RUSSELL A. FEINGOLD

ComEd Ex. 8.0

STATE OF ILLINOIS

ILLINOIS COMMERCE COMMISSION COMMONWEALTH EDISON COMPANY Annual formula rate update and revenue requirement reconciliation under Section 16-108.5 of the Public Utilities Act.

: : : : :

No. 14-____

Direct Testimony of

RUSSELL A. FEINGOLD

Vice President Financial & Regulatory Services - Energy

Black & Veatch Corporation

On Behalf Of Commonwealth Edison Company

Docket No. 14-____ ComEd Ex. 8.0

Page i

TABLE OF CONTENTS Section Page

I. INTRODUCTION 1 A. Identification of Witness 1 B. Background and Qualifications 2 C. Purpose of Testimony 3 D. Summary of Conclusions 4

II. CONCEPTUAL FOUNDATION FOR ESTABLISHING A UTILITY’S UNBUNDLED SERVICES AND RATES 6 A. The Objectives of Service and Rate Unbundling 6 B. Costing Methods Utilized in the Utility Ratemaking Process 7

III. INDUSTRY-WIDE REVIEW OF THE RATEMAKING TREATMENT OF CUSTOMER SERVICE COSTS 14 A. Scope of the Review 14 B. Results of the Review 15 C. Implications of the Results of the Review for Ratemaking Purposes 21

IV. A REVIEW OF COMED’S CUSTOMER CARE COST STUDIES 22 A. ComEd’s Switching Study 22 B. An Alternative Approach 26

V. CONCLUSION 28


Page 1 of 29

I. INTRODUCTION 1

A. Identification of Witness 2

Q. What is your name and business address? 3

A. My name is Russell A. Feingold. My business address is 2525 Lindenwood Drive, 4

Wexford, Pennsylvania 15090. 5

Q. By whom and in what position are you employed? 6

A. I am employed by Black & Veatch Corporation (“Black & Veatch”) as a Vice President, 7

and I lead the Financial & Regulatory Services - Energy Practice of Black & Veatch’s 8

Management Consulting Division. 9

Q. Please describe the firm of Black & Veatch. 10

A. Black & Veatch has provided comprehensive engineering and management services to 11

utility, industrial, and governmental entities since 1915. Its Management Consulting 12

Division delivers management consulting solutions in the energy and water sectors. Our 13

services include broad-based strategic, regulatory, financial, and information systems 14

consulting. In the energy sector, Black & Veatch Management Consulting delivers a 15

variety of services for companies involved in the generation, transmission, and 16

distribution of electricity and natural gas. From an industry-wide perspective, Black & 17

Veatch has extensive experience in all aspects of the North American natural gas and 18

electric power industries, including utility costing and pricing, gas supply and transportation 19

planning, competitive market analysis and regulatory practices and policies gained through 20

management and operating responsibilities at gas and electric distribution utilities, 21

interstate gas pipeline companies, and other energy-related companies, and through a 22


Page 2 of 29

wide variety of client assignments. Black & Veatch has assisted numerous gas and 23

electric distribution companies located in the U.S. and Canada. 24

B. Background and Qualifications 25

Q. What has been the nature of your work in the utility consulting field? 26

A. I have over thirty-eight (38) years of experience in the utility industry, the last thirty-five 27

(35) years of which have been in the field of utility management and economic 28

consulting. Specializing in the utilities industry, I have advised and assisted utility 29

management, industry trade and research organizations, and large energy users in matters 30

pertaining to costing and pricing, competitive market analysis, regulatory planning and 31

policy development, gas supply planning issues, strategic business planning, merger and 32

acquisition analysis, corporate restructuring, new product and service development, load 33

research studies and market planning. In addition to my presentation of expert testimony 34

in utility regulatory proceedings that was just discussed, I have spoken widely on issues 35

and activities dealing with the pricing and marketing of gas and electric utility services. 36

Further background information summarizing my work experience, presentation 37

of expert testimony, and other industry-related activities is included in ComEd Exhibit 38

(“Ex.”) 8.01 to my direct testimony. 39

Q. Please describe your educational and professional background. 40

A. I received a Bachelor of Science Degree in Electrical Engineering from Washington 41

University in St. Louis and a Master of Science Degree in Financial Management from 42

Polytechnic Institute of New York University. 43


Page 3 of 29

Q. Have you ever testified before any utility regulatory commissions? 44

A. Yes. I have presented expert testimony before the Federal Energy Regulatory 45

Commission (“FERC”), the National Energy Board of Canada, and numerous state and 46

provincial regulatory commissions, including the Illinois Commerce Commission (the 47

“Commission”). My expert testimony has dealt with the costing and pricing of energy-48

related products and services for gas and electric distribution and gas pipeline companies. 49

In addition to traditional utility costing and rate design concepts and issues, my testimony 50

has addressed revenue decoupling concepts and other innovative ratemaking approaches, 51

gas transportation rates, gas supply planning issues and activities, market-based rates, 52

Performance-Based Regulation (“PBR”) concepts and plans, competitive market analysis, 53

gas merchant service issues, strategic business alliances, market power assessment, 54

merger and acquisition analyses, multi-jurisdictional utility cost allocation issues, inter-55

affiliate cost separation and transfer pricing issues, seasonal rates, cogeneration rates, and 56

pipeline ratemaking issues related to the importation of gas into the United States. 57

