply comments were filed 27

August 23, 2019. ED staff shared a proposed schedule to resolve this 28

matter at the July 18th workshop, with the next step being to have the 29

Energy and Environmental Economics, Inc. consulting group model the 30

avoided T&D methodology by October 2019, and then provide these 31

modeling results via an ALJ or Commissioner Ruling in late-October 2019. 32

Further developments regarding T&D costs and DER valuation occurred in 33

the IDER (ACC Update) proceeding, described below. 34

14 Pub. Util. Code 769(b)(1).

(PG&E-2)

11-8

In D.19-05-019, the CPUC adopted new cost-effectiveness policies for 1

DERs in the electric sector to align the IDER, DRP and Integrated Resource 2

Plan (IRP) proceedings and move closer to a universal cost-effectiveness 3

framework in the future. D.19-05-019 established the Total Resource Cost 4

as the primary test of cost-effectiveness for all DERs, with consideration of 5

the Program Administrator Cost and Ratepayer Impact Measure for any 6

DER regulatory activities. Additionally, three elements of the SCT are to be 7

tested in the IRP for informational purposes only through 2020. 8

In D.19-05-019, the CPUC also adopted a regulatory process for 9

changes to the ACC with minor updates approved through a Resolution 10

process in odd years and major updates requiring a formal process to be 11

initiated in odd years and completed in even years. The ACC was originally 12

developed in 2004, and through periodic updates has continued to serve as 13

a relevant and useful tool for computing utility avoided costs. The values 14

produced in the tool (such as utility costs for energy, generation capacity, 15

T&D investments, and environmental compliance) are used in demand-side 16

proceedings to determine the cost-effectiveness of DER such as energy 17

efficiency, demand response, and distributed generation.15 The ACC or its 18

underlying methodology has also been used in other contexts, such as 19

evaluations of the impacts of behind-the-meter energy storage16 and default 20

TOU rates.17 21

The Commission has recently provided direction for the 2020 major 22

updates to the ACC model in D.20-04-010 (the IDER Decision), which 23

determined that unspecified distribution marginal costs in the ACC should 24

use a system average approach (called “Method 1” in the staff white paper), 25

and that unspecified transmission costs in the ACC should use values from 26

utility GRCs. In particular, unspecified transmission marginal costs 27

15 See D.16-06-007 at OP 1.h. 16 See 2017 Self Generation Incentive Program Advanced Energy Storage Impact

Evaluation, September 7, 2018, available at: https://www.cpuc.ca.gov/uploadedFiles/CPUC_Public_Website/Content/ Utilities_and_Industries/Energy/Energy_Programs/Demand_Side_Management/Customer_Gen_and_Storage/2017_SGIP_AES_Impact_Evaluation.pdf.

17 See Attachment 1 of Supplemental Testimony on Calculation of Cost Estimates and Greenhouse Gas Reductions, A.17-12-011, September 26, 2018.

(PG&E-2)

11-9

(i.e., transmission costs not associated with specific, identified transmission 1

upgrades) for Southern California Edison and San Diego Gas & Electric 2

should be developed by staff based on PG&E’s transmission marginal 3

cost methodology.18 4

In addition, the IDER Decision formally links up the ACC with the IRP 5

proceeding, directing that avoided energy and ancillary services costs shall 6

be based on costs from the Strategic Energy and Risk Valuation 7

Model (SERVM) production cost model, while avoided generation capacity 8

costs for the ACC shall be determined by the net cost of new entry of a 9

storage battery based on the SERVM-developed energy and ancillary 10

services costs. Details of many of the methodologies spelled out in the 11

IDER Decision were discussed in workshops on May 6-8, 2020, following 12

issuance of the draft Resolution E-5077 that adopts the 2020 ACC on 13

May 1. PG&E notes that while the IDER Decision is not binding on PG&E’s 14

GRC Phase II proceeding, its direction regarding marginal transmission, 15

energy and capacity costs generally support PG&E’s filing, in that the IDER 16

Decision explicitly references PG&E’s methodology for unspecified 17

transmission marginal costs, while specifying essentially the same capacity 18

cost methodology as PG&E describes in Chapter 2. The only significant 19

differences are that the IDER Decision specifies the use of SERVM to 20

calculate energy and ancillary service prices rather than a 21

statistically-derived model,19 and that PG&E uses short-run capacity costs 22

from an existing combined cycle unit when those are higher than the 23

long-run capacity costs of a new battery and in years when new capacity is 24

not needed for reliability. 25

18 D.20-04-10, p. 3. 19 On the other hand, the associated Staff Proposal notes that energy prices from

production simulation models are too “flat,” and proposes adjusting high and low energy prices to better match historical data. This is similar to the PG&E Marginal Energy Cost (MEC) model including the spread parameter in its objective function, as described in section B.1.d of Chapter 2.

(PG&E-2)

11-10

C. Revisiting the Summer Period: Should June Still Be Designated a 1

Summer Month? 2

In Chapter 12 of the Marginal Costs Volume, Exhibit (PG&E-9) of PG&E’s 3

2017 GRC Phase II Testimony, PG&E determined that the period June through 4

September should be chosen as the summer months, based on the number of 5

hours with high MGCs (“high cost hours”) occurring in each month. In that 6

determination, PG&E considered both the top 100 hours and the top 250 hours 7

of the year as high cost hours and used a forecast year of 2020. This summer 8

period definition was adopted by the CPUC in D.18-08-013.20 To refresh the 9

data and analysis developed for the 2017 GRC, PG&E uses the same 10

designation of top 100 and top 250 hours based on forecasted MGCs, but, in 11

accordance with the TOU Order Instituting Rulemaking decision, PG&E uses the 12

forecast year of 2025 required to be considered in the TOU period 13

analysis below. 14

To provide a basis for comparison with results from the 2017 GRC Phase II, 15

in Figure 11-2, PG&E (1) provides a reproduction of the previous case’s 16

Table 12-2 from Exhibit (PG&E-9) (which was based on a 2020 forecast); and 17

(2) presents results using updated MGCs for forecast years of 2020 and 2025. 18

The current summer period (June through September) is highlighted in green 19

background, while the forecasts based on the updated model for a forecast 20

year of 2025 are highlighted in bold, and the period with the highest number of 21

high-cost hours based on the updated MGCs is outlined in orange. 22

20 See D.18-08-013, p. 32.

(PG&E-2)

11-11

FIGURE 11-2 DISTRIBUTION OF TOP GENERATION MARGINAL COST HOURS ACROSS CALENDAR

MONTHS FROM 2017 AND 2020 GRC FORECASTS

PG&E notes that the updated forecasts for both 2020 and 2025 show 1

one fewer high-cost hour in June, and moderately more in October than those 2

identified for 2020 in PG&E’s 2017 GRC. All forecasts also show a modest 3

number of high-cost hours in early winter (November-December), and virtually 4

none in late winter and spring (January-May). 5

In the absence of customer considerations, the data displayed in 6

Figure 11-2, especially the figures for the months outlined in orange, suggest 7

that June should not be treated as a summer month for rates that apply in the 8

early to mid-2020s, while October should be treated as a summer month for 9

such rates. However, PG&E is concerned that the rate instability of changing 10

the definition of summer months to July-October only about two years after the 11

summer season definition had been changed to June-September would cause 12

customer confusion. 13

In addition, as described in Section A, PG&E also performed the same 14

analysis, using the sum of MGCs and MDCs to establish Top 250 and Top 100 15

hour designations. The results from that analysis of combined marginal costs is 16

presented in Figure 11-3. Because PG&E’s 2017 GRC analysis did not consider 17

MDCs directly to support the summer season definition, Figure 11-3 compares 18

2017 GRC 2020 GRC 2020 GRC 2017 GRC 2020 GRC 2020 GRCRow Month (Year 2020) (Year 2020) (Year 2025) (Year 2020) (Year 2020) (Year 2025)

