Marginal Cost of Service

Transcription

1 UE 294 / PGE / 1300 Macfarlane Werner BEFORE THE PUBLIC UTILITY COMMISSION OF THE STATE OF OREGON UE 294 Marginal Cost of Service PORTLAND GENERAL ELECTRIC COMPANY Direct Testimony and Exhibits of Robert Macfarlane Bruce Werner February 12, 2015

2 Macfarlane- Werner Ii Table of Contents I. Introduction and Summary... 1 II. Generation Marginal Cost Study... 2 III. Distribution Marginal Cost Study... 6 IV. Customer Service Marginal Cost Study V. Area and Streetlights VI. Qualifications List of Exhibits... 16

3 Q. Please state your names and positions. I. Introduction and Summary UE 294 I PGE I 1300 Macfarlane - Werner I 1 2 A. My name is Robert Macfarlane. I am a senior analyst in Pricing and Tariffs for PGE. I 3 am responsible, along with Mr. Werner, for the development of the marginal cost 4 studies. 5 My name is Bruce Werner. I am an analyst in Pricing and Tariffs for PGE. I am 6 also responsible for the development of the marginal cost studies. 7 Our qualifications are included at the end of this testimony. 8 Q. What is the purpose of your testimony? 9 A. Our testimony describes the methodologies and results of PGE's generation, 10 distribution, customer service, and street lighting marginal cost studies. PGE 11 Exhibit 1301 provides a summary of these marginal costs by component. The summary 12 lists costs by PGE rate schedule for subtransmission, substation, feeder backbone and 13 tapline, transformers, service laterals, meters and customer service costs. Rate schedule 14 changes are discussed in PGE Exhibit Q. What is the purpose of the distribution and customer marginal cost studies? 16 A. The purpose is to calculate the incremental, or marginal unit cost of service for various 17 categories such as distribution substations and feeders, or billing. These unit costs, 18 expressed as costs per customer, costs per kilowatt (kw) of demand, or costs per 19 kilowatt hour (kwh) are then used to allocate the functional revenue requirements as 20 described in PGE Exhibit 1400.

4 II. Generation Marginal Cost Study UE 294 I PGE I 1300 Macfarlane - Werner I 2 Q. 2 A Q. 7 A. What methodology do you propose in this docket? We propose a long-run generation methodology that explicitly takes into account the cost of marginal generation capacity, long-run marginal energy costs, and renewable requirements. This methodology is similar to the long-run methodology stipulated to in UE 283. Please describe the methodology stipulated to in UE 283. In UE 283 we defined the long-run marginal generation resource as a combined cycle 8 combustion turbine (CCCT) for baseload purposes. We used the fixed costs of 9 an "F-class" simple cycle combustion turbine (SCCT) to estimate the portion of CCCT 10 fixed costs to assign to capacity. We estimated marginal energy costs using the Q. weighted values of the energy portion of the CCCT and a wind plant. We based the weightings on the renewable portfolio standard (RPS) requirements for each year. For example, the 2020 weighting is based on 20% wind and 80% thermal. Please describe the key differences between the study stipulated to in UE 283 and 15 the current study. 16 A. The first difference between the studies is that the study stipulated to in UE included environmental assumptions and the current study does not. We excluded the 18 projected costs of carbon dioxide compliance from the thermal plant and the federal 19 production tax credits from the wind plant. The second key difference is that we 20 included fixed gas transportation as a capacity cost for the SCCT in this study whereas 21 the study stipulated to in UE 283 assumed variable gas transportation.

