BEFORE THE MINNESOTA PUBLIC UTILITIES COMMISSION. Beverly Jones Heydinger

Transcription

1 BEFORE THE MINNESOTA PUBLIC UTILITIES COMMISSION Beverly Jones Heydinger Nancy Lange Dan Lipschultz Matthew Schuerger John A. Tuma Chair Commissioner Commissioner Commissioner Commissioner In the Matter of the Application of CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas for Authority to Increase Natural Gas Rates in Minnesota ISSUE DATE: June 3, 2016 DOCKET NO. G-008/GR FINDINGS OF FACT, CONCLUSIONS, AND ORDER Contents PROCEDURAL HISTORY...1 I. Initial Filings and Orders...1 II. The Parties and Their Representatives...1 III. Proceedings Before the Administrative Law Judge...2 IV. Public Comments...2 V. Proceedings Before the Commission...3 FINDINGS AND CONCLUSIONS...3 I. The Ratemaking Process...3 A. The Substantive Legal Standard...3 B. The Commission s Role...3 C. The Burden of Proof...4 II. Rate Case Overview...5 III. Summary of the Issues...5 IV. The Administrative Law Judge s Report...8 V. Nicollet Mall Headquarters...8 A. Introduction...8 B. Positions of the Parties The Department The OAG...9 i

2 3. The Company...10 C. The Recommendation of the Administrative Law Judge...10 D. Commission Action...11 VI. PRIME Project Expenses...12 A. Introduction...12 B. Positions of the Parties...12 C. The Recommendation of the Administrative Law Judge...13 D. Commission Action...13 VII. Environmental Costs...13 A. Introduction...13 B. Positions of the Parties...13 C. The Recommendation of the Administrative Law Judge...14 D. Commission Action...14 VIII. Plant Retirements...15 A. Introduction...15 B. Positions of the Parties The OAG The Company...16 C. The Recommendation of the Administrative Law Judge...16 D. Commission Action...16 IX. Distribution-Plant Balances...17 A. Introduction...17 B. Positions of the Parties The Department The Company...17 C. The Recommendation of the Administrative Law Judge...17 D. Commission Action...17 X. Cash Working Capital...18 A. Introduction...18 B. Positions of the Parties...18 C. The Recommendation of the Administrative Law Judge...18 D. Commission Action...18 XI. Non-Qualified Pension Expense...18 A. Introduction...18 B. Positions of the Parties...19 ii

3 C. The Recommendation of the Administrative Law Judge...19 D. Commission Action...19 XII. Inflation Factors...19 A. Introduction...19 B. Positions of the Parties The OAG The Company...20 C. The Recommendation of the Administrative Law Judge...20 D. Commission Action...20 XIII. Incentive Compensation...21 A. Introduction...21 B. Positions of the Parties The Company The Department...22 C. The Recommendation of the Administrative Law Judge...23 D. Commission Action Long-Term Incentive Short-Term Incentive Short-Term Incentive Refund Mechanism and Annual Reporting ALJ Report Modification and Future Rate-Case Filing Requirements...24 XIV. Executive Compensation...25 A. Introduction...25 B. Positions of the Parties The OAG The Company...25 C. The Recommendation of the Administrative Law Judge...25 D. Commission Action...25 XV Rate-Case Expenses...26 A. Introduction...26 B. Positions of the Parties...26 C. The Recommendation of the Administrative Law Judge...26 D. Commission Action...26 XVI. American Gas Association Dues...27 A. Introduction...27 B. Positions of the Parties...27 iii

4 1. The OAG The Company...27 C. The Recommendation of the Administrative Law Judge...27 D. Commission Action...27 XVII. United Way Donation...28 A. Introduction...28 B. Positions of the Parties...28 C. The Recommendation of the Administrative Law Judge...28 D. Commission Action...28 XVIII. Property Taxes...29 A. Introduction...29 B. Positions of the Parties...29 C. The Recommendation of the Administrative Law Judge...29 D. Commission Action...29 XIX. Capital Structure...30 A. Introduction...30 B. Positions of the Parties The Company The Department The OAG...33 C. The Recommendation of the Administrative Law Judge...34 D. Commission Action The Equity Ratio Short-Term and Long-Term Debt Ratios...36 XX. Return on Equity...38 A. Introduction...38 B. The Analytical Tools...38 C. Positions of the Parties The Company The Department The OAG...40 D. The Recommendation of the Administrative Law Judge...42 E. Commission Action...43 XXI. Cost of Long-Term Debt...44 XXII.Cost of Short-Term Debt...45 iv

5 A. Introduction...45 B. Positions of the Parties The Company The Department...45 C. The Recommendation of the Administrative Law Judge...46 D. Commission Action...46 XXIII. Final Capital Structure and Overall Cost of Capital...47 XXIV. Class-Cost-of-Service Study...48 A. Background...48 B. Distribution System Summary The Classification Methods The Recommendation of the Administrative Law Judge Commission Action...53 C. Sales Expenses The Issue The Recommendation of the Administrative Law Judge Commission Action...55 D. Regulatory-Commission Expenses The Issue The Recommendation of the Administrative Law Judge Commission Action...55 E. Treatment of Historical Cost Data on Distribution Mains The Issue The Recommendation of the Administrative Law Judge Commission Action...56 F. Income Taxes The Issue The Recommendation of the Administrative Law Judge Commission Action...57 XXV. Class Revenue Apportionment...57 A. Introduction...57 B. Positions of the Parties The Department The OAG...58 v

6 3. The Company...59 C. The Recommendation of the Administrative Law Judge...59 D. Commission Action...60 XXVI. Customer Charges...61 A. Introduction...61 B. Positions of the Parties The Company The Department The OAG Fresh Energy SRA...63 C. The Recommendation of the Administrative Law Judge...64 D. Commission Action...64 XXVII. Future Rate Case Requirements...65 A. Calendar-Year Test Year Introduction Positions of the Parties The Recommendation of the Administrative Law Judge Commission Action...66 B. Sales Forecast and Weather Normalization The Issue Commission Action...67 C. Interim-Rate-Design Methodology...67 D. Formatting Financial Information...67 XXVIII. Overall Financial Schedules...68 A. Gross Revenue Deficiency...68 B. Rate Base Summary...68 C. Operating Income Summary...69 XXIX. Compliance Filing Required...71 ORDER...71 vi

7 BEFORE THE MINNESOTA PUBLIC UTILITIES COMMISSION Beverly Jones Heydinger Nancy Lange Dan Lipschultz Matthew Schuerger John A. Tuma Chair Commissioner Commissioner Commissioner Commissioner In the Matter of the Application of CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas for Authority to Increase Natural Gas Rates in Minnesota ISSUE DATE: June 3, 2016 DOCKET NO. G-008/GR FINDINGS OF FACT, CONCLUSIONS, AND ORDER I. Initial Filings and Orders PROCEDURAL HISTORY On August 3, 2015, CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas (CenterPoint or the Company) filed this general rate case seeking an annual rate increase of $54,100,000, or approximately 6.4% per year. The filing included a proposed interim rate schedule. On the same date, the Company filed a petition to establish a new base cost of gas for the period during which interim rates would be in effect; that petition was granted by order dated September 22, Also on September 22, 2015, the Commission issued three orders in this case: an order finding the rate-case filing substantially complete and suspending the proposed final rates; a notice and order for hearing referring the case to the Office of Administrative Hearings for contested-case proceedings; and an order setting interim rates for the period during which the rate case was being resolved. II. The Parties and Their Representatives The following parties appeared in this case: 1 In the Matter of the Application of CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas to Establish a New Base Cost of Gas and Reset the Purchased Gas Adjustment to Zero, to Coincide with the Implementation of Interim Rates in General Rate Case Filing, Docket No. G-008/MR , Order Setting New Base Cost of Gas (September 22, 2015). 1

8 CenterPoint Energy Minnesota Gas, represented by Eric F. Swanson and David M. Aafedt, Winthrop & Weinstine, P.A. Minnesota Department of Commerce (the Department), represented by Julia E. Anderson and Peter E. Madsen, Assistant Attorneys General. Office of the Minnesota Attorney General Residential Utilities and Antitrust Division (OAG), represented by Ryan P. Barlow and Joseph C. Meyer, Assistant Attorneys General. Suburban Rate Authority (SRA), represented by James M. Strommen, Kennedy & Graven, Chartered. Fresh Energy, represented by Hudson B. Kingston, Minnesota Center for Environmental Advocacy, 2 and Will Nissen, Senior Policy Associate, Fresh Energy. III. Proceedings Before the Administrative Law Judge The Office of Administrative Hearings assigned Administrative Law Judge (ALJ) Eric L. Lipman to hear the case. The parties filed direct, rebuttal, and surrebuttal testimony prior to the opening of evidentiary hearings. The ALJ held evidentiary hearings in Saint Paul on January 15, 19, 20, and 21, After the hearings the parties filed initial briefs, reply briefs, and proposed findings of fact and conclusions of law. The ALJ also held five public hearings in the case, on the dates and at the locations set forth below: Earle Brown Heritage Center, Brooklyn Center December 2, 2015 Sabathani Community Center, Minneapolis December 2, 2015 Central Lakes Community College, Brainerd December 3, 2015 Mankato Civic Center December 8, 2015 Normandale Community College December 8, 2015 IV. Public Comments The Administrative Law Judge held five public hearings, where the Company, the Department, the OAG, and the Commission s Consumer Affairs Office made presentations and fielded questions from members of the public. All public comments are filed in the case record. Written comments are labeled Public Comment, and oral comments appear in the public-hearing transcripts filed by the court reporter. Thirty-one members of the public filed written comments, and two members of the public spoke at the public hearings. 2 Mr. Kingston appeared for the limited purpose of defending Fresh Energy s witness, Will Nissen, during the evidentiary hearings on January 19 21, Mr. Kingston withdrew as counsel on January 25,