1 Jan 1 0 0 0 0 02 Feb 0 0 0 0 0 03 Mar 0 0 0 0 0 04 Apr 0 0 0 0 0 05 May 0 0 0 0 0 06 Jun 2 1 0 1 0 07 Jul 39 21 10 65 16 68 Aug 24 30 23 28 40 329 Sep 10 24 27 5 35 40

10 Oct 6 17 19 0 7 1511 Nov 8 7 10 0 2 612 Dec 10 0 10 1 0 1

Top 250 Hours Top 100 HoursPercent Count of High Cost Hours (Energy + Capacity)

(PG&E-2)

11-12

the Top Cost hours using MGCs plus MDCs only with the MGC-only results from 1

Figure 11-2. 2

FIGURE 11-3 DISTRIBUTION OF TOP MARGINAL COST HOURS ACROSS CALENDAR MONTHS

FROM 2020 GRC FORECASTS

The addition of MDCs to the generation marginal costs clearly shifts the 3

forecasted distribution of top hours from July-October (outlined in orange) back 4

to the current definition of June-September (outlined in black). PG&E believes 5

that the primary driver of this difference is that solar generation (which peaks in 6

June, and is only moderate in October) affects the distribution system (and thus 7

MDCs) only one third as much as it affects the generation system (and thus 8

MGCs).21 Thus MDCs not only peak earlier in the day than MGCs, they also 9

peak earlier in the year. The result is that while June is not forecast to be a high 10

generation cost month in 2025, it is forecast to be a high cost month when 11

marginal distribution as well as generation costs are considered. 12

21 Approximately one third of solar generation in California occurs at the distribution

level (chiefly rooftop photovoltaic); the other two thirds is at the transmission, or generation level.

MGC Only MGC Plus MGC Only MGC PlusRow Month (Figure 11-2) MDCC (Figure 11-2) MDCC

1 Jan 0 0 0 02 Feb 0 0 0 03 Mar 0 0 0 04 Apr 0 0 0 05 May 0 0 0 06 Jun 0 11 0 177 Jul 10 14 6 68 Aug 23 23 32 289 Sep 27 28 40 41

10 Oct 19 16 15 811 Nov 10 5 6 012 Dec 10 4 1 0

Top 250 Hours Top 100 HoursPercent Count of High Cost Hours in 2025 (2020 GRC)

(PG&E-2)

11-13

Based on both customer considerations such as rate stability and the 1

impact of MDCs on total avoided costs, PG&E proposes to maintain the 2

June-September summer season definition at this time, with the expectation that 3

the possibility of a July-October definition will be revisited in PG&E’s 2023 GRC 4

Phase II Application. 5

D. Revisiting TOU Periods: PG&E’s Dead Band Tolerance Criteria 6

1. Definition of PG&E’s Dead Band Tolerance Criteria 7

PG&E’s Dead Band Tolerance range (or equivalently, threshold to 8

exceed) comprises two parts, both of which must be met to allow PG&E to 9

consider revising TOU periods sooner than five years after the most recent 10

change in TOU period (i.e., in this GRC Phase II Application).22 The results 11

of PG&E’s analysis show that the Dead Band Tolerance range is exceeded 12

for the peak period; however, PG&E is opting not to propose changing the 13

Base TOU periods at this time, as exceeding the Dead Band Tolerance 14

range merely suggests the option to propose changed hours but does not 15

require it.23 As discussed at the end of Section A, above, PG&E has 16

serious concerns about changing its TOU periods in this proceeding, 17

because most customers will have just seen a significant shift in TOU peak 18

22 The conditions are: (1) Changed cost data justify changing either (a) the start or ending

time of the TOU period by at least one hour (in either direction), for the summer peak, winter peak, or spring SOP; or (b) the months for which particular TOU period definitions apply; and (2) Using a forecast of MGCs, or whatever other marginal costs are used to determine TOU periods in a GRC Phase II proceeding, with the forecast year set at least three years after the year the Base TOU period will go into effect, the “goodness of separation” (GOS) metrics pertaining to the summer peak period, the winter peak period or the SOP period increase under the new TOU period definition by at least five percentage points (5%) relative to the corresponding GOS metrics using the old TOU period definition.

23 D.17-01-006, Appendix 1, p. 2, Section 6: “To evaluate whether a dead band tolerance range has been exceeded and to ensure that the Commission and the public are aware of the likelihood of future Base TOU period changes, Base TOU period analysis should be provided in each GRC, even if the IOU does not propose a change in Base TOU periods. If such analysis shows that the dead band tolerance range has been exceeded, the IOU should propose revisions to Base TOU periods.”

(PG&E-2)

11-14

hours24 and PG&E believes rate stability, to avoid the risk of customer 1

confusion, counsels waiting until the 2023 GRC Phase II Application to 2

consider potential TOU period changes. 3

2. Calculation of Goodness of Separation Metrics for Peak TOU Periods 4

The definitions of GOS metrics and their components are provided in 5

Attachment A. This section describes how those components were 6

calculated and the resulting metrics for peak periods. 7

To evaluate summer and winter peak periods, PG&E determined the 8

number of high-cost hours that occur during various potential on-peak 9

periods in summer (June-September) and winter (October-May), using a 10

forecast year of 2025, for each of the ten weather scenarios described in 11

Chapter 2 of this exhibit, and then used the averages over all scenarios to 12

develop GOS metrics. For example, to develop the GOS metric for a peak 13

period of 4 p.m. – 9 p.m., PG&E set a “high cost flag” to one in each hour of 14

the year 2025 forecast for each weather scenario if the MGC was in the top 15

5 percent of year 2025 hours for that weather scenario, and set the high cost 16

flag to zero if the MGC was not in the top 5 percent of year 2025 hours. 17

PG&E then took the average of the high cost flags over all scenarios to get 18

an “average high cost flag” by hour of the year 2025 forecast. Those 19

average high cost flags were used to calculate the expected number of true 20

positive, false negative, true negative, and false negative hours for the 21

4 p.m. – 9 p.m. period, and thence the A and B factors and GOS metric. 22

The results for various summer On-Peak definitions are shown in 23

Table 11-2; results for winter On-Peak periods are in Table 11-3; while 24

Table 11-4 shows weighted average GOS metrics for the entire calendar 25

year (which is appropriate if the TOU periods are required to be the same in 26

summer and winter). 27

24 These shifts include primarily the Commercial and Agricultural TOU Time Period

transitions from the current Noon to 6 p.m. summer peak to the new 4 p.m. – 9 p.m. and 5 p.m. – 8 p.m. peak periods, as well as the default TOU transition of eligible Residential customers from a tiered rate plan with no TOU time period to a TOU rate with a 4 p.m. – 9 p.m. peak. Significant resources have been invested (and will continue to be invested during 2020-2021) in building customer awareness and understanding of the new 4 p.m. to 9 p.m. period for Residential and Commercial customers.