Q. On whose behalf are you appearing in this proceeding? 58

A. I am appearing on behalf of Commonwealth Edison Company (“ComEd”). 59

C. Purpose of Testimony 60

Q. What is the purpose of your direct testimony? 61

A. The purposes of my direct testimony are to: (1) present a conceptual foundation for 62

establishing a utility’s unbundled services and rates; (2) present and discuss the results of 63

Black & Veatch’s industry-wide review which identifies the cost allocation and 64

ratemaking treatment of customer service costs by electric utilities in states where electric 65


Page 4 of 29

deregulation has been approved and implemented; and (3) provide a recommendation to 66

the Commission as to how it should assign customer service costs to ComEd’s delivery 67

and supply functions. 68

D. Summary of Conclusions 69

Q. What are the conclusions of your direct testimony? 70

A. The main conclusions of my direct testimony are: 71

• A utility such as ComEd provides delivery services to all of its customers, 72

irrespective of whether they are provided with supply service from Retail 73

Electric Suppliers (“RESs”) or from ComEd. At the same time, the utility is 74

required to stand ready to provide supply-related services to all customers, 75

including those who may have previously been customers of a RES. 76

• From a conceptual perspective, if the level of customer service activity and 77

related costs incurred by ComEd cannot reasonably be expected to decrease 78

with customers switching to RESs, then such costs should continue to be 79

recovered through ComEd’s delivery rates for which all customers are 80

responsible. Likewise, only the costs of those customer service activities that 81

can be expected to decrease with fewer bundled service customers, or to 82

increase with greater bundled service customers, should be considered to 83

support ComEd’s supply function. 84

• The vast majority of states in which utility regulators have approved and 85

implemented electric deregulation on a full or partial basis has not assigned 86


Page 5 of 29

any customer service costs to the electric utilities’ competitive supply 87

function. 88

• The Commission should utilize the results of ComEd’s 2014 Updated 89

Switching Study (“Switching Study”) to reach the conclusion that all of its 90

customer service activities and related costs are necessary to support its 91

delivery function. It provides a reasonable basis to evaluate customer service 92

costs because it clearly demonstrates how such costs change over time with 93

changes in the level of customer switching. 94

• If the Commission decides to allocate a portion of ComEd’s customer service 95

costs to its supply function, the Commission should adopt an alternative 96

approach consisting of a two-step costing method that relies upon a 97

combination of embedded and avoided costing principles to allocate ComEd’s 98

customer service costs to its delivery and supply functions. 99

Q. Are there any exhibits attached to your direct testimony? 100

A. Yes. In addition to ComEd Ex. 8.01, ComEd Ex. 8.02 presents the results of Black & 101

Veatch’s industry-wide review which identifies the cost allocation and ratemaking 102

treatment of customer service costs by electric utilities in states where electric 103

deregulation has been approved and implemented. 104


Page 6 of 29

II. CONCEPTUAL FOUNDATION FOR ESTABLISHING A UTILITY’S 105 UNBUNDLED SERVICES AND RATES 106

A. The Objectives of Service and Rate Unbundling 107

Q. Please explain the objectives of rate unbundling for certain utility-related services 108

that are open to competition. 109

A. A utility that is required to separate or unbundle certain of its previously bundled services 110

should do so in a manner that achieves the following pricing objectives: (1) the pricing 111

approach should induce all suppliers to provide the competitive service and to establish 112

its associated rates and charges in the most economically efficient manner; (2) the pricing 113

approach should be neutral with respect to all competitors and should not impose any 114

restrictions or requirements that are not fair for all suppliers; and (3) the pricing approach 115

should provide the incumbent supplier with a fair opportunity to recover all of its 116

prudently incurred costs to provide the regulated service, including any incremental costs 117

of facilitating competitive alternatives. 118

Q. Can you please elaborate upon each of these objectives as they relate to the 119

unbundling of an electric utility’s delivery and supply functions? 120

A. Yes. The first objective related to economic efficiency means that the rate unbundling 121

process should avoid the unnecessary duplication of costs and avoid the unintended 122

creation of subsidies for alternative suppliers. If an alternative supplier cannot provide a 123

competitive service at a cost lower than the costs actually saved by the incumbent 124

supplier (by not providing that service), than these services are most efficiently provided 125

by a single firm. In addition, if the costs allocated to the service currently provided by 126

the incumbent supplier (based on the regulatory approved costing method) greatly exceed 127


Page 7 of 29

the cost reductions of the incumbent supplier (and of society at large) by providing less of 128

these services, the alternative suppliers’ ability to compete will effectively be subsidized 129

by the choice of costing method and the resulting prices of the incumbent for this service. 130