5 Macfarlane- Werner I 3 Q. What type of SCCT did you use to estimate the marginal capacity costs? 2 A. Consistent with the methodology used to establish prices in the UE 283 stipulation, we 3 use an "F-class" SCCT. This unit has lower capital costs than the LMS 100 and 4 reciprocating engine units PGE presents in its recent Integrated Resource Plans (IRPs ). 5 Q. Please describe the steps used to develop the long-run generation allocation 6 methodology. 7 A. The generation marginal cost analysis involves the following inputs and steps: 8 1. Determine both a long-run marginal energy cost and a long-run marginal capacity 9 cost by first defining the marginal long-run generation resource as a CCCT used 10 for baseload purposes From this analysis, separately estimate the capacity and energy components as 12 follows: 13 a) Estimate the marginal cost of future capacity as the fixed cost of an "F-class" 14 SCCT. 15 b) Use these SCCT fixed costs as the portion of the CCCT fixed cost that is 16 assigned to capacity with the remaining CCCT fixed costs assigned to energy. 17 c) To the SCCT capacity costs add 12% reserve requirements consistent with 18 PGE's 2013 IRP Finally, express the capacity and energy values in real levelized terms. 20 Q. What are the sources of the overnight capital costs for the resources used in the 21 model? 22 A. For the CCCT, we use the Carty Generating Station values used in this filing. For the 23 SCCT, we obtained the location specific cost estimates for the SCCT from a publication

6 Macfarlane - Werner I 4 titled "Updated Capital Cost Estimates for Utility Scale Electricity Generating Plants." Q. 6 7 A Q. 11 A Q. This publication is sponsored by the United States Department of Energy. We include a copy of this publication in the Marginal Cost Work Papers. For the wind resource, we use the Tucannon River Wind Farm costs stipulated to in UE 283. Please describe how you determined the proportion of marginal energy costs attributable to the CCCT and the generic wind farm. We weighted the marginal energy cost by the RPS target percentages for each year. For example, if the RPS target is 20% in a given year, the weighting is 20% wind and 80% thermal. What is the source of your long-term gas price forecast? We used the long-term gas price forecast dated November 2014 for the Sumas and AECO hubs. We equally weighted the projected burnertip prices from these two hubs. Did you include the projected costs of carbon dioxide compliance in your analysis? 14 A. No. Any potential carbon tax is uncertain. We also assumed no production tax credit 15 for wind for the same reason. 16 Q. What is the fully allocated cost of the wind farm? 17 A. The cost of Tucannon River Wind Farm exclusive of wheeling is estimated at 18 $66.51/MWh in real levelized 2016 dollars, consistent with the capital costs in UE Q. How did you estimate each rate schedule's long-run marginal cost of energy? 20 A. We multiply each schedule's monthly on-peak and off-peak load forecast by the 21 corresponding monthly on-peak and off-peak long-term energy value.

7 Macfarlane- Werner I 5 Q. How do you shape the annual long-run marginal cost of energy into monthly 2 on-peak and off-peak values? 3 A. We shape the annual long-run marginal energy cost into monthly on-peak and off-peak 4 values based on the monthly on-peak and off-peak Mid-Columbia forward prices used 5 in PGE's production cost model, MONET. 6 Q. In UE 283, the Citizens' Utility Board proposed a methodology for incorporating 7 energy efficiency into the generation marginal cost study as an energy resource. 8 Does your study reflect energy efficiency as a marginal energy resource? 9 A. No. In the stipulation in UE 283, parties agreed to support the opening of an 10 investigatory docket to consider the appropriateness of including energy efficiency as a 11 marginal energy resource in light of the requirements of Senate Bill 838. PGE intends 12 to participate in that proceeding. We recognize that Commission determinations in that 13 proceeding could impact our generation marginal cost study results in this case.

8 Macfarlane - Werner I 6 III. Distribution Marginal Cost Study Q. A. Q. A. Which marginal distribution costs do you calculate? We calculate marginal distribution costs (separately) for subtransmission, substations, distribution feeders (backbone facilities and local facilities), line transformers (including services), and meters. How do you calculate the marginal unit costs of subtransmission and substations? We calculate subtransmission unit costs by first summing growth-related capital expenditures over the five-year period We then annualize these capital 8 expenditures and divide by the growth in system non-coincident peak. Customers 9 IO Q. A. served at subtransmission voltage are excluded from this calculation because they supply their own substation. We calculate substation marginal costs using a recent engineering estimate of the cost to construct a substation. Then we divide the cost by the substation transformer capacity in kw, and annualize the cost per kw. Columns (A) and (B) in PGE Exhibit 1301, summarize subtransmission and substation costs. How do you calculate the marginal unit feeder costs? We estimate distribution feeder unit costs in the following manner: 1. Perform an analysis that places customers on the distribution feeder from which they are currently served. 2. Eliminate any distribution feeders from which we cannot obtain customer information, and which do not conform to "typical" standards. Examples of these "non-typical" feeders are feeders serving customers at 4 kilovolt (Kv), or feeders that serve downtown core areas.