(PG&E-2)

11-15

TABLE 11-2 GOODNESS OF SEPARATION METRICS FOR SUMMER 2025

BASED ON MARGINAL GENERATION COSTS

TABLE 11-3 GOODNESS OF SEPARATION METRICS FOR WINTER 2025

BASED ON MARGINAL GENERATION COSTS

TABLE 11-4 GOODNESS OF SEPARATION METRICS FOR CALENDAR YEAR 2025

BASED ON MARGINAL GENERATION COSTS

The GOS metrics shown in Tables 11-2 through 11-4 combine the 1

two metrics that were used to determine proposed TOU periods in PG&E’s 2

2017 GRC Phase II. In that earlier proceeding, PG&E reported the values 3

for the true positive rate A and false positive rate B, and explained that an 4

optimal TOU period would have a high value for A and a low value for B.25 5

The GOS metric, calculated as A * (1-B), combines these preferences into a 6

single metric, where a higher value for GOS generally implies a higher value 7

25 See A.16-06-013, Exhibit (PG&E-9), p. 12-14.

Line No.Summer

Peak HoursTrue Pos.

HoursFalse Neg.

HoursFalse Pos.

HoursTrue Neg.

HoursTrue Pos.

Rate AFalse Pos.

Rate BGOS

Metric1 4PM-9PM 155.7 42.3 454.3 2275.7 78.6% 16.6% 65.6%2 5PM-9PM 155 43 333 2397 78.3% 12.2% 68.7%3 5PM-10PM 192.2 5.8 417.8 2312.2 97.1% 15.3% 82.2%4 6PM-10PM 188.7 9.3 299.3 2430.7 95.3% 11.0% 84.9%

Line No.Winter

Peak HoursTrue Pos.

HoursFalse Neg.

HoursFalse Pos.

HoursTrue Neg.

HoursTrue Pos.

Rate AFalse Pos.

Rate BGOS

Metric1 4PM-9PM 211 29 1004 4588 87.9% 18.0% 72.1%2 5PM-9PM 211 29 761 4831 87.9% 13.6% 76.0%3 5PM-10PM 234.9 5.1 980.1 4611.9 97.9% 17.5% 80.7%4 6PM-10PM 184.5 55.5 787.5 4804.5 76.9% 14.1% 66.0%

(PG&E-2)

11-16

for A, a lower value for B, or some combination of the two. Thus, the TOU 1

period definitions that have the highest GOS metrics are considered to best 2

match the peak periods with high cost hours. 3

From Table 11-2, the summer peak TOU periods with the highest values 4

of GOS are 5 p.m. – 10 p.m. and 6 p.m. – 10 p.m. Both of those periods are 5

at least 1 hour later than the current summer peak period of 4 p.m. – 9 p.m., 6

and both have a GOS that exceeds the GOS for the current peak period by 7

more than five percent, so both of the criteria in PG&E’s Dead Band 8

Tolerance are exceeded for both of these later periods. 9

As for the winter peak period, Table 11-3 shows that both 10

5 p.m. – 9 p.m. and 5 p.m. – 10 p.m. have higher GOS than the current 11

4 p.m. – 9 p.m. winter peak, and 5 p.m. – 10 p.m. exceeds the GOS of the 12

current peak by the established five percent Dead Band Tolerance 13

threshold. The 6 p.m. – 10 p.m. peak period has a lower winter GOS than 14

the current peak. 15

To minimize customer confusion, PG&E proposed harmonizing the peak 16

periods between summer and winter in its 2017 GRC Phase II.26 PG&E 17

believes the same arguments for harmonization apply today. Thus, 18

PG&E does not advocate for setting the peak to 6 p.m. – 10 p.m. in summer 19

and 5 p.m. – 10 p.m. in winter (which would maximize GOS in each season 20

separately), but considers which consistent all-year TOU period would 21

maximize the overall GOS.27 The results from that analysis are presented 22

in Table 11-4, and indicate that the 5 p.m. – 10 p.m. period best matches 23

the high cost hours (highest overall GOS). This is the same period 24

PG&E proposed in its 2017 GRC Phase II, where it also found 25

26 See A.16-06-013, Exhibit (PG&E-9), p. 12-16, footnote 13. 27 The current analysis assumes that California will continue to use Daylight Saving Time

(DST). However, there are various proposals to eliminate DST or to apply it year-round, which could come into force during the life of this Application. If DST were to be eliminated (so Pacific Standard Time applied all year), a 5 p.m. – 9 p.m. peak period would likely best align with high cost hours in both summer and winter; while if DST applied all year a 6 p.m. – 10 p.m. peak period would likely best align with high cost hours in both summer and winter. However, while the impact of a DST change on average solar and wind production by TOU period is easily determined, changing the DST usage could affect load in complex ways, so more analysis would be required to confirm these expectations.

(PG&E-2)

11-17

5 p.m. – 10 p.m. to have the best match to high cost hours.28 Considering 1

this year-round metric, the later period meets the Dead Band Tolerance 2

criterion, since the GOS for 5 p.m. – 10 p.m. is 12 percent (81.1 – 69.1) 3

better than the GOS for the current 4 p.m. – 9 p.m. period.29 4

PG&E also calculated GOS metrics based on the combination of MGCs 5

and MDCs; results are shown in Tables 11-5 through 11-7. 6

TABLE 11-5 GOODNESS OF SEPARATION METRICS FOR SUMMER 2025

BASED ON MARGINAL GENERATION PLUS DISTRIBUTION COSTS

28 PG&E and other parties settled on the 4 p.m. – 9 p.m. peak period for the 2017 GRC

Phase II, partly to align PG&E’s TOU peak period with those of the other IOUs, which were proposing a 4 p.m. – 9 p.m. peak period. However, marginal cost values were neither litigated nor proposed in the settlement agreement adopted in PG&E’s 2017 GRC. PG&E is here referring to its calculated metrics based on its proposed marginal costs in that proceeding.

29 Note that in its November 22, 2019 GRC filing (in which the Dead Band Tolerance was not exceeded for the Peak period), PG&E suggested that GOS metrics were unlikely to change significantly in this July update, since the GOS metrics are based on the timing rather than the magnitude of marginal generation and distribution costs. However, PG&E now realizes that in this July update, the marginal generation capacity costs (MGCC) increased substantially compared to the November 2019 filing, while MEC were relatively unchanged or even flattened by the addition of energy storage modeling. Out of the three marginal costs considered in this chapter (MEC, MGCC and MDC), MGCC has the latest peak, because the Adjusted Net Load (ANL) that is used to calculate its 8760 shapes includes the impact of both behind- and front-of-the meter solar generation (thus peaking later than MDCs, which are only affected by behind-the-meter solar), and does not include the impact of ramping and temperature effects in the rest of the Western Electricity Coordinating Council, both of which shift the peak of MECs earlier than the peak of ANL (as discussed in Chapter 2). Thus higher MGCCs relative to the other components used in the GOS calculation tend to shift the peak later, and thus increase the difference between the GOS of the current 4 p.m. – 9 p.m. peak and the later peak periods examined here.

Line No.Summer

Peak HoursTrue Pos.

HoursFalse Neg.

HoursFalse Pos.