The second objective related to competitor neutrality means that the chosen 131

pricing approach should be fairly balanced, without unevenly applied restrictions, for the 132

incumbent supplier and all alternative suppliers so they can freely compete based on their 133

underlying business capabilities and their costs of doing business. 134

The third objective related to the recovery of costs for the incumbent supplier 135

means that the pricing approach should result in prices for the competitive and regulated 136

services that are reflective of the underlying costs of providing these services. To 137

achieve this objective, the pricing approach must provide the incumbent supplier with a 138

reasonable opportunity to fully recover its prudently incurred cost of service across all 139

services with changes in the level of the competitive services it provides over time. In 140

this way, the risk of stranded costs to the incumbent supplier will be minimized, which is 141

an appropriate consideration. 142

B. Costing Methods Utilized in the Utility Ratemaking Process 143

Q. Can you please explain the various types of costing methods available for use in 144

conjunction with the utility ratemaking process? 145


Page 8 of 29

A. Yes. There are four different types of costing methods that are available for use in 146

conjunction with the utility ratemaking process.1 They are: (1) embedded costing; (2) 147

direct costing; (3) marginal or incremental costing; and (4) standalone costing. As I 148

explain later in my direct testimony, each of these costing methods has its own purpose as 149

well as certain limitations, so care needs to be taken in recognizing these attributes when 150

determining their applicability to particular ratemaking situations. 151

Embedded costing refers to a utility’s historically-based cost of service (derived 152

from actual accounting costs), which includes a fair rate of return on the utility’s 153

investments and its various operating costs. Embedded costs in this sense represent the 154

costs incurred by the utility to provide the existing level of utility service. Embedded 155

costs are sometimes referred to as “fully loaded” or “fully distributed” costs because all 156

costs associated with a particular service or function are identified for pricing purposes. 157

From an economic perspective, embedded costs are referred to as “average costs.” 158

Direct costing – as the name suggests – refers to costs that can be specifically 159

identified and isolated in recognition of the fact that such costs are incurred exclusively to 160

support a particular service or function. The method assigns only those costs that are 161

directly related to the particular utility service or function. Shared costs or indirect costs 162

would not be allocated under a direct costing approach. 163

Marginal costing refers to the cost of producing one additional unit of output. In 164

other words, marginal cost is a measure of the resources required to provide an additional 165

1 In some jurisdictions, a combination of costing methods is used in the utility ratemaking process (e.g., Oregon uses an embedded cost method to establish a utility’s total revenue requirement and a Long-Run Incremental Cost (“LRIC”) method to evaluate the utility’s class revenue levels).


Page 9 of 29

increment of service or, alternatively, is a measure of the resources released or avoided if 166

there is a decrement in the level of service demanded. Incremental costing addresses the 167

practical difficulty in evaluating a one-unit change in output under marginal cost 168

principles by determining the change in costs over multiple units of output. Marginal or 169

incremental costing assigns all of the costs that vary with changes in output to the 170

particular service or function. 171

Finally, standalone costs refer to the costs incurred by an entity to provide a single 172

service or function on a separate basis. Standalone costing assigns all of the costs that a 173

service or function would require to be provided by an independent entity, which would 174

include costs that do not vary with output as well as the costs that do vary. 175

Q. Please explain how these concepts are reflected in the cost of service studies 176

prepared by gas and electric distribution utilities. 177

A. Embedded cost studies analyze the costs for a test period based on either the book value 178

of accounting costs (a historical period), the estimated book value of costs for a forecast 179

test year, or some combination of historical and future costs. Where a forecast test year is 180

used, the costs and revenues are typically derived from budgets prepared as part of the 181

utility’s financial plan. Typically, embedded cost studies are used to allocate the revenue 182

requirement between jurisdictions, classes, and between customers within a rate or 183

service class. 184

Marginal cost studies do not reflect actually incurred costs, but rely on estimates 185

of the expected changes in cost associated with changes in utility service. Marginal cost 186

studies are forward-looking to the extent permitted by available data. Marginal cost 187


Page 10 of 29

studies are particularly useful for rate design and can also be used as a guide to determine 188

how a utility’s total revenue requirement should be allocated to its classes of service. 189

Where it is important to send appropriate price signals associated with additional energy 190

consumption by customers, an understanding of marginal cost may be useful. 191

Q. How are these different types of costing methods relied upon by utilities in the 192

ratemaking process? 193

A. Embedded costing is used widely in the North American utilities industry in the 194

determination of a utility’s total revenue requirement, as a guide in the setting of class 195

revenues by rate or customer class, and for purposes of evaluating individual rate levels 196

and rate structures. Incremental costing is utilized more often as a guide in the 197

determination of rate structures, rate levels, and individual charges. Direct and 198

standalone costing methods are used much less frequently in the setting of utility rates, 199

but they nevertheless provide insights into how utilities and alternative suppliers incur 200

costs and how such costs should be recovered through their rates and prices. 201

Q. Can you please explain the applicability of each of these costing methods to the 202

establishment of a utility’s unbundled services and rates, and in particular, to the 203

allocation of customer service costs to ComEd’s delivery and supply functions? 204

A. Yes. To do this, however, it is first necessary to understand the general nature of 205

ComEd’s customer service activities and its delivery and supply functions as defined by 206

the specific activities that support these functions. 207

Q. What is the general nature of ComEd’s customer service activities? 208


Page 11 of 29

A. ComEd’s customer service activities reside within its Customer Operations area and 209

include Meter Reading, Field and Meter Services, Advanced Metering Infrastructure 210

(“AMI”) Implementation, Billing, Revenue Management, Revenue Protection, Cash 211

Processing, the Customer Contact Centers, Customer Relations, Demand Management, 212