9 Nearly all commenting members of the public either opposed the rate increase entirely or argued that it was too high. The objections raised most frequently were that the increase would cause hardship for low-income households, that the amount of the increase should not exceed the inflation rate or the latest Social Security cost-of-living adjustment, and that the Company was not controlling costs sufficiently, especially in light of increased supplies of natural gas and lower natural-gas prices. V. Proceedings Before the Commission On March 17, 2016, the Administrative Law Judge filed his Findings of Fact, Summary of Public Testimony, Conclusions of Law, and Recommendation (the ALJ s Report). The following parties filed exceptions to the ALJ s Report under Minn. Stat and Minn. R : the Company, the Department, the OAG, and the SRA. On April 28 and May 5, 2016, the Commission heard oral argument from and asked questions of the parties. On May 5, 2016, the record closed under Minn. Stat , subd. 2. Having examined the entire record in this case, and having heard the arguments of the parties, the Commission makes the following findings, conclusions, and order. I. The Ratemaking Process FINDINGS AND CONCLUSIONS A. The Substantive Legal Standard The legal standard for utility rate changes is that the new rates must be just and reasonable. 3 The Minnesota Supreme Court has described the Commission s statutory mandate for determining whether proposed rates are just and reasonable as broadly defined in terms of balancing the interests of the utility companies, their shareholders, and their customers, citing Minn. Stat. 216B.16, subd That statute is set forth in pertinent part below: The commission, in the exercise of its powers under this chapter to determine just and reasonable rates for public utilities, shall give due consideration to the public need for adequate, efficient, and reasonable service and to the need of the public utility for revenue sufficient to enable it to meet the cost of furnishing the service, including adequate provision for depreciation of its utility property used and useful in rendering service to the public, and to earn a fair and reasonable return upon the investment in such property. B. The Commission s Role While the Public Utilities Act provides baseline guidance on the ratemaking treatment of different kinds of utility costs, it generally makes only threshold determinations on rate recoverability, leaving to the Commission the tasks of determining (a) the accuracy and validity of claimed costs; 3 Minn. Stat. 216B.16, subds. 4, 5, and 6. 4 In re Interstate Power Co., 574 N.W.2d 408, 411 (Minn. 1998). 3

10 (b) the prudence and reasonableness of claimed costs; and (c) the compatibility of claimed costs with the public interest. In ratemaking, therefore, the Commission must decide a wide range of issues, ranging from the accuracy of the financial information provided by the utility, to the prudence and reasonableness of the underlying transactions and business judgments, to the proper distribution of the final revenue requirement among different customer classes. These diverse issues require different analytical approaches, involve different burdens of proof, and require the Commission to exercise different functions and powers. In ratemaking the Commission acts in both its quasi-judicial and quasi-legislative capacities: As a quasi-judicial body it engages in traditional fact-finding, and as a quasi-legislative body it applies its institutional expertise and judgment to resolve issues that turn on both factual findings and policy judgments. As the Supreme Court has explained, [I]n the exercise of the statutorily imposed duty to determine whether the inclusion of the item generating the claimed cost is appropriate, or whether the ratepayers or the shareholders should sustain the burden generated by the claimed cost, the MPUC acts in both a quasi-judicial and a partially legislative capacity. To state it differently, in evaluating the case, the accent is more on the inferences and conclusions to be drawn from the basic facts (i.e., the amount of the claimed costs) rather than on the reliability of the facts themselves. Thus, by merely showing that it has incurred, or may hypothetically incur, expenses, the utility does not necessarily meet its burden of demonstrating it is just and reasonable that the ratepayers bear the costs of those expenses. 5 C. The Burden of Proof Under the Public Utilities Act, utilities seeking a rate increase have the burden of proof to show that the proposed rate change is just and reasonable. 6 Any doubt as to reasonableness is to be resolved in favor of the consumer. 7 On purely factual issues, the Commission acts in its quasi-judicial capacity and weighs evidence in the same manner as a district court, requiring that facts be proved by a preponderance of the evidence. On issues involving policy judgments, the Commission acts in its quasi-legislative capacity, balancing competing interests and policy goals to arrive at the resolution most consistent with the broad public interest. Utilities seeking rate changes must therefore prove not only that the facts they present are accurate, but that the costs they seek to recover are rate-recoverable, that the rate recovery mechanisms they propose are permissible, and that the rate design they advocate is equitable, under the just and reasonable standard set by statute. As the Court of Appeals explained, quoting the Supreme Court, 5 In re N. States Power Co., 416 N.W.2d 719, (Minn. 1987) (citation omitted). 6 Minn. Stat. 216B.16, subd Minn. Stat. 216B.03. 4

11 A utility seeking to change its rates has the burden of proving by a preponderance of the evidence that its proposed rate change is just and reasonable. Preponderance of the evidence is defined for ratemaking proceedings as whether the evidence submitted, even if true, justifies the conclusion sought by the petitioning utility when considered together with the Commission s statutory responsibility to enforce the state s public policy that retail consumers of utility services shall be furnished such services at reasonable rates. 8 II. Rate Case Overview The Company said that the main reason for this rate case was the need to make major capital investments to comply with new federal regulations on pipeline safety and integrity. 9 No one disputed this claim, and the Company s filing confirmed it. The Company testified that, for the ten calendar years from 2002 through 2011, its annual capital expenditures averaged $65,000,000, while, for the seven years from 2015 through 2021, its annual capital expenditures are projected to average $200,000, Similarly, the Department recognized the substantial increase in the Company s capital expenditures since its last rate case, noting that its rate base had grown by 30.3% (or some $212,572,000) over the two-year period since its last rate-case filing. 11 While these infrastructure investments in large part drove the Company s filing, they were not the main points of contention in this case. No one challenged the investments themselves, and the only related cost that was challenged was the cost of a records-management project designed to collect, integrate, and manage infrastructure data, the PRIME Project, 12 discussed under the Financial Issues section of this order. The Company s test year runs from October 1, 2015, to September 30, III. Summary of the Issues In its Notice and Order for Hearing the Commission listed 17 corporate-allocations issues on which the Company should provide more information, and the Company provided detailed information on all 17 issues in response to a Department discovery request. In subsequent discussions between the Company and the Department, both parties agreed that the full amount of aircraft expenses included in test-year costs ($34,083) should be removed, and the Administrative Law Judge concurred. No one challenged any other cost related to the 17 issues listed in the order. 8 In re Minn. Power & Light Co., 435 N.W.2d 550, 554 (Minn. App. 1989) (citations omitted). 9 CPE Ex. 7, at 6 (Vortherms Direct). 10 Id. at DOC Ex. 503, at 6 (Johnson Direct). 12 Permanent Records Integrity Management Excellence Project. 5

12 Many initially contested issues were resolved in the course of evidentiary proceedings. The Administrative Law Judge found that the resolutions reached by the parties were reasonable and supported by record evidence; he recommended accepting them. 13 The Commission concurs. Other issues remained contested. The following issues were either contested or otherwise require discussion. Financial Issues PRIME (Permanent Records Integrity Management Excellence) Project Costs Should the Company be permitted to recover the cost of this project, which collects, integrates, and converts to an electronic format detailed information on the age, size, composition, and location of the components of its transmission and distribution system? Environmental Cleanup Costs Should the Company be permitted to recover $2,000,000 in manufactured-gas-plant cleanup costs? And should the Commission require the Company to restore $7,100,000 in cleanup funds paid from its insurance recovery account tracker? Nicollet Mall Headquarters Has the Company demonstrated that the cost of its new headquarters was prudently incurred, properly allocated between regulatory jurisdictions, properly allocated between intra-company business units, and properly accounted for in Company depreciation schedules? Plant Retirements Are the Company s projected test-year plant retirement projections reasonable, or should it be required to recalculate them on a category-by-category basis? Distribution-Plant Balances Should the Company s end-of-test-year distribution-plant balances be adjusted in light of actual start-of-test-year data? Cash Working Capital Has the Company adequately documented its cash working capital, including payments to the parent corporation s service company? Non-Qualified Pension Expenses Should the Company be permitted to recover pension costs that exceed tax-deductible levels set by the IRS? Inflation Rates Are the inflation factors used by the Company in calculating test-year costs reasonable and prudent? Incentive Compensation Should the Company be permitted to recover the costs of its long-term and short-term incentive-compensation plans? Executive Compensation Should the Company be denied rate recovery of the salary of the Executive Chairman of the Board of Directors due to its heavy dependence on long-term performance incentives? 2013 Rate-Case Costs Should the Company be permitted to recover $11,736 in 2013 rate-case costs incurred in 2014, despite the Commission establishing an approved recovery for 2013 rate case costs in a previous order? American Gas Association Dues Should the Company be permitted to recover some $177,584 in dues for its membership in the American Gas Association? 13 ALJ s Report