HoursTrue Neg.

HoursTrue Pos.

Rate AFalse Pos.

Rate B GOS Metric1 4PM-9PM 255.4 60.6 354.6 2257.4 80.8% 13.6% 69.9%2 5PM-9PM 244.1 71.9 243.9 2368.1 77.2% 9.3% 70.0%3 5PM-10PM 291.1 24.9 318.9 2293.1 92.1% 12.2% 80.9%4 6PM-10PM 263.3 52.7 224.7 2387.3 83.3% 8.6% 76.2%

(PG&E-2)

11-18

TABLE 11-6 GOODNESS OF SEPARATION METRICS FOR WINTER 2025

BASED ON MARGINAL GENERATION PLUS DISTRIBUTION COSTS

TABLE 11-7 GOODNESS OF SEPARATION METRICS FOR CALENDAR YEAR 2025 BASED ON MARGINAL GENERATION PLUS DISTRIBUTION COSTS

Comparing Table 11-5 to Table 11-2, both True Positive and False 1

Negative hours are greater for the combined costs than for MGCs alone. 2

This is because MDCs are concentrated in the summer more than MGCs 3

are, so all TOU periods shown here have more high cost hours total in the 4

summer when Distribution costs are included. The GOS for the 5

4 p.m. – 9 p.m. and 5 p.m. – 9 p.m. periods are greater when Distribution 6

costs are included; later TOU periods (especially the 6 p.m. – 10 p.m. 7

period) show lower GOS when Distribution costs are included. 8

Comparing Table 11-6 to Table 11-3, both True Positive and False 9

Negative hours are lower for the combined costs than for MGCs alone, while 10

all winter TOU periods except 5 p.m. – 10 p.m. show higher GOS when 11

Distribution costs are included. 12

Finally, comparing Table 11-7 to 11-4, all year-round TOU periods 13

except for 4 p.m. – 9 p.m. show lower GOS metrics when Distribution costs 14

are included, with the 5 p.m. – 10 p.m. period showing greater reductions 15

than the other TOU periods. Based on this analysis, if MDCs were included 16

Line No.Winter

Peak HoursTrue Pos.

HoursFalse Neg.

HoursFalse Pos.

HoursTrue Neg.

HoursTrue Pos.

Rate AFalse Pos.

Rate B GOS Metric1 4PM-9PM 114.8 7.2 1100.2 4609.8 94.1% 19.3% 76.0%2 5PM-9PM 111 11 861 4849 91.0% 15.1% 77.3%3 5PM-10PM 117.3 4.7 1097.7 4612.3 96.1% 19.2% 77.7%4 6PM-10PM 96.7 25.3 875.3 4834.7 79.3% 15.3% 67.1%

Line No.All-Year

Peak HoursTrue Pos.

HoursFalse Neg.

HoursFalse Pos.

HoursTrue Neg.

HoursTrue Pos.

Rate AFalse Pos.

Rate B GOS Metric

1 4PM-9PM 370.2 67.8 1454.8 6867.2 84.5% 17.5% 69.7%2 5PM-9PM 355.1 82.9 1104.9 7217.1 81.1% 13.3% 70.3%3 5PM-10PM 408.4 29.6 1416.6 6905.4 93.2% 17.0% 77.4%4 6PM-10PM 360 78 1100 7222 82.2% 13.2% 71.3%

(PG&E-2)

11-19

in the TOU period analysis, the 5 p.m. – 10 p.m. TOU period would again be 1

shown to be the most aligned with marginal Generation plus Distribution 2

costs. Furthermore, the Dead Band Tolerance threshold of five percent 3

would again be exceeded, since the GOS for 5 p.m. – 10 p.m. is 7.7 percent 4

(77.4 – 69.7) better than the GOS for the current 4 p.m. – 9 p.m. period. 5

However, based on the discussion above regarding providing rate stability 6

for customers as they are transitioned to new rates and new TOU periods 7

over the next few years, PG&E believes that the peak period for both 8

summer and winter should remain 4 p.m. – 9 p.m. at this time. 9

3. Calculation of GOS Metrics for Super Off-Peak TOU Periods 10

PG&E performed similar calculations to test various combinations of 11

months and hours for the SOP period. For this analysis, PG&E considered 12

both the current start time of 9 a.m. and the SOP start time of 8 a.m. that 13

applies in Southern California Edison Company’s (SCE) rates;30 and hourly 14

ending times from the current 2 p.m. through 5 p.m. In addition, PG&E 15

considered three seasonal definitions: (1) the current SOP season of 16

March-May; (2) March-June; and (3) November-June (i.e., all months except 17

for the summer months of July-October that were shown to be preferred in 18

Section C). The calculations and resulting GOS metrics for all combinations 19

are shown in Figure 11-4. 20

30 SCE has an 8 a.m. – 4 p.m. SOP during the winter (October-May) in its TOU-D-4-9PM

rate, and a winter SOP from 8 a.m. – 5 p.m. in its TOU-D-5-8PM rate. See https://pages.email.sce.com/RatePlanOptions/en (accessed November 14, 2019).

(PG&E-2)

11-20

FIGURE 11-4 GOS CALCULATIONS FOR VARIOUS SOP PERIODS

FOR CALENDAR YEAR 2025 BASED ON MARGINAL GENERATION COSTS

In the above figure, each set of three rows correspond to the 1

three seasonal definitions using the designated TOU period; the sets of rows 2

are organized such that lower rows have either earlier starting times or later 3

ending times (thus, longer SOP periods). The current SOP definition 4

(9 a.m. – 2 p.m., March-May) is shown on the first row, and actually has the 5

lowest GOS of all combinations tested. 6

Within each set of three rows, the highest GOS is highlighted; the 7

highest GOS over all combinations (corresponding to the March-May SOP 8

season and SOP hours of 8 a.m. – 5 p.m.) is highlighted over the entire row. 9

PG&E notes that while the 8 a.m. start time always yields a higher GOS 10

SeasonHB

StartHE

EndTruePos.

FalseNeg.

FalsePos.

TrueNeg.

True Pos A

False Pos B GOS

Mar-May 9 14 412.8 335.1 47.2 1412.9 55.2% 3.2% 53.4%Nov-Jun 712.3 482.4 497.7 4115.6 59.6% 10.8% 53.2%Mar-Jun 534.7 420.8 75.3 1897.2 56.0% 3.8% 53.8%Mar-May 8 14 470.1 277.8 81.9 1378.2 62.9% 5.6% 59.3%Nov-Jun 802 392.7 650 3963.3 67.1% 14.1% 57.7%Mar-Jun 613.6 341.9 118.4 1854.1 64.2% 6.0% 60.4%Mar-May 9 15 496.3 251.6 55.7 1404.4 66.4% 3.8% 63.8%Nov-Jun 850.3 344.4 601.7 4011.6 71.2% 13.0% 61.9%Mar-Jun 639.6 315.9 92.4 1880.1 66.9% 4.7% 63.8%Mar-May 8 15 553.6 194.3 90.4 1369.7 74.0% 6.2% 69.4%Nov-Jun 940 254.7 754 3859.3 78.7% 16.3% 65.8%Mar-Jun 718.5 237 135.5 1837 75.2% 6.9% 70.0%Mar-May 9 16 575.4 172.5 68.6 1391.5 76.9% 4.7% 73.3%Nov-Jun 963.6 231.1 730.4 3882.9 80.7% 15.8% 67.9%Mar-Jun 736.4 219.1 117.6 1854.9 77.1% 6.0% 72.5%Mar-May 8 16 632.7 115.2 103.3 1356.8 84.6% 7.1% 78.6%Nov-Jun 1053.3 141.4 882.7 3730.6 88.2% 19.1% 71.3%Mar-Jun 815.3 140.2 160.7 1811.8 85.3% 8.1% 78.4%Mar-May 9 17 641.4 106.5 94.6 1365.5 85.8% 6.5% 80.2%Nov-Jun 1042.1 152.6 893.9 3719.4 87.2% 19.4% 70.3%Mar-Jun 813.7 141.8 162.3 1810.2 85.2% 8.2% 78.2%Mar-May 8 17 698.7 49.2 129.3 1330.8 93.4% 8.9% 85.1%Nov-Jun 1131.8 62.9 1046.2 3567.1 94.7% 22.7% 73.3%Mar-Jun 892.6 62.9 205.4 1767.1 93.4% 10.4% 83.7%