Electric Supplier Services Department (“ESSD”) and Market Research. In addition, 213

these activities are supported by ComEd’s Information Technology, Support Services, 214

and Large Customer Solutions areas. The specific customer service activities are 215

described in detail in the direct testimony of Mr. Ronald E. Donovan. 216

Q. What is the nature of ComEd’s delivery and supply functions? 217

A. According to the Electric Service Customer Choice and Rate Relief Law of 1997, 218

Delivery services means those services provided by the electric utility that are necessary 219

in order for the transmission and distribution systems to function so that retail customers 220

located in the electric utility's service area can receive electric power and energy from 221

suppliers other than the electric utility, and shall include, without limitation, standard 222

metering and billing services.”2 223

Supply services - for retail customers taking bundled electric service, the 224

Company is obligated to procure all the component services the Company requires to 225

meet retail customer instantaneous electric power and energy requirements at any given 226

time under the Company’s tariffs, applicable tariffs on file with the FERC, and other 227

applicable law, including, without limitation, all required electric energy, energy to 228

2See Illinois Public Utilities Act, 220 ILCS 5/16-102.


Page 12 of 29

satisfy losses, electric generation capacity, volumetric risk management, transmission 229

services, ancillary transmission services, renewable energy resources, administrative 230

services, and other necessary services procured by the Company. Such obligations are 231

met by the Company through contractual arrangements and purchases in the PJM 232

Interconnection L.L.C., or its successor (PJM) administered markets or wholesale 233

electricity markets, as applicable.3 234

ComEd provides delivery services to all of its customers, irrespective of whether 235

they are provided with supply service from a RES or from ComEd. As a result, a 236

customer shifting to a competitive supply service provided by a RES should continue to 237

be responsible for the costs incurred by ComEd in providing delivery service. From a 238

conceptual perspective, if the level of customer service activity and related costs incurred 239

by ComEd cannot reasonably be expected to decrease with increased customer switching 240

to RESs, then such costs should continue to be recovered through ComEd’s delivery 241

service rates for which all customers are responsible. Likewise, only the costs of those 242

customer service activities that can be expected to decrease with fewer bundled service 243

customers, or to increase with greater bundled service customers, should be considered to 244

support ComEd’s supply function. 245

At the same time, the utility is required to stand ready to provide supply-related 246

services to all customers, including those who may have previously been customers of a 247

RES. This continuing utility supply obligation is known as the Provider of Last Resort 248

3 See Rider PE – Purchased Electricity.


Page 13 of 29

(“POLR”). Due to this requirement, the utility simply cannot avoid costs which support 249

its supply function since it is unable to eliminate that required utility function. 250

Q. With this as a backdrop, can you please continue with your explanation of the 251

applicability of the above-described costing methods to the establishment of a 252

utility’s unbundled services and rates? 253

A. Yes. With regard to embedded costs, its primary strength is that the method is widely 254

used in the industry to establish a utility’s total revenue requirement, as well as its class 255

revenues and rates. However, embedded costs cannot reflect a competitive price for the 256

unbundled services that are open to competition and will encourage inefficient entry into 257

the market by alternative suppliers. This limitation exists because of the inherent 258

inefficiencies associated with using embedded costing to set competitive rates, which 259

results in higher rates compared to the level of rates using avoided or incremental costing. 260

As a result, a subsidy will be provided to customers who choose to take an unbundled 261

service from the alternative supplier, at the expense of the utility’s remaining bundled 262

supply customers. In addition, use of embedded costs will not be able to reflect the true 263

savings to the utility for having the service provided by an alternative supplier, with 264

either the utility or its remaining bundled service customers bearing the burden of 265

increased rates caused by a reduction in utility revenues that is greater than the avoided 266

costs for the competitive service in question. 267

With regard to marginal or incremental costs, this costing method is the most 268

reasonable when consideration is given to an economically efficient basis for pricing, and 269

therefore, by extension, for unbundling a utility’s functions. However, this method is not 270


Page 14 of 29

strictly compatible with the basis upon which a utility’s total revenue requirement is 271

determined, which is through the use of embedded costs. In addition, many state utility 272

regulatory commissions, including this Commission, utilize embedded cost concepts as a 273

guide for setting a utility’s class revenues and rates. 274

This brief explanation highlights that there are certain tradeoffs between these 275

costing methods which need to be understood and weighed when selecting a particular 276

costing method that can properly recognize and reflect in utility rates both cost causative 277

principles and the integrated nature of the utility’s currently available bundled service 278

offerings. I revisit these tradeoffs later in my direct testimony when I present a review of 279

ComEd’s cost studies and how the Commission should consider these studies for the 280

purposes of allocating ComEd’s customer service costs to its delivery and supply 281

functions. 282

III. INDUSTRY-WIDE REVIEW OF THE RATEMAKING TREATMENT OF 283 CUSTOMER SERVICE COSTS 284

A. Scope of the Review 285

Q. What was the scope of the industry-wide review of the ratemaking treatment of 286

customer service costs conducted by Black & Veatch? 287

A. ComEd requested that Black & Veatch conduct a review of how the electric utilities 288

operating in states where electric deregulation has been approved and implemented treat 289

customer service costs for ratemaking and cost recovery purposes among the utilities’ 290


Page 15 of 29

delivery and supply functions. To guide our review, we selected particular states based 291

on their relative electric industry restructuring status.4 We included in our review: 292