13 United Way Stock Donation What recovery is permissible for the Company s indirect donation of Time-Warner stock to United Way? Property Taxes Has the Company adequately supported its estimate that pending school-district referendums would result in a 2% increase in test-year property taxes? Cost-of-Capital Issues Capital Structure What percentages of equity, long-term debt, and short-term debt should make up the Company s capital structure? Cost of Short-Term Debt What is the Company s cost of short-term debt? Return on Equity What is a fair and reasonable rate of return on equity for this Company, on this record, at this time? Class-Cost-of-Service-Study (CCOSS) Issues CCOSS Procedures for Classifying and Allocating Distribution-System Costs Should the Commission rethink its historical reliance on minimum-system modeling and instead adopt the Alternative Minimum System Method (AMSM), the Basic Customer Approach, or the Peak-and-Average Approach to classify and allocate between customer classes the costs of the distribution system? CCOSS Treatment of Sales Expense, FERC Accounts Is the Company s interclass allocation of sales costs by number of customer locations the most reasonable allocation method in the record? CCOSS Treatment of Regulatory Commission Expense, FERC Account 928 Is the Company s interclass allocation of regulatory commission expense by number of customer locations the most reasonable allocation method in the record? CCOSS Treatment of Historical Cost Data on Distribution Mains In classifying and allocating distribution-main costs, should costs be set at original investment or updated to present cost, using the Handy-Whitman index? CCOSS Treatment of Income Tax Is it just and reasonable for the Company to classify income-tax costs, in their entirety, as customer costs? Rate-Design Issues Interclass Revenue Apportionment What percentage of the revenue requirement should be allocated to each customer class? Customer Charges At what level should the Commission set the fixed monthly charge for Residential, Commercial A and Large-Volume Sales and Transportation customers? 7

14 Future Rate-Case Issues Calendar-Year Test Year Should the Commission vary Minn. R , subp. 17 and require the Company to use a calendar year for the test year in its next rate case? Sales Forecast and Weather Normalization Should the Company be required to provide detailed discussion of 10-year, 15-year, and 20-year weather patterns as part of the sales-forecast analysis in its next rate case? Interim Rate Design Methodology Should the Company be required to remove the cost of gas from its interim rate design in future rate cases? These issues are examined individually below, with issues on which the Commission declines to accept the ALJ s recommendation discussed in greater detail. IV. The Administrative Law Judge s Report The Administrative Law Judge s Report is well reasoned, comprehensive, and thorough. The ALJ held four days of formal evidentiary hearings and five public hearings. He reviewed the testimony of 24 expert witnesses and related hearing exhibits. He heard testimony from members of the public and read some 31 written comments submitted by members of the public. The ALJ received and reviewed initial and reply post-hearing briefs from the parties, as well as their proposed findings of fact and conclusions of law. He made some 611 findings of fact and conclusions of law and made recommendations on all stipulated, settled, and contested issues based on those findings and conclusions. The Commission has itself examined the record, considered the report of the Administrative Law Judge, considered the exceptions to that report, and heard oral argument from the parties. Based on the entire record, the Commission concurs in most of the Administrative Law Judge s findings and conclusions. On some issues, however, the Commission reaches different conclusions, as delineated and explained below. And on a few issues it provides technical corrections and clarifications. On all other issues, the Commission accepts, adopts, and incorporates the ALJ s findings, conclusions, and recommendations. V. Nicollet Mall Headquarters A. Introduction FINANCIAL ISSUES In April 2015, the Company relocated its headquarters to a building it purchased at 505 Nicollet in Minneapolis. Before moving, the Company s headquarters was in leased space, with the lease set to expire in Owning rather than leasing property results in a higher rate base upon which the company could earn a return but potentially reduces expenses enough to offset the increased authorized recovery from ratepayers, resulting in a net savings over time. Accordingly, a change from leasing to 8

15 ownership of the Company s headquarters building presents a question of reasonableness to the Commission. CenterPoint included a total of $19,387,586 of the $26,300,000 capital costs for the headquarters in the rate base for the Minnesota-jurisdictional portion of the Company. The test year also included the building s operations and maintenance (O&M), property tax, and depreciation expenses. The Department and the OAG have challenged inclusion of portions of the building s rate-base allocation and expenses on a variety of grounds, explained below. B. Positions of the Parties 1. The Department The Department asserted that CenterPoint had not met its burden to show that the rate-base allocation of costs for the new headquarters was reasonable. Specifically, the Department objected to the building jurisdictional allocations for expenses and rate-base costs, and to the claim that owning the building was more cost-effective than leasing. The Department also challenged the Company s depreciation assumptions and resulting depreciation expense amount. In its original filing, the Company allocated 77.82% of the building s capital value to Minnesota ratepayers, which the Department argued was inconsistent with the Company s allocation of other building-related costs and with the allocation during the Company s lease arrangement. The Department contended that the Company should be required to reduce the rate-base allocation to 48%, consistent with the Company s allocation of operating expenses for the same building. The Department also disputed the reasonableness of the Company s net-present-value analysis of purchasing the building rather than continuing to lease. The Department challenged the rigor of the Company s analysis offered to support the claim that increased capital and other expenses arising from ownership were reasonable in light of the anticipated cost of leasing. Finally, the Department disputed the Company s decision to account for building depreciation assuming a 36-year useful life and a -10% salvage value. The Department provided evidence supporting a 50-year useful life and a +10% salvage value. The depreciation assumptions proposed by the Department would reduce the test-year revenue requirement, and therefore the amount approved for recovery from ratepayers. The Department also recommended that the Company be required to complete a full depreciation study. 2. The OAG The OAG agreed with the Department s criticisms of CenterPoint s building rate-base and expense allocations, and with the Department s proposed depreciation assumptions. The OAG also expressed concern that the Company s calculations did not completely exclude costs associated with a portion of the building used by the Meet Minneapolis, Convention and Visitors Association. The OAG also supplemented the Department s claim that ownership was not justified, by arguing that owning the building was more costly than leasing the Company s previous headquarters. Based on this argument, the OAG recommended that the Commission exclude the difference in cost between the Company s previous headquarters and its new headquarters. It calculated the difference to be $1,243,845, including $551,094 in building expenses. 9

16 3. The Company According to the Company, the purchase of the Nicollet Mall building was a reasonable and cost-effective decision in light of the available alternatives, including leasing. The lease at its former headquarters building was set to expire, and the Company anticipated that the cost to lease the same space would increase dramatically. Additionally, the Company wanted more space, more control over its operating costs, and to maintain access to public transportation and its presence in Minneapolis. For these reasons, the Company testified, it began considering its headquarter-location options in The Company concluded that purchasing the Nicollet Mall building was the least-cost long-term option that met the Company s goals, and provided three net-present-value analyses in support of its conclusion. It argued that leasing would be both more costly and more risky over the anticipated life of the purchased building. The Company agreed to the use of a 50-year life and a 10% salvage value for the building, for ratemaking purposes. The Company asserted that a full depreciation study was not warranted based on the dispute that arose over the headquarters building depreciation, but agreed that at least an analysis of FERC account (General Plant: Structures and Improvements, Nonleasehold) was justified. It offered to perform an analysis of FERC account in its next rate case, in lieu of the full depreciation study proposed by the Department. CenterPoint also acknowledged that approximately 5,000 square feet of the building is used by the Meet Minneapolis, Convention and Visitors Association, and agreed that rate base and expenses related to the Meet Minneapolis lease should be excluded. However, the Company continued to support its Minnesota-regulated allocations of building costs, and contended that the adjustments recommended by the Department and the OAG were unjustified. The Company asserted that the Department s analysis depended on a misunderstanding of the way the Company had labeled its allocations, and that the OAG s proposed reduction was flawed because it relied on an inapplicable and unavailable alternative for comparison. C. The Recommendation of the Administrative Law Judge The ALJ found that the record supported the Company s decision to buy its Nicollet Mall headquarters, and that disallowing recovery as the Department and OAG proposed would be an unreasonable outcome. He concluded that the Department s proposed adjustment to the Minnesota-regulated allocation of building costs would be unreasonable, and that the Company s allocations, though confusingly labeled, were more reasonable than the Department s proposed alternative allocations. He found the Company s decision-making process to be sound and to have produced the best available result for ratepayers. He recommended that expenses and a return on the property be recovered through rates, provided that costs associated with Meet Minneapolis were removed and the facility was depreciated according to a 50-year life and 10% building salvage value. 10

17 D. Commission Action The Commission concludes that CenterPoint s purchase of the Nicollet Mall building for its headquarters was prudent. Therefore, O&M expenses associated with CenterPoint s Nicollet Mall headquarters are reasonably included in the test year. Despite the Company s net-present-value analyses being incomplete and not included in its initial filing, the Commission concludes that the record is nevertheless sufficient to show that the building purchase was a prudent decision among the reasonably comparable and available alternatives provided the adjustments discussed below are made. CenterPoint will be required to depreciate Nicollet Mall over 50 years, with a positive 10% salvage value, as it has agreed and as the ALJ recommended. The Company must also ensure that it excludes all rate-base and O&M costs related to the Meet Minnesota lease, as it agreed to do. The Commission also agrees with the Department concerning building-cost allocations, and will require that the Company s rate base be adjusted to reflect the Department s recommended building-cost allocation of 48% Regulated and 52% Non-Regulated. This will allow CenterPoint to include $12,025,047 in rate base for the Nicollet Mall building. To be consistent with the modified allocation, the Company must also reduce its test-year Depreciation Expense by $490,338. In support of these determinations, and in lieu of the ALJ s findings on this issue, the Commission will adopt the Department s proposed amended findings. 14 The Commission concludes that these findings better reflect the record and the conclusions that should be drawn from it. Specifically, the Commission finds that the Company did not adequately support its rate-base asset allocations between regulated and non-regulated uses. Justifications for the allocations were incomplete in the Company s initial filing, and insufficient in its responses to requests for information by the Department and the OAG. CenterPoint has not adequately explained why it is reasonable to adopt different regulated/ nonregulated allocations for rate-base asset value and for building expenses such as depreciation. Absent adequate explanation, the Commission is persuaded that adopting different allocations in this fashion is likely to be unreasonable. The Department s investigation has therefore raised reasonable doubt about the Company s claims concerning its rate-base allocation for the Nicollet Mall building. Doubt about the reasonableness of these allocations must be resolved in favor of ratepayers. 15 The Commission therefore concludes that the Department s proposed building rate-base allocation is more reasonable than the Company s. A 48% allocation to Minnesota regulated use is consistent with the Company s 48% allocation of the building s depreciation expense, O&M expense, and property taxes. 14 Exceptions of the Minnesota Department of Commerce to the ALJ s Report, at These findings are reproduced in full in the ordering paragraphs below. 15 Minn. Stat. 216B