(PG&E-2)

11-21

than a 9 a.m. start for the same season and end-time (i.e., the GOS for each 1

8 a.m. start row is greater than the GOS for the 9 a.m. start three rows up), 2

the seasonal pattern is much less consistent, except that all the SOP 3

periods that end at 4 p.m. or 5 p.m. (i.e., the last four sets of rows) show the 4

highest GOS metric with the current March through May SOP season, with 5

March through June GOS metrics at most 2% worse. 6

PG&E also calculated GOS metrics for the SOP using combined 7

Generation and Distribution costs; results are shown in Figure 11-5. The 8

results shown in Figure 11-5 are almost identical to those in Figure 11-4, 9

with individual GOS metrics changing by at most 1.2 percent, and the 10

ordering of TOU periods and seasonal combinations (e.g., within each set of 11

three and among TOU periods) almost identical between the MGC-only and 12

combined cost results. 13

(PG&E-2)

11-22

FIGURE 11-5 GOS CALCULATIONS FOR VARIOUS SOP PERIODS

FOR CALENDAR YEAR 2025 BASED ON MARGINAL GENERATION AND DISTRIBUTION COSTS

Whether or not the analysis includes Distribution costs, the SOP 1

definition that most aligns with the incidence of very low-cost hours (with 2

marginal costs at or below zero) runs from March through May and applies 3

from 8 a.m. to 5 p.m. The end of that period aligns with the start of the peak 4

period with the highest GOS metric determined in Subsection 2, above. So, 5

if the peak period were to be changed to 5 p.m. to 10 p.m., the SOP in 6

earlier months would align with it, and therefore be easy to remember. 7

PG&E also notes that the GOS metric for the March-June 8 a.m. – 5 p.m. 8

SOP definition is the second highest of all tested combinations. Thus, if the 9

SeasonHB

Start HE EndTruePos.

FalseNeg.

FalsePos.

TrueNeg.

True Pos A

False Pos B GOS

Mar-May 9 14 408.6 323.4 51.4 1424.6 55.8% 3.5% 53.9%Nov-Jun 695 458 515 4140 60.3% 11.1% 53.6%Mar-Jun 518.1 396.6 91.9 1921.4 56.6% 4.6% 54.1%Mar-May 8 14 464.6 267.4 87.4 1388.6 63.5% 5.9% 59.7%Nov-Jun 782.3 370.7 669.7 3985.3 67.8% 14.4% 58.1%Mar-Jun 594.6 320.1 137.4 1875.9 65.0% 6.8% 60.6%Mar-May 9 15 490.4 241.6 61.6 1414.4 67.0% 4.2% 64.2%Nov-Jun 828.4 324.6 623.6 4031.4 71.8% 13.4% 62.2%Mar-Jun 618.4 296.3 113.6 1899.7 67.6% 5.6% 63.8%Mar-May 8 15 546.4 185.6 97.6 1378.4 74.6% 6.6% 69.7%Nov-Jun 915.7 237.3 778.3 3876.7 79.4% 16.7% 66.1%Mar-Jun 694.9 219.8 159.1 1854.2 76.0% 7.9% 70.0%Mar-May 9 16 568.1 163.9 75.9 1400.1 77.6% 5.1% 73.6%Nov-Jun 937.1 215.9 756.9 3898.1 81.3% 16.3% 68.1%Mar-Jun 710.8 203.9 143.2 1870.1 77.7% 7.1% 72.2%Mar-May 8 16 624.1 107.9 111.9 1364.1 85.3% 7.6% 78.8%Nov-Jun 1024.4 128.6 911.6 3743.4 88.8% 19.6% 71.4%Mar-Jun 787.3 127.4 188.7 1824.6 86.1% 9.4% 78.0%Mar-May 9 17 630.6 101.4 105.4 1370.6 86.1% 7.1% 80.0%Nov-Jun 1007.9 145.1 928.1 3726.9 87.4% 19.9% 70.0%Mar-Jun 780.4 134.3 195.6 1817.7 85.3% 9.7% 77.0%Mar-May 8 17 686.6 45.4 141.4 1334.6 93.8% 9.6% 84.8%Nov-Jun 1095.2 57.8 1082.8 3572.2 95.0% 23.3% 72.9%Mar-Jun 856.9 57.8 241.1 1772.2 93.7% 12.0% 82.5%

(PG&E-2)

11-23

summer definition were changed to July-October it could make sense to 1

continue the SOP season through June to align its end with the start of 2

summer. Both the March through May, 8 a.m. – 5 p.m. and March through 3

June, 8 a.m. – 5 p.m. definitions have GOS metrics that exceed the 4

five percent threshold improvement. Finally, the March-May, 8 a.m. – 4 p.m. 5

and 9 a.m. – 4 p.m. SOP definitions also show greater than five percent 6

GOS improvement over the current definition. 7

If PG&E were to propose a change to the SOP at this time, the most 8

customer-friendly update that would improve alignment with the incidence of 9

2025 forecast very low-cost hours would be a change to 8 a.m. – 4 p.m. 10

This would match SCE’s TOU-D-4-9PM SOP period and would align the end 11

of the SOP with the start of the current peak period, rather than the start of 12

the summer shoulder peak period as at present. An alternative would be 13

9 a.m. – 4 p.m., which keeps the current 9 a.m. start time and would also 14

align with the start of the peak period. 15

However, as with the definition of the summer season and the TOU 16

peak period, PG&E believes that changing the definition of the SOP period 17

so soon after its implementation in Commercial and Industrial rates in 18

2019-2020 would cause customer confusion and should not be adopted at 19

this time. The next opportunity to re-examine the definition of the SOP 20

would be in PG&E’s 2023 GRC Phase II, and should be based on a holistic 21

examination of seasonal and TOU period definitions, as well as other 22

considerations at that time. 23

E. Conclusion 24

PG&E presents in this Chapter the time-differentiated portion of MDCs, and 25

provides information regarding TOU-based applications at FERC, and the status 26

of DER valuation proceedings. PG&E also considers whether TOU periods, and 27

seasons, should shift based on updated marginal costs and the Dead Band 28

Tolerance criteria established in Advice Letter 5037-E-A. Whether the analyses 29

of TOU periods and seasons are based on just MGCs or also include MDCs, 30

PG&E proposes to maintain the same seasons and TOU periods that were 31

adopted in PG&E’s 2017 GRC Phase II, to avoid customer confusion so soon 32

after the current seasons and TOU periods are implemented in 2020 and 2021. 33

(PG&E-2)