• Those states where electricity prices are competitively determined for all retail 293

customers (i.e., both standard offer service and retail access). In these states, 294

retail access is permitted for all customers and electric utilities generally do 295

not own generation. 296

• Those states where competitively priced electricity is limited to retail access 297

customers. In these states, retail access is permitted on at least a limited basis. 298

Electricity prices for standard offer service customers remain regulated and 299

the electric utilities generally remain vertically integrated. 300

Based on these criteria, we conducted a review of the electric utility regulatory activities 301

in twenty-one (21) states, including the District of Columbia. 302

B. Results of the Review 303

Q. Have you prepared an exhibit which presents the results of Black & Veatch’s 304

industry-wide review? 305

A. Yes. ComEd Ex. 8.01 presents the results of Black & Veatch’s industry-wide review. 306

Q. Can you please present a summary of the results of Black & Veatch’s review? 307

A. Yes. Based on our review, we determined that the states in which utility regulators have 308

approved and implemented full or partial electric deregulation have treated customer 309

4 Regulatory Research Associates (part of SNL Energy) categorizes states that have undergone some level of electric deregulation according to three Tiers. Black & Veatch started with the states in Tiers I and II, and then researched them to confirm their current status with regard to electric deregulation.


Page 16 of 29

service costs for ratemaking purposes among the utilities’ delivery and supply functions 310

in the following ways: 311

1. No customer service costs are assigned to the electric utility’s competitive 312

supply function in the states of Arizona, California, Connecticut, Delaware, 313

Maine, Maryland, Massachusetts, Michigan, Montana, Nevada, New 314

Hampshire, New Jersey, Ohio, Oregon, Pennsylvania, Rhode Island, Virginia, 315

Washington, and the District of Columbia. 316

2. Some level of customer service costs is being assigned to the electric utility’s 317

competitive supply function in the state of New York. 318

3. The electric utility has eliminated its supply function as part of the 319

deregulation process in the state of Texas, so the treatment of customer 320

service costs related to the supply function is not addressed. 321

It should be noted that any administrative service costs incurred by the electric utility 322

related to its energy procurement portfolio for bundled service customers are not 323

considered to be customer service costs, based on the nature of ComEd’s customer 324

service activities. This particular definitional aspect of our review is detailed, where 325

appropriate, in the Explanatory Comments provided in ComEd Ex. 8.01. 326

Q. Please explain the method that was adopted to allocate customer service costs to the 327

supply function in the one state where some level of customer service costs was 328

assigned to the electric utility’s competitive supply function. 329


Page 17 of 29

A. In New York, the issue of service and rate unbundling was addressed in multiple 330

regulatory proceedings over many years5 in conjunction with a wide range of issues 331

related to the structure and implementation of electric deregulation. Importantly, the 332

initial regulatory proceedings had occurred during the early stages of retail access in New 333

York, where relatively few customers had chosen to receive supply service from an 334

alternative energy supplier. The New York Public Service Commission (“NYPSC”) 335

recognized that “retailing costs”6 associated with selling utility services were not easily 336

disaggregated and quantified by individual function since they are performed by the 337

utility for both supply and delivery services using common facilities and labor. 338

After much discussion and exploration of alternative costing methods occurred 339

among the participating parties to address the treatment of customer service costs and 340

other related costs, the NYPSC directed the utilities to structure their embedded cost of 341

service studies to use historic costs as a surrogate for the costs a utility would avoid in the 342

long-run if retailing functions were no longer performed by the utility for supply service. 343

At the same time, however, the NYPSC did acknowledge the limitations of such a costing 344

approach to sufficiently address the treatment of customer service costs.7 Based on this 345

5 On March 21, 2000, the NYPSC instituted a proceeding to determine, among other issues, the future role regulated utilities should play in providing electricity and natural gas in competitive markets (Case 00-M-0504, Proceeding on Motion of the Commission Regarding Provider of Last Resort Responsibilities, the Role of Utilities in Competitive Energy Markets, and Fostering the Development of Retail Competitive Opportunities). 6 “Retailing Costs” were defined as those costs that are directly related to maintaining a relationship with the customer for the purposes of providing all of the utilities’ services, both competitive and non-competitive. 7 In its Statement of Policy on Unbundling and Order Directing Tariff Filings in Case 00-M-0504 (dated August 24, 2004), the NYPSC noted that, “(we) asked the parties to begin with traditionally calculated


Page 18 of 29

costing method, the electric utilities in subsequent rate proceedings established separate 346

merchant function charges that were designed to recover a portion of the costs of their 347

customer and collections and bill processing and payment activities from their full service 348

or bundled customers. 349

Q. Did the issue of the unbundling of customer service costs arise in conjunction with 350

the utility regulatory activities that implemented electric deregulation in 351

Pennsylvania? 352

A. Yes. In Pennsylvania, the issue of service and rate unbundling also was addressed over a 353

number of years starting in May 2007 with the issuance of a Final Policy Statement by 354

the Pennsylvania Public Utility Commission (“PPUC”) on Default Service and Retail 355

Energy Markets.8 The resulting regulations addressed, among other things, the default 356

service cost elements that should be recovered in the Price-to-Compare (“PTC”) 357

charge(s).9 Aside from the wholesale costs of acquiring energy supplies, the regulations 358

identified the includable “Administrative Costs” as the utility activities of billing, 359

collection, education, regulatory, litigation, tariff filings, working capital, information 360

system and associated administrative and general expenses related to default supplier 361

service. However, the PPUC did not direct the electric utilities to immediately make the 362

required changes to their rates and charges to accommodate this requirement. Instead, 363

embedded cost of service studies, but we recognized that the application of traditional cost-causation principles, ordinarily used to allocate costs between customer classes and to design regulated rates, may not be of assistance in allocating costs between competitive and non-competitive services.” (page 16). 8 PPUC Docket No. M-00072009, Final Policy Statement, dated May 10, 2007. 9 52 Pa. Code § 69.1808 (a).