18 The Commission has determined not to require a full depreciation study as requested by the Department. Instead, it will require CenterPoint to provide a depreciation analysis of FERC Account (General Plant: Structures and Improvements, Nonleasehold) in its next rate case. The Commission agrees that the large changes in that specific account, attributable to the headquarters and to a warehouse purchase, warrant a closer analysis of the account in the Company s next rate case. A broader look at depreciation is not necessary at this time. VI. PRIME Project Expenses A. Introduction CenterPoint has proposed to include $1,442,858 in test year expenses related to its Permanent Records Integrity Management Excellence (PRIME) project. According to the Company, the project is intended to integrate currently disjointed sources of data, information, and records which would allow the Company to more effectively maintain a safe pipeline network. The OAG argued that the PRIME project expenses should be excluded from the test year, and also recommended that the Commission disallow $80,144 of building operations and maintenance (O&M) costs that it calculated are attributable to the PRIME project. B. Positions of the Parties CenterPoint characterized the PRIME project as an advancement in its information systems necessary to analyze its physical infrastructure, facilitating more efficient responses to safety concerns and regulatory changes. According to the Company s witness, the PRIME project will allow the Company to layer-in data from outside sources, such as system data from other operators, and weather and soil data. With the intelligence of the data in the system and the ability to significantly expand the data set, the Company can execute advanced analytics against the data set to identify possible trends earlier [than] with our current record sets. 16 The Company reduced its initially proposed test-year expense amount from $1,778,759 in response to a Department recommendation. CenterPoint and the Department agreed that inclusion of $1,442,858 in PRIME project expenses is warranted. In the OAG s view, the PRIME project serves to remediate pre-existing shortcomings in the Company s recordkeeping. On that basis, the OAG recommended that the Commission exclude PRIME project-related expenses, including building O&M expenses attributable to the project. The OAG asserts that ratepayers should not be required to pay for a program designed to address gaps, inaccuracies, and other deficiencies in the Company s records. The OAG also disputed that the Company adequately supported the expense amount ultimately agreed to by the Department and recommended by the ALJ. 16 Ex. 20, at 19 (Centers Rebuttal). 12

19 C. The Recommendation of the Administrative Law Judge The ALJ found that PRIME was a necessary part of the Company s pipeline safety and integrity management efforts, and a reasonable method of fulfilling its duties under state and federal law. He recommended that the Commission include $1,442,858 of PRIME system-related costs in test-year expenses. The ALJ did not make a specific recommendation concerning the disputed $80,144 of building O&M costs. D. Commission Action The Commission will approve recovery of CenterPoint s PRIME Project s test year expenses totaling $1,442,858, and recovery of related O&M expenses. Continual improvement of the Company s information-management systems is expected by the Commission, and necessary for efficient implementation of the Company s Transmission Integrity Management Program (TIMP) and Distribution Integrity Management Program (DIMP). The Commission agrees with the ALJ that the PRIME project serves these purposes. Improving the Company s information infrastructure together with these significant infrastructure projects is reasonable and not an indication that it had suffered from neglect. Therefore, the Commission concurs with the ALJ that $1,442,858 in PRIME project expenses has adequate support in the record, is a necessary part of the Company s pipeline safety and integrity management efforts, and reasonably facilitates compliance with state and federal obligations. For these reasons, Commission will also permit recovery of the building O&M costs related to the PRIME project. VII. Environmental Costs A. Introduction CenterPoint has included in its test year $2,000,000 of environmental cleanup costs related to a manufactured gas plant formerly owned by the Company. The expense is in addition to $7,100,000 in insurance proceeds, intended for the purpose of environmental cleanup, which were also applied to cover cleanup costs for the plant. On July 10, 2015, the Commission approved the payment of insurance proceeds for cleanup, and ordered the Company to file direct testimony on the issue in this rate case. 17 B. Positions of the Parties CenterPoint and the Department agreed that recovery of the $2,000,000 in costs was reasonable, and agreed on an accounting method. The agreement would allow the Company to recover the identified cleanup costs, amortized over a two-year period, and cap recovery at the approved amount. 17 Order, Docket No. G-008/GR

20 The OAG argued that including the environmental cleanup payments in the test year was inappropriate. In the OAG s view, the two-year recovery period amounted to an effort to use accounting to push cleanup costs into the test year. The OAG argued that the cleanup costs were incurred and paid entirely outside the test year, and should therefore be excluded. The OAG further asserted that the time to request deferred accounting for the expense was when it was incurred, not in a future rate case, rendering the Company s recovery proposal untimely. With regard to the payment of insurance proceeds, the OAG argued that the Company had not established that the expense was reasonable. It further argued that the payment did not absolve the Company of future liability for the site. For these reasons, the OAG argued, the Company should be required to restore the amount of the payment to the Company s insurance-recovery account tracker. C. The Recommendation of the Administrative Law Judge Relying on the record and prior Commission decisions, the ALJ found that the environmental costs at issue were recoverable in rates. He concluded that the Company properly applied the insurance proceeds as directed by the Commission. He also concluded that recovery of the expense was consistent with the Commission s prior decisions concerning recovery of the Company s environmental cleanup costs. Accordingly, the ALJ recommended that the identified environmental cleanup expenses should be amortized over two years, with a sunset to prevent over-recovery, as agreed to by the Company and the Department. D. Commission Action Environmental remediation costs for the Company s former manufactured gas plants have previously been recognized as prudent. Here, the OAG challenges both the reasonableness and the timing of the recovery request. These cleanup costs are not materially different from any that the Commission has authorized for recovery over the past two decades. Liabilities for remediating former manufactured gas plants have been generally approved for recovery as a ratepayer responsibility, though subject to dispute over reasonable and appropriate ways to account for the expense. The Commission agrees with the ALJ and will approve recovery of CenterPoint s $2,000,000 in identified environmental cleanup costs (net, after application of the insurance proceeds) over the next two years. Recovery for this expense will be capped at $2,000,000. The Company will be required to track any over-recovery and discuss in the initial filing of its next rate case how it proposes to handle any over-recovery. 14

21 The Commission will also not revisit its prior decision to authorize the payment from the insurance recovery account tracker. That the amount of the payment was arrived at through mediation and did not completely absolve the Company of future liability at the site does not render the amount unreasonable. But CenterPoint did receive a release of liability for past cleanup expenses and from liability for future cleanup costs or other environmental claims or damages subject to some specific and limited exceptions. 18 The Commission is persuaded that the cleanup expense is reasonable. Approval of this recovery is limited to the facts presented in this case. Specifically, the Commission recognizes that its July 10, 2015, order could be read as an implicit invitation for the Company to seek a recovery determination on these cleanup costs in this rate case. The Commission agrees with the OAG that, in general, deferred-accounting recovery requests should be made when an expense arises. Possible recovery of future out-of-period environmental costs will require prior deferred accounting approval and will otherwise be subjected to normal ratemaking treatment. VIII. Plant Retirements A. Introduction As a part of calculating the Company s test-year rate base the assets upon which the Company may earn a rate of return from ratepayers the Company has proposed a method of calculating the assets retired during the test year. When an asset is retired, it is no longer a part of the Company s rate base. 19 The OAG has disputed the method the Company used to calculate its test-year retirements. A method that computes a higher level of retirements would indirectly reduce the Company s rate base. B. Positions of the Parties 1. The OAG The OAG asserted that several of the Company s plant retirement calculation results were unreasonable. In particular, the OAG argued that the Company s method calculated no plant retirements in 14 FERC accounts, which was inconsistent with the Company s historical average for those accounts over the past five years. In other words, the Company s calculation is inconsistent with plant retirements in the recent past and, the OAG argued, so inconsistent as to be an unreasonable model for test-year retirements. An alternative calculation using FERC plant categories rather than individual FERC accounts was proposed by the OAG. Using the Company s historical five-year average plant additions and retirements as a comparison, the OAG argued that its method offered more reasonable results. Using the broader categories produces a higher test-year retirement amount that is more in line with the Company s five-year annual average retirement percentage. 18 Direct Testimony of Mr. Kirk R. Nesvig, at A corresponding adjustment is also made to the Company s depreciation reserve. 15

22 2. The Company CenterPoint disputed that its retirement calculations were unreasonable, and argued that the OAG s alternative calculation had its own flaws. It rejected the OAG s implicit assumption that retirements in the test year should be consistent with five-year historical averages. The Company also criticized the OAG s method of using plant categories, arguing that it can produce less precise results compared to the Company s FERC-account-based analysis. The Company highlighted that it has used its plant-retirement calculation method for the past several rate cases, and recommended that the Commission approve its method. C. The Recommendation of the Administrative Law Judge The ALJ found that the Company s calculation of plant retirements is reasonable and appropriate. D. Commission Action Having considered the plant retirement calculations in the circumstances of this case, the Commission agrees with the OAG that CenterPoint s method of projecting test-year plant retirements is inaccurate, and because of the nature of those inaccuracies, is less reasonable than the OAG s proposed alternative. The Commission will require CenterPoint to calculate its test-year Plant Retirements using the OAG s category-by-category methodology. As the OAG pointed out, the Company s methodology led to contradictory and unsupportable retirement projections for several FERC accounts. The Company s method is particularly susceptible to under-calculating retirements when retiring assets in a FERC account and not also adding assets in that same FERC account. The Company s calculation included no retirements in 14 FERC accounts, including the cast iron mains account. These results are unsupportable in light of the Company s ongoing capital improvement projects and its specific commitment to replace all of its cast iron mains in the near future. The unlikely results calculated by the Company suggest that its method likely under-calculated retirements in the test year. In light of the Company s ongoing, significant investment and asset-replacement efforts, the Commission concludes that a plant retirement calculation method shown to be more consistent with its recent history is more likely to reflect its test-year activity. The Commission is persuaded that using broader plant categories is a more reasonable way to calculate plant retirements in this case. The Commission acknowledges that the OAG s method may have its own shortcomings because it is less fine-grained than the Company s method. Accordingly, the Commission will require in the Company s next rate case that it quantify the plant retirement impact using both the FERC-account method and the category-by-category method. This will allow the Commission and interested parties to examine the differences between the two methods. 16