11-24

PG&E requests the Commission accept PG&E’s showing of MDCs and 1

other required information and retain the most recently adopted TOU periods 2

and seasons as proposed by PG&E in this testimony. 3

(PG&E-2)

PACIFIC GAS AND ELECTRIC COMPANY

CHAPTER 11

ATTACHMENT A

PG&E SUPPLEMENTAL ADVICE LETTER 5037-E-A

(PG&E PROPOSED DEAD BAND TOLERANCE RANGE)

STATE OF CALIFORNIA EDMUND G. BROWN JR., Governor

PUBLIC UTILITIES COMMISSION 505 VAN NESS AVENUE

SAN FRANCISCO, CA 94102-3298

January 2, 2019Advice Letter 5037-E

5037-E-A

Erik JacobsonDirector, Regulatory RelationsPacific Gas and Electric Company77 Beale Street, Mail Code B10CP.O. Box 770000San Francisco, CA 94177

SUBJECT: Pacific Gas and Electric Company’s Proposed Dead Band Tolerance Range for Determining When a Change Would Trigger TOU Period Revisions More Frequently Than Five Year Intervals, in Compliance with Decision 17-01-006

Dear Mr. Jacobson:

Advice Letter 5037-E and supplemental 5037-E-A are effective as of January 2, 2019per Resolution E-4948 Ordering Paragraphs.

Sincerely,

Edward RandolphDirector, Energy Division

11-AtchA-1

(PG&E-2)

Erik Jacobson Director Regulatory Relations

Pacific Gas and Electric Company 77 Beale St., Mail Code B13U P.O. Box 770000 San Francisco, CA 94177 Fax: 415-973-3582

December 28, 2018

Advice 5037-E-A(Pacific Gas and Electric Company ID U 39 E)

Public Utilities Commission of the State of California

Subject: Supplemental: Pacific Gas and Electric Company’s Proposed Dead Band Tolerance Range for determining when a change would trigger TOU period Revisions More Frequently than Five Year Intervals, in Compliance with Decision 17-01-006 and Resolution E-4948

Purpose

This Tier 2 Advice Letter (AL) describes Pacific Gas and Electric Company’s (PG&E’s) modified dead band tolerance range for determining when a change in the time pattern of electricity costs would trigger time-of-use (TOU) period revisions more frequently than every two General Rate Case (GRC) cycles, coupled with a mechanism for implementation, in compliance with Decision (D.) 17-01-006 (Decision), Decision on Adopting Policy Guidelines to Assess Time Periods for Future Time-of-Use Rates andEnergy Resource Contract Payments,1 and Resolution E-4948.

Background

On January 23, 2017, the California Public Utilities Commission (Commission or CPUC) issued D.17-01-006 requiring PG&E, Southern California Edison Company (SCE), and San Diego Gas & Electric Company (SDG&E) (collectively the IOUs) to each submit Tier 3 advice letters setting forth their proposals for determining when a change in the time pattern of electricity costs would be sufficiently large to trigger aproposal to revise TOU periods more frequently than every two GRC cycles, along witha mechanism for implementation.2 The general principles adopted in the Decision,with respect to development and implementation of changes in Base TOU periods,include general principle number 6, which states that:

[T]o evaluate whether a dead band tolerance range has been exceeded and to

1 Also, on 2/16/2017 the CPUC issued Decision 17-02-017, titled “Order Correcting Errors in

Decision 17-01-006”.2 D.17-01-006, mimeo, p.78.

11-AtchA-2

(PG&E-2)

Advice 5037-E-A - 2 - December 28, 2018

ensure that the Commission and the public are aware of the likelihood of future Base TOU period changes, Base TOU period analysis should be provided ineach general rate case, even if the IOU does not propose a change in BaseTOU periods. If such analysis shows that the dead band tolerance range hasbeen exceeded, the IOU should propose revisions to Base TOU periods.3

On March 30, 2017, PG&E submitted Advice Letter 5037-E, which described PG&E’sdead band tolerance proposal, and a proposed mechanism for implementation.

On November 29, 2018, the Commission issued Resolution E-4948, approving with modifications the dead band tolerance proposals of PG&E and the other IOUs and directing the IOUs to modify their proposals via supplemental compliance Advice Letter (AL) within 30 days of the effective date of the order.

This Tier 2 Advice Letter provides PG&E’s modified dead band tolerance proposal incompliance with Resolution E-4948, and a proposed mechanism for implementation.In the “Dead Band Tolerance Proposal” section below, substantive modifications to PG&E’s original deadband tolerance proposal are in double underline or strikeout font.

Dead Band Tolerance Proposal

PG&E proposes a two-part test for a dead band tolerance range (or equivalently, threshold to exceed) as the trigger for whether revised TOU periods could beconsidered sooner than 5 years after the most recent change in TOU period.Specifically, PG&E proposes that TOU period definitions should remain constant for at least 5 years unless both of the following conditions 1 and 2 are met:

1. Changed cost data justify changing either (a) the start or ending time of the TOU period by at least one hour (in either direction), for the summer peak, winter peak, or spring super-off-peak (SOP), or (b) the months for which particular TOU period definitions apply;4 and

2. Using a forecast of marginal generation costs (MGC), or whatever other marginal costs are used to determine TOU periods in a GRC Phase II proceeding,5

3 D.17-01-006, mimeo, p. 6.4 For example, if changing cost patterns justify modifying the months which are defined to be in

the summer season or, as an alternative example, modifying the months which are considered to be spring season (in which low super-off-peak rates apply).

5 The Commission directed PG&E in Resolution E-4948 (at p. 11) to “ensure that the marginal costs used as inputs in determining TOU periods in their GRC Phase II proceedings aremirrored in their deadband tolerance analysis going forward. Thus, because marginal transmission costs were not included as inputs for calculating the TOU periods determined inPG&E’s recent GRC Phase II decision (D.18-08-013), marginal transmission costs should also

11-AtchA-3

(PG&E-2)

Advice 5037-E-A - 3 - December 28, 2018

with the forecast year set at least three years after the year the Base TOU period will go into effect, the “goodness of separation” (GOS) metrics pertaining to the summer peak period, the winter peak period or the super-off-peak (SOP) period increase under the new TOU period definition by at least three five percentage points (35%)relative to the corresponding GOS metrics using the old TOU period definition.

Specifically, for the summer and winter peak period, the GOS metric is calculated as follows:

(1) Peak Period GOS Metric = A * (1 – B).6

In equation (1), A represents the percent of high-cost hours correctly captured by the peak period definition, and B represents the percentage of low-cost hours incorrectly captured by the peak period definition (i.e., the “false positive” rate for the peak period). Mathematically, A and B are defined in equations (2) and (3) below:7

(2) A = TP / (TP + FN), where

TP = the number of high cost hours (95th percentile and above) falling within candidate period; and

FN = the number of high-cost hours falling in other periods.

(3) B = FP / (FP + TN), where

FP = the number of low-cost hours within the candidate period; and

TN = the number of low-cost hours falling in other periods.