Page 19 of 29

the PPUC indicated its preference to defer addressing the further unbundling of services 364

and rates until the next distribution rate case for each electric utility. 365

In the ensuing years, the electric utilities addressed the issue in both distribution 366

rate cases and generation/energy procurement proceedings. Through the settlement 367

process, although the unbundling issue was raised by some parties, the only 368

administrative costs of generation/energy that were recovered through the PTC (i.e., 369

generation supply charges) were costs associated with the procurement and management 370

of the energy portfolio, with no customer service costs allocated to the competitive 371

supply function. 372

In more recent times, the PPUC once again addressed the service and rate 373

unbundling issue in an investigative proceeding of Pennsylvania’s Retail Electricity 374

Market: End State of Default Services. However, in its Final Order,10 the PPUC declined 375

to specifically address the unbundling issue (at the urging of some parties) on either a 376

generic basis or to “promulgate regulations requiring further unbundling as we believe 377

these measures would be a significant undertaking and require the time and resources of 378

many stakeholders.”11 379

Q. Do you have any views on how the unbundling issue related to customer service 380

costs was addressed in New York and Pennsylvania? 381

10 PPUC Docket No. I-2011-2237952, Final Order (dated February 14, 2013). 11 PPUC Docket No. I-2011-2237952, Final Order, at page 21.


Page 20 of 29

A. Yes. In New York, I believe the lengthy regulatory process that was followed to 382

determine an appropriate costing method for the treatment of customer service costs 383

occurred because of the NYPSC’s desire to address a number of competing 384

considerations, and that coming to a reasoned and balanced result was not an easy 385

process.12 386

In Pennsylvania, while there are specific regulations specifying the costs 387

characterized as “Administrative Costs” that should be recovered in the electric utilities’ 388

PTC charges, the PPUC chose not to be prescriptive in their application. In its previously 389

referenced Final Policy Statement, the PPUC stated that, “we are open to the concept of 390

addressing the allocation of costs between generation and distribution rates through a 391

collaborative process. We further agree that cost allocation should reflect the level of 392

service, or lack of service, provided to default service and non-default service 393

customers.”13 As I pointed out above, however, the inclusion of customer service costs 394

in the electric utilities’ competitive supply charges never occurred as an outcome of the 395

settled distribution rate cases or generation/energy procurement proceedings. 396

12 In In its Statement of Policy on Unbundling and Order Directing Tariff Filings in Case 00-M-0504 (dated August 24, 2004), the NYPSC stated that, “(it) does seem clear, however, that statutory obligation-to-serve costs, including a portion of what the parties called ‘retailing costs,’ should be borne by all customers on the network. This suggests that appropriate cost allocations include assigning some portion of such costs to both the competitive and non-competitive functions. The uncertainties associated with differentiating unavoidable obligation-to-serve costs from costs avoidable only in the long-run also suggest that there may be no single right answer, but a range of reasonable answers depending on how the parties’ varied interests are otherwise balanced.” (page 17). 13PPUC Docket No. M-00072009, Final Policy Statement, dated May 10, 2007, page 9.


Page 21 of 29

C. Implications of the Results of the Review for Ratemaking Purposes 397

Q. Based on the results of Black & Veatch’s industry-wide review of the ratemaking 398

treatment of customer service costs, do you have any comments on how the results 399

should be interpreted for purposes of allocating customer service costs to ComEd’s 400

delivery and supply functions? 401

A. Yes. Clearly, the vast majority of regulatory commissions in states where electric 402

deregulation has occurred have decided that no customer service costs should be assigned 403

to the electric utility’s competitive supply function. And even in those few states where 404

some level of customer service costs either is allocated, or is being considered for 405

allocation, to the electric utility’s competitive supply function, there was by no means 406

unanimous agreement among the parties on the particular activities, and the manner in 407

which the costs of these activities should be determined, for purposes of unbundling these 408

activities from the utility’s delivery function with recovery through its supply-related 409

charges. 410

I believe this situation occurred because of concerns over the significantly 411

different rate impacts to the various parties in those regulatory proceedings inherent in 412

the choice of a particular cost allocation method. Also affecting the ultimate resolution 413

of the unbundling issue was the NYPSC’s strong desire at that time to foster the growth 414

of retail access in New York. Very simply, if the utilities’ competitive supply rates were 415

set too high using a cost allocation method that was too generalized in its treatment of 416

customer service costs, the rates for non-competitive services like delivery service would 417

be set too low, resulting in greater revenue shortfalls to the utility and eventually higher 418

utility rates than would otherwise be necessary. Conversely, if the utilities’ competitive 419