9 Macfarlane - Werner I 7 3. Perform an inventory of the wire types and sizes for each feeder. Standardize these 2 wire types and sizes to current specifications and then calculate the cost of 3 rebuilding these feeders in today's dollars Segregate the wire types and sizes into mainline feeders and taplines. Mainline 5 feeders are typically capable of carrying larger loads and are generally closer to the 6 substations from which they originate. Taplines are typically capable of carrying 7 smaller loads and can be remote from substations For each feeder, allocate the mainline cost responsibility of each rate schedule based 9 on the rate schedules' proportionate contribution to non-coincident peak (NCP). 10 Calculate a unit cost per kw by totaling the feeder cost responsibilities and dividing 11 by the sum of each schedule's NCP For each feeder, allocate the tapline cost responsibility of each rate schedule based 13 on the rate schedules proportionate design demand (estimated peak at the line 14 transformer). Calculate a unit cost per kw for both poly and single phase customers 15 by totaling the feeder cost responsibilities and dividing by the sum of each 16 schedule's design demand Annualize the mainline and tapline unit costs by applying an economic carrying 18 charge Separately estimate the unit costs of customers greater than 4 MW who are typically 20 on dedicated distribution feeders. Calculate these marginal unit costs (per 21 customer) as the average distance between the substation and the customer-owned 22 facilities. Finally, apply the annual carrying charge to annualize the cost per 23 customer.

10 Macfarlane - Werner I 8 9. Separately estimate the per-customer costs of customers served at subtransmission Q. A. Q. A. voltage. This is done by first calculating the average distance from the point at which subtransmission voltage customers connect into the subtransmission system from their substation. Then we multiply this average distance by the current cost per wire mile and annualize the costs. Columns (C) and (D) in PGE Exhibit 1301 summarize feeder mainline and tapline costs. Please describe any other considerations in calculating unit feeder costs. Currently, many municipalities require undergrounding of taplines within subdivisions and commercial areas. Therefore, we used the current cost of underground facilities exclusively in our marginal feeder tapline cost calculations. How do you calculate marginal transformer and service costs? We calculate each schedule's marginal transformer and service costs by estimating the cost of providing the average customer within specific load sizes with a service lateral and a line transformer (secondary delivery voltage only). Primary delivery voltage customers don't incur the cost of a transformer, but do incur the cost of the facilities necessary to interconnect them to the distribution feeder. Service and design costs, and any wire costs not captured in the feeder portion of the study are included in these estimates. For smaller customers such as those on Schedules 7 and 32, we estimate the average number of customers on a transformer in order to appropriately calculate the per customer share of transformer costs. Column (E) in PGE Exhibit 1301 summarizes transformer and service costs.

11 Q. Please describe how you calculate the marginal costs of meters. UE 294 I PGE I 1300 Macfarlane - Werner I 9 2 A. We calculate marginal meter costs as the weighted installed cost of an Advanced 3 Metering Infrastructure (AMI) meter for each rate schedule or load size, and then 4 apply an annual carrying charge. Column (F) in PGE Exhibit 1301, summarizes meter 5 costs. 6 Q. How do you allocate distribution operations and maintenance (O&M) to each 7 distribution category and ultimately to each rate schedule? 8 A. We allocate test-period distribution O&M by distribution category to the rate 9 schedules in proportion to each schedule's respective usage added to the per unit 10 marginal capital cost. All of the distribution costs by functional category in PGE 11 Exhibit 1301, Summary of Distribution and Customer Marginal Cost Studies, are 12 inclusive of test-period distribution O&M. 13 Q. The UE 283 Second Partial Stipulation required PGE to perform a 14 kilovolt-amperes reactive (KVAR) cost study and present the results at a pricing 15 workshop prior to filing this general rate case. Has PGE met this requirement? 16 A. Yes, PGE presented the results of the KVAR cost study at the January 20, pricing workshop to representatives of the Oregon Public Utility Commission Staff, 18 Kroger, and the Industrial Customers of Northwest Utilities. The results of the KVAR 19 study demonstrated that PGE's current reactive demand charges for large 20 nonresidential customers are appropriate, relative to the costs of mitigating reactive 21 power. The specific calculations of the study are contained in the PGE Exhibit work papers.