23 IX. Distribution-Plant Balances A. Introduction In its initial filing the Company projected beginning- and end-of-test-year accounting balances for its distribution-plant assets. As the rate case progressed, the actual beginning-of-test-year balance became available and was used by the parties. The Company agreed that, based on the actual data, it was appropriate to decrease its beginning distribution-plant balance by $2,268,003. Based on this updated information, the Department recommended reducing the rate base by the same amount, with a commensurate reduction in the test year s depreciation expense. The Company disagreed with the Department s proposal, and argued a smaller reduction was appropriate. B. Positions of the Parties 1. The Department The Department argued that both the beginning- and end-of-test-year balances should be reduced by the same amount. Doing so would reduce the rate base, which uses an average of the two figures, by the full $2,268,003. In support of its argument, the Department asserted that the Company did not provide any evidence to support adjusting the beginning balance to reflect the actual data, but not adjusting the ending balance. Adjusting only the beginning balance would implicitly adopt a new, $2,268,003 higher forecast for new test year distribution plant. The Department argued that the record provided no support to conclude that such an increased forecast was reliable or reasonable. 2. The Company CenterPoint agreed to reduce its distribution-plant beginning balance to conform to the updated amount, but did not agree that the end-of-year balance should be reduced by the same amount. It argued that additional projected activity justified leaving the end-of-year balance unadjusted. Applying the adjusted beginning balance, the Company contended that it should reduce its rate base by half of the adjusted amount, or $1,135,000. C. The Recommendation of the Administrative Law Judge The ALJ found the Department s position reasonable, and recommended reducing the Company s test-year rate base and test-year depreciation expense by $2,268,003 and $42,493, respectively. He found inadequate support in the record for the Company s position. D. Commission Action The Commission agrees with the ALJ and the Department. The implicit assumption that during the test year the Company would increase its actual net distribution-plant balance to that which it had originally forecast, and therefore only the starting balance should be adjusted in light of the new information, is not supported by the record. The balance is starting from a lower point, and applying the company s own forecast, will end at a lower point. Adjusting both balances in light of actual data, and in the absence of evidence supporting a modified forecast, is more reasonable. 17

24 For these reasons, the Commission adopts the ALJ s recommendation and will require that CenterPoint s net distribution balances and test-year rate base be reduced by $2,268,003, resulting in a test-year depreciation expense reduction of $42,493. X. Cash Working Capital A. Introduction In its initial filing, the Company included an adjustment for cash working capital, which included payments the Company made to the Service Company for services provided. The OAG contested the payments to the Service Company, recommending that the cash-working-capital amount exclude $570,619 for the disputed payments. B. Positions of the Parties In the OAG s view, the Company did not adequately substantiate that it made the payments to the Service Company. The OAG recommended finding that the Company failed to show that it actually paid cash for the costs assigned to it by the Service Company. The core of the OAG s argument is that it believes the documentation provided by the Company does not establish that it actually made the payments in question. CenterPoint disagreed, stating that it had accounted for its cash working capital using the same method it has used in previous rate cases. It described that payments to the Service Company are reflected in intercompany accounting entries that are supported by the record and no different from the entries that would be made for payments to third parties. It argued that it had adequately substantiated the payments in question. It recommended that the OAG s proposed disallowance be rejected. C. The Recommendation of the Administrative Law Judge The ALJ found that the Company s accounting for cash working capital is consistent with adjustments for other vendors and in previous rate cases. He therefore recommended the Company s approach over the OAG s proposed disallowance. D. Commission Action Because the Commission agrees with the ALJ, it will approve CenterPoint s cash-working-capital methodology. The Company s evidence of the disputed payments which is consistent with the record it would provide for payments to third parties is adequate. The Company will be required to update its final calculation of cash working capital to reflect the decisions made in this order. XI. Non-Qualified Pension Expense A. Introduction CenterPoint requests recovery of $34,841 in non-qualified pension expense. The Company offers the non-qualified plan to employees because qualified pension plan benefits are limited by IRS caps on annual earnings, benefit contributions, or benefits payable. The non-qualified plan supplements the pension plans for high-earning employees that are subject to the IRS cap on those benefits. Non-qualified pension plans do not receive the same tax-favored status as qualified plans. 18

25 B. Positions of the Parties The Department argued that the non-qualified pension and savings plan expenses should be excluded from the Company s test year. It asserted that these plans only benefit highly-compensated employees, and that the Company had not established that the identified expenses were necessary and reasonable. The Department also argued that the Commission should likewise exclude the test year non-qualified pension costs from the Service Company ($984,052) and associated capital costs ($12,115). CenterPoint objected to excluding the non-qualified pension expense. It argued that it had established that its overall compensation for all employees is reasonable, which includes the Company s compensation for employees who receive non-qualified pension benefits. C. The Recommendation of the Administrative Law Judge The ALJ recommended excluding the non-qualified pension costs, concluding that the Company had not established that absent these costs it could not secure the employees necessary to provide safe, reliable service. D. Commission Action The Commission will adopt the ALJ s recommendation and deny recovery of all non-qualified pension costs. Having concluded that the record lacked evidence adequate to establish that the expenses were reasonable and necessary costs of providing utility service, the ALJ recommended disallowing the expenses. The Commission agrees because the Company did not establish that these expenses are reasonable and necessary to provide utility service, it will disallow recovery of the expenses. While technically open to all employees, benefits under these plans only accrue to the Company s highest earners. The record lacks evidence that absent these benefits the Company could not secure employees necessary to provide safe, reliable service. CenterPoint will be required to adjust its rate base to remove any associated capitalized costs commensurate with the disallowance. XII. Inflation Factors A. Introduction To calculate its Operations and Maintenance (O&M) costs for its October 1, 2015, to September 30, 2016, test year, the Company began with its actual 2014 O&M costs, made 40 adjustments for known and measurable changes, and applied inflation factors to all costs affected by inflation. CenterPoint used three inflation factors: one for direct payroll costs (5.32%), based on actual union wage contracts; one for secondary payroll costs (4.07%), based on wages, payroll taxes, and employee-benefit costs; and a general factor for other costs (3.96%), calculated using an average of Producer Price Index (PPI) values the Bureau of Labor Statistics index measuring cost changes for manufacturers and wholesalers. 19

26 The Company has used this process in all of its recent rate cases. These inflation factors accounted for $4,065,842 in test-year costs. B. Positions of the Parties 1. The OAG The OAG challenged the Company s payroll and general inflation factors and recommended alternatives. In lieu of the 5.32% factor for direct payroll costs, the OAG recommended a 5.25% factor, arguing that some reduction was appropriate to account for employee turnover in the test year. It also recommended that CenterPoint be required to assume a general 3.45% deflation factor for other costs. It disputed the Company s decision to use a 31-year historical average of the PPI as the basis for its general inflation factor, arguing that the data were stale and that recent economic conditions supported a conclusion that costs are decreasing rather than increasing. 2. The Company CenterPoint stated that it used the same approach to developing its inflation factors in this case that it has used in its last several rate cases and that the Commission has consistently approved them. The Company argued that the OAG s proposed payroll inflation factor was inappropriate because it did not take into account the effects of compounding and because it relies on an assumption about employee turnover that is not supported by the record. The Company also disputed the OAG s proposed general deflation factor. It argued that the PPI can fluctuate significantly over the short term, and that using a historical average smooths out those fluctuations. C. The Recommendation of the Administrative Law Judge The ALJ found that the Company s proposed payroll inflation factor and general inflation factor were supported by the hearing record, and recommended their adoption. He noted that in the past the Commission has approved the use of an averaged general inflation factor to smooth short-term fluctuations in the PPI. The ALJ also found that the company s proposed general inflation factor accounted for declines in commodity prices that the OAG highlighted in its argument. D. Commission Action The Commission concurs with the ALJ s findings, conclusions, and recommendations. The Commission finds that a 5.32% payroll inflation factor and a 3.96% general inflation factor should be used in this proceeding, as the ALJ recommended and the Company proposed. CenterPoint s proposed inflation factors are sound, supported in the record, and consistent with longstanding practice. The Commission is not persuaded that the OAG s proposed payroll factor is supported by the record, and will not accept the OAG s invitation to use a 21-month average PPI as the basis of the general inflation factor. A 21-month average would over-emphasize short-term fluctuations, limiting the predictive value of the result. 20