Page 22 of 29

supply rates were set too low using a cost allocation method that was too narrowly 420

focused in its assessment of customer service costs, the rates for non-competitive services 421

would be set too high. This resulted in claims from some parties that customers would be 422

overpaying for customer service activities because the costs of certain of these activities 423

would be reflected in both the utility’s delivery rates and the alternative supplier’s 424

competitive supply charges. Finally, recognizing that New York and Pennsylvania were 425

at the early stages of retail access and electric deregulation, some parties also claimed 426

that without full unbundling of customer service costs, the competitive marketplace for 427

supply services would not develop. 428

In my view, balance is essential in resolving this unique ratemaking issue. This is 429

no different than how other utility ratemaking issues are addressed. While any one of the 430

utility costing methods traditionally relied upon by utilities and regulators to set rates 431

provides an objective measure of the cost basis for rates, the reality is that non-cost 432

factors are often considered when determining rates for the utility that are deemed to be 433

just and reasonable. 434

IV. A REVIEW OF COMED’S CUSTOMER CARE COST STUDIES 435

A. ComEd’s Switching Study 436

Q. Have you had an opportunity to review ComEd’s updated allocation and switching 437

studies presented in this filing? 438

A. Yes, I have. I am familiar with the overall structure, conceptual underpinnings, 439

operational grounding, and input data sources of ComEd’s updated allocation and 440

switching studies that are presented by Mr. Ronald E. Donovan (ComEd Exs. 7.04 and 441


Page 23 of 29

7.05). I also understand the basis for ComEd’s selection of cost allocation methods and 442

its derivation of cost allocation factors in both cost studies. Finally, I have reviewed 443

ComEd’s earlier versions of these customer care cost studies that were filed in a previous 444

rate proceeding. 445

Q. Based on your familiarity with both of ComEd’s customer care cost studies, which 446

study do you recommend the Commission should rely upon for purposes of 447

determining how customer service costs should be allocated to ComEd’s delivery 448

and supply functions? 449

A. I recommend that the Commission rely upon ComEd’s Switching Study to determine 450

how customer service costs should be allocated to ComEd’s delivery and supply 451

functions. On that basis, I also believe the Commission should conclude that all of the 452

costs of ComEd’s customer service activities are necessary to support its delivery 453

function. 454

Q. Why do you believe ComEd’s Switching Study should be relied upon by the 455

Commission for this purpose? 456

A. ComEd’s Switching Study is a direct means of determining whether customer service 457

costs are inherently related to delivery service, or to supply service. Very simply, the fact 458

that the costs of a particular customer service activity do not change with changes in the 459

level of switching indicates that the activity is required to meet ComEd’s ongoing 460

requirement to provide delivery service for all of its customers. This effectively means 461

that the associated costs of this activity are fixed in nature, relative to ComEd’s supply 462

function, because they do not vary with changes in the number of customers electing to 463


Page 24 of 29

have RES-provided supply service. The underlying avoided costing concepts in 464

ComEd’s Switching Study serve to establish the important one-to-one relationships 465

between the costs incurred and the levels of service that help to define which specific 466

activities are necessary to support each utility function. 467

Q. Why is an embedded costing method not your preferred approach to unbundle 468

ComEd’s traditional bundled utility service into its individual delivery and supply 469

components? 470

A. I believe there are fundamental limitations in using an embedded costing method to 471

unbundle traditional bundled utility service into its individual delivery and supply 472

components, because the method assumes inappropriately that all of ComEd’s customer 473

service costs are variable in nature. ComEd’s 2014 Updated Allocation Study 474

(“Allocation Study”) assumes from the outset that a portion of each customer service cost 475

category that cannot be directly assigned to either the delivery or supply utility function 476

will not be attributable to the utility’s delivery function. 477

In addition, an embedded costing method is not able to properly reflect a utility’s 478

economies of scale and scope in the resulting rates, which is an appropriate consideration 479

when establishing economically efficient prices for utility services. ComEd’s Allocation 480

Study does not attempt to discern the actual costs attributable to the basic or core delivery 481

service provided by ComEd. This important step is accomplished in ComEd’s Switching 482

Study by starting from the utility’s existing bundled service base and existing systems 483

and examining changes in the level of service provided. Scope economies are accounted 484

for by including only costs that vary directly with the service being provided. For 485


Page 25 of 29

example, if a cost is truly common to both a utility’s delivery and supply functions, it will 486

not be avoided if the service level is reduced for only one service, but it may be avoided 487

for a reduced level of service for a combined package of these two services. Scale 488

economies are accounted for by considering the level of business lost or gained and 489

examining the avoided or incremental cost relative to that new reduced or increased 490

service level. It would be difficult to account for such cost changes using an embedded 491

costing method that is not tied to the actual amount of service that will be lost. 492

Q. Can you provide an example of these limitations inherent in using an embedded 493

costing method, which ComEd’s Allocation Study is premised upon, to unbundle 494

customer service costs between its delivery and supply functions? 495

A. Yes. For example, to allocate the costs of bill printing activities in its Allocation Study, 496

ComEd measured the surface area of the bill dedicated to its delivery and supply charges. 497

Based on that assessment, approximately 83% of the cost of ComEd’s bill printing 498

activity was allocated to the delivery function and the remaining 17% of the costs were 499

allocated to the supply function. In reality, however, ComEd would not realize any 500

savings (i.e., avoided costs) for delivery service because bills for this required utility 501

service still must be printed even if 100% of its customers chose to receive supply service 502

from RESs. 503

Q. Since ComEd’s Switching Study is based on avoided cost concepts, isn’t it 504

incompatible with the way in which ComEd’s total revenue requirement is 505

determined using embedded costs? 506


Page 26 of 29

A. No. The starting point for both ComEd’s Switching Study and its Allocation Study is 507

identical, which is the customer service costs from the pool of costs for 2013. While 508