12 Macfarlane - Werner I 10 IV. Customer Service Marginal Cost Study Q. 2 A Q. 7 What is the purpose of the customer service marginal cost study? The purpose is to calculate the incremental cost of customer service to each customer class. PGE incurs costs in managing its relationship with customers, including handling customer communications, measuring usage, maintaining records, and billing. As such, customer service costs increase as the number of customers PGE serves increases. Does PGE use the forecasted test year expenses in the customer marginal cost study? 8 A. Yes. PGE uses forecasted costs for the 2016 test period and 2014 actual costs to 9 develop the 2016 test year customer marginal costs (CMC). These costs are found in FERC Accounts 902, 903, 905, 908, and 909. referred to as budget amounts in this testimony. The 2016 forecasted costs are also 12 Q. 13 A. 14 Is the study's methodology the same as in PGE's last rate case- UE 283? Yes, the methodology is the same. As in UE 283, the costs are allocated by PGE accounts directly on the basis of cost causation. A few accounts are allocated based on 15 a sub-allocation of the other account costs. After the costs are spread across rate Q. schedules, the final result is marginal costs for each rate schedule by each of the three functionalized categories: metering, billing, and other services. Does this cost study identify a similar amount of costs relative to the cost study 19 used in UE 283? 20 A. This cost study is improved by the inclusion of considerably more customer cost 21 categories as allocators than the previous cost study. Inclusion of the addional 22 categories affects the allocation percentages between the rate schedules. Schedule 89,

13 Macfarlane - Werner I 11 for example, is allocated a lower percentage of costs relative to other rate schedules 2 compared to UE 283. Examples of Customer Marginal Cost Calculations 3 Q. Please provide an example of how you calculate metering marginal costs. 4 A. The 2016 forecasted budget amount for FERC account 902, Field Collection 5 Department, is allocated based on manual meter reads and a weighted percentage of 6 customers (less unmetered lighting and signals). 7 Q. Please provide examples of how you calculate billing marginal costs. 8 A. Examples include: 9 The costs for Retail Receivables and Field Collections are allocated based on 10 percentage of adjusted write-offs by rate schedule. 11 Customer Information System billing costs are allocated by the number of 12 customers, except streetlights and traffic signals. 13 The costs for Printing and Automated Mail Services are allocated based on the 14 number of paper bills delivered. 15 Network Data Operation costs are allocated based on the number of customers with 16 meters, which excludes unmetered lighting and traffic signals. 17 Q. Please provide examples of how you calculate other consumer service marginal 18 costs. 19 A. Examples include: 20 The budget amount associated with the Customer Contact Operations is allocated by 21 the number of customers on rate schedules using up to 200 kw.

14 Macfarlane - Werner I 12 The budget amount for the Direct Access Operations Department is allocated by the 2 number of customers participating in the direct access program. 3 The budget amount for the Special Attention Operations Department is allocated 4 based on the number of residential customers. 5 The Solar Payment Option and Net Metering Operations budget amounts are 6 allocated by the number of customers participating in the programs.

