PACIFIC GAS AND ELECTRIC COMPANY 2020 GENERAL RATE … · (PG&E-2) 1-2 1 Metering (NEM)3 and...
Transcript of PACIFIC GAS AND ELECTRIC COMPANY 2020 GENERAL RATE … · (PG&E-2) 1-2 1 Metering (NEM)3 and...
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 [email protected]. For changes to any other service list, please contact the Commission’s Process Office at (415) 703-2021 or at [email protected]. Send all electronic approvals to [email protected]. 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✔[email protected]
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: [email protected]
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)[email protected]
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)