27 Although the Commission will approve the Company s general inflation factor based on the PPI and the Company s chosen historical averaging period, the Commission acknowledges the OAG s concern that recent economic conditions are underrepresented in a data set that stretches back 31 years. To increase future inflation calculations sensitivity to recent economic conditions, the Commission will require CenterPoint to use a 15-year average of the PPI for future rate cases. XIII. Incentive Compensation A. Introduction CenterPoint sought rate recovery of short-term and long-term incentive compensation costs. It proposed to include in its test year long-term incentive compensation expenses of $241,945, plus $1,188,176 as part of allocations from the Service Company; and short-term incentive compensation expenses of $2,179,656, plus $1,670,305 (on a pre-allocated basis) from the Service Company allocations. 20 Incentive compensation is paid in addition to base pay and is contingent on the Company meeting specified financial or operational goals. Fifty percent of short-term incentive compensation is automatically owed when certain metrics are met; the rest is payable at the discretion of management. Short-term incentive compensation is paid in cash in a one-time annual payment. Short-term incentive compensation is available to all employees. According to the Company, for most employees a short-term incentive award level is five, seven, or ten percent of the employee s base salary. Since the Company s last rate case, the performance indicators governing eligibility for short-term incentive compensation have changed, as follows: 2013 Operating income 25% Gas operations operating income 15% Operational performance 30% Customer service 10% Safety 20% 2014 CenterPoint Energy, Inc., operating income 55% O&M expenses 25% Safety 10% Operational Performance 10% 20 ALJ finding 412 mistakes which of the two short-term incentive compensation amounts were stated on a pre-allocated basis. 21

28 2015 CenterPoint Energy, Inc., operating income 55% O&M expenses 25% Safety 10% Customer satisfaction and reliability 10% Long-term incentive performance indicators are commonly aligned with shareholder interests. In , the Company measured performance toward total shareholder return and operating income goals. In the Company s last rate case, the Commission disallowed rate recovery of long-term incentive compensation costs, capped recovery of short-term incentive compensation costs at 25% of base pay, and required the Company to track amounts paid to employees and refund to ratepayers all amounts built into rates and not paid to employees. B. Positions of the Parties 1. The Company CenterPoint argued that both long- and short-term incentive compensation programs were standard in the utility industry, that they were necessary to attract and retain qualified employees for key leadership positions, and that they benefitted ratepayers by contributing to the strong financial performance required for a financially healthy operating utility. The Company opposed any cap on either incentive compensation plan, arguing that a cap was not supported by record evidence or objective analysis. 2. The Department The Department recommended disallowing rate recovery of all long-term incentive compensation costs, capping recovery of short-term incentive compensation costs at 25% of base pay, and disallowing recovery of all expenses for Service Company short-term incentive compensation for lack of proof. In its argument, the Department pointed out that the Commission has consistently taken a hard look at incentive compensation, usually disallowing rate recovery of long-term plans (because they target only shareholder goals) and capping rate recovery of short-term plans at 15%-25% of base pay to avoid over-rewarding short-term thinking and to reflect the value to shareholders. Here, the Department recommended disallowing all rate recovery of the Company s long-term program. The Department stated that the long-term plan had not been substantially changed since the last rate case and that it continued to be fair and reasonable for shareholders, not ratepayers, to bear its costs. It argued that the Commission has a long history of rejecting recovery of long-term incentive compensation, including in CenterPoint s last two rate cases. The Department also recommended a 25% cap for short-term incentive. It argued that the Commission has historically limited short-term incentives. It further recommended disallowing any recovery of Service Company short-term incentive compensation expenses, because the Company disclosed late in the evidentiary proceeding the significant incentive program expenses allocated to the Company as part of the Service Company s expenses. 22

29 In the alternative, if the Commission chose not to exclude the Service Company expense entirely, the Department proposed that a 100% payout target and a cap of 25% of base pay for Service Company short-term incentive compensation would best balance ratepayer and shareholder interests. C. The Recommendation of the Administrative Law Judge The ALJ concluded that recovery of short-term incentive expenses was reasonable and appropriate provided that the amounts were capped at 25% of base pay, and a 100% payout was assumed. The ALJ therefore recommended, subject to those requirements, that short-term incentive costs be included in the test year. He agreed with the Department and recommended exclusion of long-term incentive compensation expenses. He concluded that the Department s recommendation was reasonable in light of prior Commission decisions and because the program focuses primarily on the interests of shareholders rather than ratepayers. D. Commission Action The Commission concurs with the Administrative Law Judge and accepts his findings, conclusions, and recommendations, with the following modifications. 1. Long-Term Incentive CenterPoint s long-term incentive compensation program is substantially unchanged from the one rejected in the last rate case; its costs should therefore continue to be borne by shareholders. It is designed chiefly to serve shareholders interests; its benefits to ratepayers are indirect and could be better served by other means; and its time horizon for rewarding corporate financial performance carries the potential to divert attention from the much longer planning horizons critical to providing safe, reliable, and affordable utility service. 2. Short-Term Incentive Short-term incentive compensation costs should continue to be recovered from ratepayers subject to a cap of 25% of base pay. That cap continues to strike the right balance between the interests of ratepayers and shareholders and between the goals of rewarding solid day-to-day financial management and protecting the long-term thinking vital to good utility management. The cap responds appropriately to the design of the short-term program. While the program does tie employee compensation in part to performance goals that directly serve ratepayers safety and operational efficiency it also ties compensation substantially to Company financial performance. Accordingly, the Commission agrees with the ALJ that the 25% cap is appropriate and should remain in place. Continued under-emphasis of ratepayer-oriented goals in the Company s short-term incentive program is cause for concern. In future rate cases, if the short-term incentive goals do not place greater weight on customer-service components, the Commission may lower the cap or deny recovery of the short-term incentive plan costs. 23

30 Costs allocated to CenterPoint for the Service Company s short-term incentive plan will also be capped at 25% of employees base pay, for the reasons stated above. The Commission will also require the test-year amount to reflect a 100% payout target, which the Company has agreed to do. Though the Department s concerns about the timeliness of the information supporting this expense are warranted, the Commission concludes that the record is adequate and supports recovery subject to these limitations. In light of discrepancies in the various schedules documenting incentive compensation expenses, the Commission will require the Company to recalculate the approved expense amounts consistent with the decisions in this order, resolving any discrepancies in favor of its consumers. 3. Short-Term Incentive Refund Mechanism and Annual Reporting As it has in the past, the Commission will require CenterPoint to track all test-year incentive compensation amounts approved but not actually paid out, and refund them. It would be inconsistent with test-year principles and unfair to ratepayers to authorize full and unconditional recovery of these contingent, discretionary costs. CenterPoint will be required to continue filing an annual report on incentive compensation within 30 days after incentive compensation is normally scheduled for payout. Because the Service Company s short-term incentive expense is also approved subject to a 100% payout assumption, the Company will be required to provide a separate reporting of the regulated portion of the Service Company incentive plan amount actually paid, compared to the amount included in base rates. Unless the Company shows that corporate allocations to Minnesota jurisdictional utility operations are the reason for lower payout, the Commission will require the Company to return to ratepayers the amounts of unpaid Service Company incentive compensation built into rates. In its annual incentive compensation reports, in addition to the refund-related information, the Company must identify each performance indicator and its associated scorecard information, such as the measure, the goal for various attainment levels (threshold, target, maximum), its funding weight and the actual result achieved; and must report the overall plan payout percentage attained relative to the target goal of 100%. This will assist the Commission in evaluating the program s effectiveness at advancing ratepayer interests in the future. 4. ALJ Report Modification and Future Rate-Case Filing Requirements The Commission adopts the following modified ALJ Finding 418, to more precisely reflect the Department s position on short-term incentive expenses: The Department recommended that the entire Service Company STI amount be denied because CPE did not meet its burden of proof to justify its proposed recovery. Additionally, Alternatively, the Department recommended that if any STI from Service Company is included in corporate allocations, it should similarly be capped at 25 percent. DOC Initial Brief at

31 Finally, to facilitate analysis of these issues, when seeking recovery of short-term incentive costs in future rate cases CenterPoint must include and identify in its initial filing the incentive plan program documents. XIV. Executive Compensation A. Introduction CenterPoint initially included $2,285,183 in test-year expenses for the Minnesota-jurisdictional portion of compensation for its executive leadership. Historically, the Commission has allowed recovery of executive compensation to the extent it is consistent with the interests of ratepayers. During the contested case, CenterPoint reduced the amount by $592,571 for executives that had departed, and agreed to exclude non-utility-allocated expenses, leaving an expense of $1,206,783. The OAG recommended further disallowances related to the compensation of one executive: the Company s executive chairman, Mr. Milton Carroll. B. Positions of the Parties 1. The OAG The OAG argued that Mr. Carroll s executive role primarily serves shareholders rather than ratepayers. In support of its argument the OAG pointed to the structure of Mr. Carroll s compensation two-thirds is long-term incentive compensation, and he is ineligible to participate in the short-term incentive compensation program. According to the OAG, this compensation structure aligns Mr. Carroll s incentives with shareholders and not with ratepayers, and therefore his compensation should not be recovered from ratepayers. 2. The Company CenterPoint argued that its executive-compensation amount was appropriate for recovery, but did not respond to the OAG s arguments concerning Mr. Carroll s compensation structure. It did dispute that its initial calculations included non-utility-allocated expenses, but agreed that they are properly excluded from the calculated $1,206,783 expense amount. C. The Recommendation of the Administrative Law Judge The ALJ found that the same principle that applies to recovery of long-term incentive expenses should apply with the longer-term features of Mr. Carroll s compensation. He recommended that, to the extent that Mr. Carroll s compensation for his service as executive chairman includes long-term incentives, those amounts should be excluded from test-year expenses. D. Commission Action To promote the alignment of executive compensation and ratepayer interests, the Commission will approve recovery of the executive-compensation expense amount, minus Mr. Carroll s compensation costs for amounts in excess of his Minnesota-jurisdictional share of compensation for participating on the independent board of directors. The Commission is persuaded that Mr. Carroll s incentive structure as executive chairman is sufficiently out of step with ratepayer interests to exclude recovery of a portion of his compensation expense. 25