15 V. Area and Streetlights UE 294 I PGE I 1300 Macfarlane - Werner I 13 Q. Please describe the changes you propose in the pricing of Area Lights and 2 Streetlights. 3 A. We propose to price the investment portion (poles and luminaires) of providing lighting 4 service using a real levelized annual revenue requirement rather than nominal. This 5 change reduces the investment cost by removing the effects of inflation and is 6 consistent with the methodology used in our other marginal cost studies. 7 Q. Please describe how you calculate the amount of outdoor lighting maintenance. 8 A. Similar to UE 283, we propose to base the test period lighting maintenance amount on 9 the incurred maintenance amounts during PGE's most recent 5-year re-lamping cycle 10 ( ). More specifically, we express the historical maintenance amounts on a 11 per-light basis, and then escalate this per-light maintenance figure for inflation. A 12 further reduction is made for Light-Emitting Diode (LED) street and area lights since 13 (1) their maintenance is significantly less than other lights, and (2) the years used in the 14 most recent 5-year re-lamping cycle do not include LEDs. Following this, we allocate 15 maintenance to each type of luminaire based on the marginal cost of maintenance 16 study. 17 Q. How do the maintenance amounts calculated in the marginal cost study compare 18 to the maintenance amounts calculated using the historical re-lamping cycle as a 19 base? 20 A. The amounts are close; the total amount of maintenance proposed for the 2016 test 21 period - based on the historical re-lamping cycle - is approximately $229,000 lower 22 than the amount calculated in the marginal cost study.

16 Macfarlane - Werner I 14 Q. Do you provide a summary exhibit of the proposed pole and luminaire prices? 2 A. Yes. This summary is provided in PGE Exhibit 1406.

17 Macfarlane - Werner I 15 VI. Qualifications Q. 2 A Q. 10 A Mr. Macfarlane, please state your educational background and experience. I received a Bachelor of Arts business degree from Portland State University with a focus in finance. Since joining PGE in 2008, I have worked as an analyst in the Rates and Regulatory Affairs Department. My duties at PGE have included pricing, revenue requirement, Public Utility Regulatory Policies Act avoided costs, and regulatory issues. From 2004 to 2008, I was a consultant with Bates Private Capital in Lake Oswego, OR, where I developed, prepared, and reviewed financial analyses used in securities litigation. Mr. Werner, please state your educational background and qualifications. I received a Bachelor of Arts degree with an emphasis in Fine Arts from Montana State University in Since joining PGE in 1999 I have worked as an analyst on a variety of pricing issues and cost studies in the Rates and Regulatory Affairs Department. From 1979 to 1999 I worked at PacifiCorp in several different capacities starting with Weatherization and Energy Efficiency programs and finishing as a Senior Cost of Service Analyst in their Rates and Regulatory Affairs Department. 16 Q. Does this conclude your testimony? 17 A. Yes.

18 Macfarlane- Werner I 16 List of Exhibits PGE Exhibit Description 1301 Marginal Cost Study

19 UE 294 I PGE I Exhibit 1301 Macfarlane - Werner Page 1 PORTLAND GENERAL ELECTRIC 2016 MARGINAL ENERGY COSTS Bus bar Schedule Energy {MWh} Schedule 7 8, 141,584 Schedule 15 17,425 Schedule 32 1,711,772 Schedule 38 41,909 Schedule 47 22, 135 Schedule49 67, 115 Schedule 83 2,992,243 Schedule ,000 kw 2,412,281 Schedule MW 967,084 Schedule 89 GT 4 MW 977,909 Schedule 90-P 1,578,439 Schedule 91/95 79,651 Schedule 92 3,465 Totals 19,013,012 Marginal Energy Cost $401,864,226 $792,164 $83,982,444 $2, 104,884 $1, 115,665 $3,268,684 $146,402,339 $118,430,943 $47,281,555 $47, 132,916 $75,881,616 $3,621,007 $166, 109 $932,044,551