For the spring SOP period, the GOS metric is calculated as:

(4) SOP Period GOS Metric = A’ * (1 – B’). be excluded in the deadband tolerance analysis unless and until future GRC Phase II decisions incorporate them into TOU periods.”PG&E notes that in its 2017 GRC Phase II, the peak and super off-peak periods used only MGCs to inform the period definitions, while the part-peak definitions also considered distribution marginal costs. Thus, in accordance with the direction in E-4948, PG&E will consider only MGCs when evaluating whether the deadband tolerance has been exceeded, unless and until a future GRC Phase II decision incorporates distribution and/or transmission marginal costs to set peak and super off-peak periods.6 See Workpaper “2017 GRC Ph 2 Exh 2 Vol 1 Tables 12-2 to 12-4 12-6 to 12-8.xlsb”, tabs

“TESTIMONY TABLES-Summer” and “TESTIMONY TABLES-Winter” in PG&E’s 2017 GRC Phase II Application.

7 See Exh. PG&E-2, p. 12-13 in PG&E’s 2017 GRC Phase II Application

11-AtchA-4

(PG&E-2)

Advice 5037-E-A - 4 - December 28, 2018

In equation (4), A’ represents the percent of very low-cost hours captured by the SOP period definition, and B’ represents the percentage of non-very-low-cost hours incorrectly captured by the SOP period definition (i.e., the “false positive” rate for the SOP period). Mathematically, A’ and B’ are defined in equations (5) and (6) below:

(5) A’ = TP’ / (TP’ + FN’), where

TP’ = the number of very low-cost hours (i.e., hours with negative or zero MGCs) falling within the SOP period; and

FN’ = the number of very low-cost hours falling in other periods.

(6) B’ = FP’ / (FP’ + TN’), where

FP’ = the number of non-very-low-cost hours (i.e., hours with positive MGCs) within the SOP period; and

TN’ = the number of non-very-low-cost hours falling in other periods.

For example, suppose that, in PG&E’s 2017 General Rate Case (GRC) Phase II proceeding, the Commission finds that the summer TOU peak period definition for PG&E’s non-residential customers should be the hours from 4 p.m. through 10 p.m.,based on a forecast made for three years after the new TOU periods would go into effect (i.e., 2022). Further suppose that, in preparing its 2020 GRC Phase II testimony, PG&E determines that the then-existing forecast data support a somewhat later Base summer TOU peak period, from 5 p.m. through 10 p.m., based on a forecast made for three years after any such new TOU periods could go into effect (presumably 2023 or 2024).

PG&E would then check whether the potential new Base summer TOU peak period meets both conditions, as follows.

For condition 1, the start time of the proposed peak period differs by (at least) 1 hour from the old peak, so condition 1 is satisfied.

For condition 2, we compute the GOS metric for the old 4-10 p.m. peak period (using the updated marginal cost forecast) and compare it to the GOS of the potential new 5-10 p.m. peak period to see whether the GOS increases by at least 35%.

11-AtchA-5

(PG&E-2)

Advice 5037-E-A - 5 - December 28, 2018

As a hypothetical example, suppose the GOS metrics for the old and potential new peak periods using the updated marginal cost forecast turned out to be as follows:8

For Scenario S-17 (4-10 p.m.)A = TP/(TP + FN) = 9593%B = FP/(FP + TN) = 20%GOS = A*(1 – B) = 7674.4%

For Scenario S-26 (5-10 p.m.)A = TP/(TP + FN) = 94%B = FP/(FP + TN) = 15%GOS = A*(1 – B) = 79.9%

Because the GOS of the potential new 5-10 p.m. peak period is at least 35% higherthan the GOS of the old 4-10 p.m. period, condition 2 is satisfied. In this example, both criteria are met so PG&E should may propose the new Base summer TOU period in its 2020 GRC Phase II.

As a second example, suppose instead that for the old 4-10 p.m. period the GOS was the same as in the first example, but for the proposed 5-10 p.m. period the GOS turned out to be as follows:

For Scenario S-26 (5-10 p.m.)A = TP/(TP + FN) = 94%B = FP/(FP + TN) = 18%GOS = A*(1 – B) = 77%

In this second hypothetical example, the GOS of the proposed peak period is less than 35% higher than the GOS of the old period, so condition 2 is not satisfied and PG&E may not propose the new Base summer TOU period in its 2020 GRC Phase II.

A similar calculation would be made for the Winter Peak and the Super Off Peak separately; if the Summer Peak change satisfies both conditions but the Winter Peak and Super Off Peak do not, PG&E would only consider changes to the Summer Peak period definition.

PG&E believes its proposed dead band tolerance methodology is reasonable because of its requirement that the new TOU period must show at least a 35% improvement in Goodness of Separation, and must cause at least a full hour shift from the previously-adopted TOU Periods. PG&E’s approach is relatively conservative, to address the CPUC’s underlying concern, expressed in D.17-01-006, that “a degree of stability is

8 These metrics are similar to the data displayed in Table 12-4 in Exhibit PG&E-9 in PG&E’s

2017 GRC Phase II Application, but are numerically different to show an illustrative example that might apply three years hence.

11-AtchA-6

(PG&E-2)

Advice 5037-E-A - 6 - December 28, 2018

needed after new TOU periods are adopted, with the significant marketing efforts needed to make customers aware of changes in TOU periods,”9 preferring the assumed time-frame to maintain newly adopted TOU periods be at least five years (i.e., two GRC cycles). However, the Decision also recognized that forecast assumptions underlying TOU time periods may deviate over time as more up-to-date data becomes available.” To build in flexibility should the new data “deviate significantly,” the Decision allowed for the possibility that an “adjustment in TOU time periods more frequently than once every five years may be warranted.”10 Thus the CPUC stated that, in every GRC Phase IIproceeding, it would review forecast data for Base TOU period development, using the dead band tolerance methodology to be adopted through this Advice Letter process.

PG&E believes its proposed dead band methodology’s requirement of at least a 35%deviation, will reliably identify whether cost-data deviations are “significant” enough to warrant consideration of an interim revision in TOU time periods. Also, by requiring the deviation to cause at least a one-hour shift in the base TOU periods, PG&E’s proposed dead band methodology would not cause proposed changes of less than an hour, which would be more difficult to communicate to customers. As the Decision noted, the several-hour shift shown in the CAISO’s as well as PG&E’s data – warranting movingthe legacy Noon – 6pm peak period to the evening hours - reflects the steep increase in “deployment of grid-connected and behind-the-meter solar…[increasing] the availability of energy during the afternoon and decreased load on the grid.” Because these solar installations are long-lived, and because the hours during which the sun shines still end in the evening, the multi-hour shift in updated TOU peak periods being considered and adopted by the CPUC in proceedings like PG&E’s 2015 RDW (D.15-11-013) and in PG&E’s GRC Phase II (A.16-06-013) are unlikely to repeat.

Finally, even if the dead band were exceeded in a future GRC after the updated TOU peak periods are adopted, the Decision only allows parties to propose modifications to the TOU periods to align with costs, it does not require the IOUs to propose a change or for the CPUC to adopt it in that interim GRC. Therefore, if after considering a potential shift of one hour or more in an interim GRC Phase II the IOU or the CPUC still has concerns about stability, neither the Decision nor this dead band methodology requires the IOU to propose or the CPUC to actually adopt such a change sooner than “at least 5 years” after the last change.