ComEd utilizes an avoided costing method to evaluate the nature of customer service 509

activities and to assign the related costs to its delivery and supply functions, the 510

composition and level of costs included in ComEd’s Switching Study is determined on an 511

embedded cost basis. 512

Q. Could the level of switching actually experienced by ComEd at the time it prepares 513

the Switching Study impact the results? 514

A. No. In ComEd’s Switching Study (which is premised upon the current switching level of 515

69%), the results indicate that there is no net reduction in customer service costs 516

associated with customer switching. In fact, at the 100% level, costs are expected to 517

increase based on 2013 expenses. In ComEd’s previous Switching Study filed in Docket 518

No. 10-0467 (which was premised upon a switching level of 1%), the results were similar 519

based on 2009 expense levels, thereby demonstrating a consistency of results over time. 520

B. An Alternative Approach 521

Q. Do you have an alternative costing method that the Commission should consider if it 522

believes that from a policy perspective some level of ComEd’s customer service costs 523

should be allocated to its supply function? 524

A. Yes. If the Commission believes that from a policy perspective some level of ComEd’s 525

customer service costs should be allocated to its supply function, then it should consider 526

adoption of the following two-step costing method as an alternative approach to 527

allocating ComEd’s customer service costs to its delivery and supply functions: 528


Page 27 of 29

1. Utilize ComEd’s current Switching Study (which is based upon an avoided 529

costing method) to identify the specific customer service activities that are 530

most sensitive to changes in the level of customer switching between ComEd 531

and RESs and the related costs that are avoided by ComEd when customers 532

switch to RESs. 533

2. Utilize ComEd’s Allocation Study (which is based upon an embedded costing 534

method) to quantify the portion of customer service costs to be allocated to 535

ComEd’s supply function for the specific customer service activities that are 536

most sensitive to changes in the level of switching. 537

In my opinion, a cost analysis such as ComEd’s Switching Study – which focuses on 538

evaluating changes in costs with changes in the level of utility service – most effectively 539

identifies the utility activities that are inherently related to the utility’s delivery and/or 540

supply functions. As I discussed previously, this type of cost analysis establishes 541

important one-to-one relationships between costs and levels of service that help to define 542

which specific activities are necessary to support each utility function. Once the specific 543

activities and related costs have been identified that are truly common to both ComEd’s 544

delivery and supply functions, its Allocation Study will quantify the portion of these 545

common costs that should be attributed to the delivery and supply functions, respectively. 546

Q. Can you please provide an example of how this two-step process would be used for 547

treating the costs of a specific customer service activity? 548

A. Yes. ComEd’s Switching Study (ComEd Ex. 7.05) identifies one particular 549

department/activity in which costs decrease (i.e., are avoided) with changes in the 550


Page 28 of 29

assumed level of switching from current levels to higher levels. If we examine the 551

Customer Contact Center activity (Line No. 6) in this exhibit, it shows that the costs for 552

this activity are expected to decrease at a 100% switching level. If we now refer to 553

ComEd’s Allocation Study (ComEd Ex. 7.04), the allocation of Customer Contact Center 554

costs to ComEd’s delivery and supply functions is indicated on Line No. 6, with 555

approximately $4.7 million allocated to the supply function as indicated under Col. (c). 556

Q. Has ComEd prepared an analysis that reflects this alternative costing method for 557

allocating customer service costs to its delivery and supply functions? 558

A. Yes. ComEd Ex. 7.06, which accompanies Mr. Donovan’s direct testimony, presents the 559

results of the allocation of customer service costs to ComEd’s delivery and supply 560

functions using the alternative costing method I have described above. 561

V. CONCLUSION 562

Q. Do you have any concluding remarks related to ComEd’s proposed allocation of 563

customer service costs to its delivery and supply functions? 564

Q. Yes. In my opinion, the unbundling of the costs of a utility’s customer service activities 565

between its delivery and supply functions is not an easy or straightforward undertaking, 566

and is not generally undertaken elsewhere in the electric utility industry. Based on the 567

results of ComEd’s Switching Study (which reflects a significant increase over the last 568

few years in the number of customers embracing retail access and customer choice), I 569

recommend that the Commission accept those results and conclude that all of ComEd’s 570

customer service activities and related costs are necessary to support its delivery function. 571

I believe this approach achieves the right balance between the competing objectives of 572


Page 29 of 29

protecting customers from unfair pricing, providing ComEd with a reasonable 573

opportunity to recover its costs, and continuing to promote supply-related competition. 574

However, if the Commission finds that a portion of the costs related to ComEd’s 575

customer service activities should be allocated to the supply function for policy reasons, 576

then the two-step costing method I present as an alternative approach attempts to 577

recognize the common cost aspect associated with ComEd’s customer service activities 578

and the desire to quantify the portion of the associated costs that can be allocated to 579

ComEd’s supply function using acceptable and widely recognized costing principles. 580

Q. Does this complete your direct testimony? 581

A. Yes. 582

RUSSELL A. FEINGOLD

Documents