32 The Commission has already concluded, above, that long-term incentive compensation is not appropriate for rate recovery because it primarily aligns with shareholder interests rather than ratepayers. The rationale similarly applies to Mr. Carroll s compensation to the extent he receives compensation inconsistent with other independent board of directors members. The Company has chosen to compensate him in a manner that aligns his interests with shareholders more than ratepayers, and so the Commission will disallow recovery for compensation so structured. 21 In order to allow recovery of an amount that the Commission recognizes as consistent with executive service to ratepayers, CenterPoint will be authorized to recover the amount attributable to Mr. Carroll s independent board of directors membership. XV Rate-Case Expenses A. Introduction CenterPoint included in this case $11,736 in expenses attributable to its 2013 rate case. The OAG objected to these expenses. The Commission previously approved recovery of $1,915,335 in 2013 rate-case expenses. 22 B. Positions of the Parties The OAG argued that the rate-case expenses were already considered and decided by the Commission, and that it was unreasonable to now seek recovery for additional expenses. CenterPoint argued that the expenses were incurred in 2014 and properly included in its request for recovery in this proceeding. C. The Recommendation of the Administrative Law Judge The ALJ concluded that the identified rate-case expenses were appropriately incurred in 2014, and therefore are appropriate for recovery in this case. D. Commission Action Because CenterPoint s authorized 2013 rate-case expenses were determined in its 2013 rate case, additional recovery will not be permitted. CenterPoint Energy will be required to remove all employee expense items related to its 2013 Rate Case Expense, reducing test-year expenses by $11,736. In the Company s last rate case, the Commission authorized recovery of a regulated rate case expense amount of $1,915, The Commission concludes that it determined the authorized 2013 rate-case expense in its June 9, 2014, order, and will not revisit the issue. 21 This disallowance therefore excludes from recovery any Long-Term Incentive Compensation expenses for Mr. Carroll. 22 In the Matter of an Application by CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas for Authority to Increase Natural Gas Rates in Minnesota, Docket No. G-008/GR , Findings of Fact, Conclusions of Law, and Order, at (June 9, 2014). 23 Id. 26

33 XVI. American Gas Association Dues A. Introduction CenterPoint is a member of the American Gas Association (AGA), an industry organization that provides a variety of benefits to the Company and to ratepayers, such as industry reports, information exchange, and access to the current version of the Gas Engineers Operating Practices manual. AGA dues in the test year total $177,584. Some portion of the AGA s activities include lobbying state and federal government policymakers. Historically, the Commission has excluded lobbying expenses from rate recovery to the extent that the lobbying is not demonstrated to advance ratepayer interests. At issue is whether the Company should be permitted to recover AGA dues from ratepayers. B. Positions of the Parties 1. The OAG The OAG objected to recovery of lobbying expenses, specifically including the portion of AGA dues that support AGA lobbying efforts. Recognizing that AGA membership provides legitimate value to ratepayers, the OAG recommended only disallowing a portion of the AGA dues expenses. However, because the Company did not provide evidence establishing what portion of AGA dues go to lobbying, the OAG recommended disallowing 90% of the expense. The OAG believed this disallowance was reasonable in the absence of more precise evidence of the extent to which dues support AGA lobbying. 2. The Company CenterPoint argued that the AGA s primary function is not lobbying, and that it provides a number of other benefits to the Company and to ratepayers. Included among these benefits are helping the company improve its programs and practices through the sharing of industry knowledge, and coordinating the activities of AGA members to do more collectively than they could individually. The Company recommended that the Commission approve recovery of the full amount of its AGA dues. C. The Recommendation of the Administrative Law Judge The ALJ recommended that the Commission allow recovery of the AGA dues. He found that the dues provide value to ratepayers, and concluded that scrutinizing the nature of the lobbying to determine what portion of the dues should be excluded would lead to an unreasonable result. D. Commission Action As the ALJ implicitly determined, it is impossible to determine on this record the portion of AGA dues that are used for lobbying, or to perform a reasoned analysis about whether or to what extent the lobbying served ratepayer interests. The Commission therefore agrees with the ALJ that attempting to sort the expense into allowable and disallowable portions could lead to an unreasonable result. 27

34 But the Commission concludes that the most appropriate response is to disallow recovery of the entire amount. The Commission does not wish to tacitly reward the commingling of unrecoverable lobbying expenses with expenses that may otherwise be justified for recovery. The burden of justifying this expense rests on the Company, and doubt must be resolved in the ratepayers favor. 24 For these reasons, the Commission will not adopt the ALJ s recommendation and will require that American Gas Association membership dues be excluded from recoverable test-year costs. XVII. United Way Donation A. Introduction CenterPoint s test year included $33,526 in expenses related to a Service Company donation of Time Warner stock to United Way. The Company paid the Service Company for the value of appreciated stock at the date of the donation, and the donation resulted in a tax benefit to the Service Company. B. Positions of the Parties The OAG argued that recovery should be limited to the original value of the stock rather than the appreciated value at the time of the donation. If the Commission allowed the requested recovery, it asserted, the Company would retain a non-taxable gain at the expense of ratepayers the tax benefit arising from the charitable donation. The OAG argued that allowing recovery only of an amount equal to the acquisition value of the stock would avoid this outcome. CenterPoint disagreed, arguing that the full amount should be recoverable as reasonable and in the public interest. According to the Company, the Service Company acquired and owned the stock until it was donated, and Minnesota ratepayers were not charged for the acquisition of the stock. C. The Recommendation of the Administrative Law Judge The ALJ recommended that the Company be allowed to recover the expenses associated with the donation of the appreciated stock as if it were cash, having found that the donation was reasonable and in the public interest. D. Commission Action Recovery for charitable contributions is permitted, provided recovery is limited to 50% of the contribution and the Commission deems the contribution prudent. 25 The Commission will deny cost recovery for the Company s indirect contribution to the United Way. The benefit to ratepayers of this expense is at best unclear. The tax advantage for the charitable donation accrued to the Service Company. 24 Minn. Stat. 216B Minn. Stat. 216B.16, subd

35 In effect, the Company appears to be asking for the ratepayers to completely subsidize a charitable donation when the benefit to ratepayers will, at best, be extremely attenuated and the magnitude of benefit flowing to ratepayers is unaccounted for in the record. The Commission is therefore unpersuaded that the expense is reasonable and appropriate for recovery from Minnesota ratepayers. XVIII. Property Taxes A. Introduction CenterPoint included in its test year an estimated $6,383,543 increase in property taxes. According to the Company, the increase is attributable to a higher market value of utility property. During the contested case, the Company recalculated its test-year property-tax expense based on Truth in Taxation statements it had received. Built into its revised estimated increase was a 2% upward adjustment (equaling $593,802) for possible increases resulting from school referenda passed in November elections. The Department provisionally accepted the Company s 2% estimate, but recommended that the Company submit evidence of the actual school referendum increases prior to the close of the record. The school referendum information was not added to the record. B. Positions of the Parties Because the Department was unable to confirm the amount of increase attributable to school referenda, the Department recommended that the placeholder 2% increase be removed from the test-year property-tax expense. It also recommended that the expense be reduced by $1,451,686 to conform to the Company s updated calculations. The Department calculated that both reductions would total a decrease of $2,045,488. CenterPoint did not object to adjusting its test-year property-tax expense to match its updated calculations. It argued that its estimated 2% increase was reasonable given the large number of referenda and stated that it had not received the updated statements necessary to substantiate the increase in the record. C. The Recommendation of the Administrative Law Judge The ALJ agreed with the Department and recommended a reduction in property-tax expense of $2,045,488. D. Commission Action The Commission agrees with and will adopt the ALJ s recommendation. CenterPoint may recover its property-tax expenses, as adjusted, and excluding the 2% referendum contingency. The burden of justifying this expense rests on the Company, and doubt must be resolved in the ratepayers favor. 26 There does not appear to be a record basis or compelling analysis to support the size of the adjustment attributed to referenda. Accordingly, the approved recovery will exclude the 2% referendum adjustment. 26 Minn. Stat. 216B

36 COST OF CAPITAL ISSUES Utilities meet their capital needs by issuing stock, known as equity, and by incurring long-term and short-term debt. These three components make up the utility s capital structure. Generally, equity is the most expensive form of financing, followed by long-term debt and then short-term debt. The percentage of the capital structure made up of each of these three components therefore has a substantial impact on costs and rates, as does the cost assigned to each component during the ratemaking process. In this case the Company, the Department, and the OAG proposed different ratios for equity, long-term debt, and short-term debt and different costs for the equity component of the capital structure. The Department and the Company also proposed different costs for short-term debt. All parties agreed on the cost of long-term debt, and the administrative law judge concurred. The Commission will address the issues of capital structure and the cost of each of its components below. XIX. Capital Structure A. Introduction CenterPoint is a division of CenterPoint Energy Resources, which is a subsidiary of CenterPoint Energy, Inc., a public-utility holding company. The Company therefore has no capital structure of its own and must be assigned a hypothetical capital structure for ratemaking purposes. In April 2003, the Commission issued an order (the ratepayer-protection order), setting broad capital-structure parameters for the Company s predecessor, Minnegasco, which became the present company under a subsequent restructuring. Those parameters were part of a larger set of ratepayer protections adopted in response to financial difficulties at the parent company, and they remain in force today. Those parameters are set forth below: 7. Without releasing Minnegasco from commitments and requirements made and imposed in this and related dockets, Minnegasco shall adhere to the following commitments: a. on its Minnesota jurisdictional books and for regulatory purposes, Minnegasco shall recognize capitalization structure and applicable cost of financing typical of an A rated utility; and b. Minnegasco shall maintain approximately a 50/50 debt equity ratio, with each debt instrument reflecting the costs associated with that of an A-rated utility at the time that the debt instrument is booked In the Matter of an Inquiry into Possible Effects of the Financial Difficulties at Reliant Energy, Inc. on Reliant Energy Minnegasco and Its Customers, Docket No. G-008/CI , Order Requiring Filings to Protect Minnesota Ratepayers, at 12 (April 8, 2003). 30