For all of these reasons, the CPUC should approve PG&E’s revised dead bandmethodology included in this advice letter.

Timing and Implementation

As discussed above, the Decision directs the IOUs to propose changes to the guidelines to clarify the mechanics and timing of the dead band tolerance trigger and

9 D.17-01-006, mimeo, p. 46.10 D.17-01-006, mimeo, p. 47.

11-AtchA-7

(PG&E-2)

Advice 5037-E-A - 7 - December 28, 2018

that Base TOU period analysis should be provided in each general rate case (even if the IOU does not propose a change in Base TOU periods). Also, Appendix 3 of D.17-01-006 outlines an anticipated Schedule for TOU Period Implementation Based on Current Rate Case Plan.11

Protests

Pursuant to CPUC General Order 96-B, Section 7.5.1, PG&E hereby requests the protest period be waived.

Effective Date

PG&E requests that this Tier 2 advice submittal become effective on regular notice, January 27, 2019, which is 30 calendar days after the date of submittal.

Notice

In accordance with General Order 96-B, Section IV, a copy of this advice letter is being sent electronically and via U.S. mail to parties shown on the attached list and the parties on the service list for R.15-12-012. Address changes to the General Order 96-Bservice list should be directed to PG&E at email address PGETariffs@pge.com. For changes to any other service list, please contact the Commission’s Process Office at (415) 703-2021 or at Process_Office@cpuc.ca.gov. Send all electronic approvals to PGETariffs@pge.com. Advice letter filings can also be accessed electronically at: http://www.pge.com/tariffs/.

/S/Erik JacobsonDirector, Regulatory Relations

cc: Michael Campbell, Program Manager, ORAJeanne B. Armstrong, Counsel for SEIAEdward Randolph, Director CPUC Energy DivisionPaul Phillips, Supervisor, CPUC Energy DivisionService List for R.15-12-012

11 D.17-01-006, mimeo, Appendix 3, pp.1-2.

11-AtchA-8

(PG&E-2)

ADVICE LETTER S U M M A R YENERGY UTILITY

Company name/CPUC Utility No.:

Utility type:Phone #:

EXPLANATION OF UTILITY TYPE

ELC GAS

PLC HEAT

MUST BE COMPLETED BY UTILITY (Attach additional pages as needed)

Advice Letter (AL) #:

WATERE-mail:E-mail Disposition Notice to:

Contact Person:

ELC = ElectricPLC = Pipeline

GAS = GasHEAT = Heat WATER = Water

(Date Submitted / Received Stamp by CPUC)

Subject of AL:

Tier Designation:

Keywords (choose from CPUC listing):AL Type: Monthly Quarterly Annual One-Time Other:If AL submitted in compliance with a Commission order, indicate relevant Decision/Resolution #:

Does AL replace a withdrawn or rejected AL? If so, identify the prior AL:

Summarize differences between the AL and the prior withdrawn or rejected AL:

Yes No

Yes No

No. of tariff sheets:

Estimated system annual revenue effect (%):

Estimated system average rate effect (%):

When rates are affected by AL, include attachment in AL showing average rate effects on customer classes (residential, small commercial, large C/I, agricultural, lighting).

Tariff schedules affected:

Service affected and changes proposed1:

Pending advice letters that revise the same tariff sheets:

1Discuss in AL if more space is needed.

Pacific Gas and Electric Company (ID U39E)

(415)973-2094✔PGETariffs@pge.com

Yvonne.Yang@pge.com

Yvonne Yang

5037-E-A 2

Supplemental: Pacific Gas and Electric Company’s Proposed Dead Band Tolerance Range for determining when a change would trigger TOU period Revisions More Frequently than Five Year Intervals, in Compliance with Decision 17-01-006 and Resolution E-4948

Compliance✔

Resolution E-4948

No

✔

✔

1/27/19 N/A

N/A

N/A

N/A

N/A

Clear Form11-AtchA-9

(PG&E-2)

CPUC, Energy DivisionAttention: Tariff Unit505 Van Ness AvenueSan Francisco, CA 94102 Email: EDTariffUnit@cpuc.ca.gov

Protests and all other correspondence regarding this AL are due no later than 20 days after the date of this submittal, unless otherwise authorized by the Commission, and shall be sent to:

Name:Title:Utility Name:Address:City:State:Telephone (xxx) xxx-xxxx:Facsimile (xxx) xxx-xxxx:Email:

Name:Title:Utility Name:Address:City:State:Telephone (xxx) xxx-xxxx: Facsimile (xxx) xxx-xxxx:Email:

Zip:

Zip:

Director, Regulatory RelationsPacific Gas and Electric Company

77 Beale Street, Mail Code B13USan Francisco, CA 94177

Erik Jacobson, c/o Megan Lawson

California 94177(415)973-2093

(415)973-3582PGETariffs@pge.com

District of Columbia

Clear Form11-AtchA-10

(PG&E-2)

PG&E Gas and Electric Advice Filing List General Order 96-B, Section IV

Pioneer Community Energy PraxairRegulatory & Cogeneration Service, Inc. SCD Energy Solutions

SCESDG&E and SoCalGas

SPURRSan Francisco Water Power and Sewer

Downey & Brand

Ellison Schneider & Harris LLP Energy Management ServiceEvaluation + Strategy for Social InnovationGenOn Energy, Inc. Goodin, MacBride, Squeri, Schlotz & RitchieGreen Charge Networks Green Power Institute Hanna & Morton

Seattle City Light

ICFSempra Utilities

International Power Technology Southern California Edison Company

Intestate Gas Services, Inc. Southern California Gas Company

Kelly Group Spark Energy

Ken Bohn Consulting Sun Light & Power

Keyes & Fox LLP Sunshine Design

Leviton Manufacturing Co., Inc. Tecogen, Inc.

LindeTerraVerde Renewable Partners

Los Angeles County Integrated Waste Management Task Force

Tiger Natural Gas, Inc.

Los Angeles Dept of Water & Power TransCanada

MRW & Associates Troutman Sanders LLP

Manatt Phelps Phillips Utility Cost Management

Marin Energy Authority Utility Power Solutions

McKenzie & Associates Utility Specialists

Modesto Irrigation District Verizon

Morgan Stanley Water and Energy Consulting

NLine Energy, Inc. Wellhead Electric Company

NRG Solar Western Manufactured Housing Communities Association (WMA)

Office of Ratepayer Advocates Yep Energy

OnGrid Solar

AT&TAlbion Power CompanyAlcantar & Kahl LLP

Anderson & Poole

Atlas ReFuelBART

Barkovich & Yap, Inc.Braun Blaising Smith Wynne P.C. CalCom SolarCalifornia Cotton Ginners & Growers AssnCalifornia Energy CommissionCalifornia Public Utilities CommissionCalifornia State Association of CountiesCalpineCasner, SteveCenergy PowerCenter for Biological DiversityCity of Palo Alto

City of San Jose Clean Power Research Coast Economic Consulting Commercial Energy County of Tehama - Department of Public WorksCrossborder Energy Crown Road Energy, LLC Davis Wright Tremaine LLP Day Carter Murphy

Dept of General Services Don Pickett & Associates, Inc.Douglass & Liddell

Pacific Gas and Electric Company

11-AtchA-11

(PG&E-2)