20 UE 294 I PGE I Exhibit 1301 Macfarlane - Werner Page2 PORTLAND GENERAL ELECTRIC 2016 UNIT MARGINAL ENERGY AND CAPACITY COSTS Thermal SCCT Capacity Fixed Gas SCCT Transport VERBS Year $/kw-year $/kw-year $/kw-year Wind Thermal Wind Thermal Capacity Marginal Marginal Capacity wverbs Energy Energy $/kw-year $/kw-year $/MWh $/MWh Weighted Weighted Capacity Marginal Costs Energy RPS $/kw-year $/MWh 15.00% % % % % % % % % % % % % % % % % % % % Real Levelized $83.86 $40.60 $14.68 $ $ $43.26 $66.51 $ $47.97 NPV $1,078 Nominal Levelized $97.30 Real Levelized $83.86 $522 $189 $47.11 $17.03 $40.60 $14.68 $1,600 $1,788 $556 $855 $ $ $50.19 $77.17 $ $ $43.26 $66.51 $1,638 $617 $ $55.66 $ $47.97 Composite Income Tax Rate Property Tax Rate Inflation Rate Capitalization: Preferred Common All Equity Debt Cost of Capital 39.94% 1.50% 1.93% 0.00% 0.00% 0.00% 50.00% 9.75% 4.88% 50.00% 4.88% 50.00% 5.77% 2.89% 7.76% After-Tax Nominal Cost of Capital After-Tax Real Cost of Capital 6.61% 4.59%

21 VE 294 / PGE I Exhibit 1301 Macfarlane - Werner Page3 PORTLAND GENERAL ELECTRIC SUMMARY OF DISTRIBUTION AND CUSTOMER MARGINAL COST STUDIES FEEDER FEEDER TRANSFORMER SUBTRANSMISSION SUBSTATION MAINLINE TAPLINE &SERVICE METER CUSTOMER COSTS COSTS COSTS COSTS COSTS COSTS COSTS SCHEDULE ($/kw) ($/kw) ($/kw) ($/kw) ($/Customer) ($/Customer) ($/Customer) (A) (B) (C} (D) (E) (F) (G) Schedule 7 Residential Single-phase $12.38 $11.39 $23.97 $16.18 $73.98 $20.22 $68.88 Three-phase $12.38 $11.39 $23.97 $16.18 $ $57.47 $68.88 Schedule 15 Residential $12.38 $11.39 $24.76 $16.86 $5.44 N/A $68.24 Schedule 15 Commercial $12.38 $11.39 $24.76 $16.86 $5.44 N/A $54.45 Schedule 32 General Service Single-phase $12.38 $11.39 $27.91 $23.61 $ $18.32 $70.46 Three-phase $12.38 $11.39 $27.91 $9.43 $ $70.94 $70.46 Schedule 38 TOU Single-phase $12.38 $11.39 $34.05 $19.37 $ $52.41 $ Three-phase $12.38 $11.39 $34.05 $13.45 $ $ $ Schedule 47 Irrigation Single-phase $12.38 $11.39 $73.00 $49.64 $10.05 $57.42 $76.64 Three-phase $12.38 $11.39 $73.00 $25.88 $19.03 $81.34 $76.64 Schedule 49 Irrigation Single-phase $12.38 $11.39 $76.09 $32.76 $ $59.88 $ Three-phase $12.38 $11.39 $76.09 $26.05 $ $69.56 $ Schedule 83 Secondary General Service Single-phase $12.38 $11.39 $24.36 $19.94 $ $52.33 $ Three-phase $12.38 $11.39 $24.36 $8.96 $ $ $ Schedule 85 Secondary General ServicE $12.38 $11.39 $20.95 $6.84 $1, $ $ Schedule 85 Primary General Service $12.38 $11.39 $20.95 $6.84 $ $1, $ Schedule 85 Secondary 1-4 MW $12.38 $11.39 $21.35 $4.89 $4, $ $ Schedule 85 Primary 1-4 MW $12.38 $11.39 $21.35 $4.89 $ $1, $ ($/Customer) Schedule 89 Secondary GT 4 MW $12.38 $11.39 $85,119 NIA $13, $ $5, ($/Customer) Schedule 89 Primary GT 4 MW $12.38 $11.39 $85, 119 NIA $2, $1, $5, ($/Customer) Schedule 89 Subtransmission $12.38 N/A $86,451 NIA N/A $17, $5, ($/Customer) Schedule 90 Primary $12.38 $11.39 $269,070 NA $2, $1, $17, Schedules 91 & 95 Streetlighting $12.38 $11.39 $24.76 $16.86 $3.28 NIA $ Schedules 92 Traffic Signals $12.38 $11.39 $24.76 $9.16 $8.06 N/A $829.74