37 In this rate case, the Company, the Department, and the OAG all supported different capital structures, but all recommended an equity ratio of at least 50%, citing the ratepayer-protection order. And, to varying degrees, all three parties relied for corroboration on the capital structures of the companies in the proxy groups they used in their respective Discounted Cash Flow (DCF) cost-of-equity studies. Proxy groups are used in the DCF analytical model to determine the cost of equity for utilities like CenterPoint that do not have their own capital structures and do not issue their own stock. Proxy groups are made up of utilities with business and risk profiles that match as closely as possible the business and risk profile of the utility under study. By the end of the case, the Company and the Department had the same seven-company proxy group, and the OAG had a five-company proxy group, with all five companies also appearing in the Company and Department groups. The capital structure recommended by each party, the capital structure recommended by the Administrative Law Judge, and the average capital structure of the two proxy groups are set forth below: Capital Structure Component Company Department OAG ALJ DOC/ Company Proxy Group 31 OAG Proxy Group Common Equity 53.43% 52.60% 50.00% 52.60% 49.77% 48.05% Long-Term Debt 44.71% 39.70% 48.00% 44.71% 41.99% 44.80% Short-Term Debt 1.86% 7.70% 2.00% 2.69% 8.24% 7.15% B. Positions of the Parties 1. The Company The Company proposed a capital structure with more equity, more long-term debt, and less short-term debt than its current capital structure: Current Capital Structure Company-Proposed Capital Structure Common Equity 52.60% 53.43% Long-Term Debt 40.16% 44.71% Short-Term Debt 7.24% 1.86% The Company argued that it needs to rely more heavily on long-term financing especially equity, but also long-term debt as it invests more heavily in the long-term capital improvements needed to update its infrastructure and meet new safety and reliability standards. The Company also pointed out that since the financial crisis of 2008, corporations have generally been removing as much debt from their capital structures as possible and argued that higher equity ratios reduce risk, thereby reducing the costs of both debt and equity. The Company argued that its 53.43% proposed equity ratio was close to the 50/50 guideline set in the ratepayer-protection order and that its proposed increase in the long-term-debt ratio and accompanying reduction in short-term debt would benefit ratepayers because the Company

38 could secure long-term debt on very favorable terms. The Company pointed out that, with its most recent long-term-debt issuance, its overall average cost of long-term debt fell more than 50 basis points below Moody s A-Rated Utility Bond index. 28 The Company also argued that its proposed equity ratio was within the range of those of both proxy groups used by the parties in their DCF analyses. And finally, the Company argued that reducing its equity ratio from current levels, as recommended by the OAG, could be viewed by investors and ratings agencies as a lack of regulatory support, which could in itself increase the cost of capital. 2. The Department The Department proposed a capital structure with the same amount of equity, slightly less long-term debt, and slightly more short-term debt than the Company s current capital structure, set in its last rate case: Current Capital Structure Department-Proposed Capital Structure Common Equity % % Long-Term Debt % % Short-Term Debt 7.24 % 7.70 % The Department recommended retaining the current equity ratio of 52.60% for two main reasons: a) that ratio is and was found in the last rate case to be compliant with the guideline set in the ratepayer-protection order; and b) that ratio is closer to the ratios in the DCF proxy group used by the Company and the Department than the Company s proposed 53.43% ratio. In fact, the 53.43% ratio is higher than the ratio of all but one company in that group and higher than the group s average equity ratio by 6.55 percentage points. 29 The Department recommended setting the short-term-debt ratio at 7.70%, the Company s actual short-term-debt ratio for fiscal year The Department opposed the Company s proposal to reduce that ratio from the currently authorized 7.24% to 1.86%, arguing that the reduction was unsupported by record evidence, inconsistent with recent Company short-term debt levels, and inconsistent with industry-average short-term debt levels. The Department pointed out that the Company s proposed short-term debt ratio was 6.38 percentage points below the Company/Department proxy group s average 30 and was much lower than the Company s actual short-term-debt ratio of 7.70% in the fiscal year preceding the rate case, FY CenterPoint Initial Brief, at Department Initial Brief, at The average short-term-debt ratio of the proxy group was 8.24%. Department Reply Brief, at

39 In FY 2014, the Company carried average monthly short-term debt of $58,700,000; for the test year, it proposed to carry average monthly short-term debt of only $17,000, The Department argued that it would be unreasonable to accept the claim that the Company s short-term financing needs would drop that dramatically in that time frame. The Department claimed that reducing short-term debt to the extent the Company proposed carried the risk that the Company would finance short-term needs with more expensive long-term debt. The agency pointed to testimony and responses to discovery requests in which the Company explained that the increase in long-term debt would reduce its need for short-term debt, 32 which it argued corroborated this concern. 3. The OAG The OAG recommended setting the Company s equity/debt ratio at 50/50, the ratio the Company was ordered to approximate in the 2003 ratepayer-protection order, and adjusting its proposed long-term and short-term debt ratios proportionally: Current Capital Structure OAG-Proposed Capital Structure Common Equity % % Long-Term Debt % % Short-Term Debt 7.24 % 2.00 % The OAG emphasized that the 53.43% equity ratio proposed by the Company fell substantially above the average equity ratio for both DCF proxy groups, and challenged the Company s reliance on the ratio being within the range of those ratios, arguing that the purpose of a DCF proxy group is to provide a credible average or composite of companies similar to the one being examined, making the average more probative than the range. The OAG pointed out that the Company s proposed equity ratio is higher than the equity ratio currently assigned to any other Minnesota natural-gas utility. In fact, the only Minnesota gas or electric utility with a higher equity ratio (54.29%) is Minnesota Power, an electric utility with a very different risk profile, whose capital structure was set during the recession following the financial crisis of Finally, the OAG opposed the Company s proposal to reduce its short-term debt ratio to 1.86%. Like the Department, the OAG argued that such a reduction was unsupported in the record and inconsistent with the Company s own recent financial history. The OAG also pointed out that the proposal was inconsistent with Company expectations for the future, since the Company reported that it expected to double its short-term debt the month after the test year ends (October 2016) and to double it again by December Department Exceptions, at Department Reply Brief, at OAG Initial Brief, at OAG Initial Brief at

40 The OAG argued that this pattern of past and future short-term debt levels together with a lack of record evidence supporting the Company s proposed 1.86% short-term debt level required rejecting it. C. The Recommendation of the Administrative Law Judge The Administrative Law Judge recommended adopting a capital structure incorporating the Department s recommended equity ratio, the Company s recommended long-term debt ratio, and a short-term debt ratio closest to the one recommended by OAG: Current Capital Structure Capital Structure Recommended by the Administrative Law Judge Common Equity % % Long-Term Debt % % Short-Term Debt 7.24 % 2.69 % He found the Department s proposed equity ratio the most reasonable in the record, because it had been found reasonable in the Company s last rate case, was closer than the Company s proposal to the 50/50 guideline established in the ratepayer-protection order, and was closer than the Company s proposal to the average equity ratios in the DCF proxy group used by the Company and the Department. He rejected the Company s proposal as unnecessarily increasing ratepayer costs. He recommended adopting the Company s proposed long-term debt ratio, finding it was consistent with the long-term debt ratios of companies in the DCF proxy group, similar to the long-term debt ratio adopted in the Company s last rate case, and consistent with the increase in long-term capital investment projected for the test year. He also found that rejecting the Company s preferred long-term debt ratio could signal a lack of regulatory support to investors and ratings agencies, thereby increasing capital costs. Finally, the Administrative Law Judge recommended a short-term debt ratio of 2.69 as the most reasonable alternative. He found that the 2.69 ratio was similar to the proposals of the Company and the OAG, but was closer than the Company s proposal to the average ratio in the DCF proxy groups. He also found it similar to the short-term debt ratio approved in the rate case preceding the Company s last rate case. D. Commission Action The Commission will set the Company s capital structure at 50% equity, 42.3% long-term debt, and 7.7% short-term debt, based on the following considerations: a) The ratepayer-protection order requires the Company to maintain an approximately 50% equity ratio. b) The average capital structures of both proxy groups confirm the reasonableness of a 50% equity, 42.3% long-term debt, and 7.7% short-term debt capital structure. c) The Company s actual, short-term-debt levels immediately preceding the test year and its projected, short-term-debt levels immediately following the test year do not support its proposed 1.86% short-term-debt ratio and in fact point to a short-term-debt level approximating 7.7%. 34

41 The equity and debt ratios will be considered in turn. 1. The Equity Ratio The Commission concurs with the parties and the Administrative Law Judge that the ratepayer-protection order s 50% equity-ratio guideline remains the starting point and touchstone in setting the Company s capital structure. That guideline and related ratepayer protections were developed in a fact-intensive proceeding with the active participation of the Company, the Department, and the OAG. They were developed to ensure that the Company could deliver safe, adequate, and affordable natural-gas service to Minnesota ratepayers, notwithstanding serious financial difficulties at that time at CenterPoint s parent company. They have proven effective in meeting that purpose, they remain in effect, and they will be honored here. No party recommended setting the equity ratio below the 50% guideline, and the Commission concurs. At the same time, the Commission concurs with the OAG that the record does not demonstrate a need to set the equity ratio above that guideline and since higher equity levels normally translate directly into higher rates the Commission will not set the equity ratio above 50% in the absence of demonstrated need. Further, the best evidence in the record on what CenterPoint s capital structure would be if it were a free-standing utility the average equity ratios of the two carefully vetted proxy groups confirms the reasonableness of a 50% equity ratio. Both proxy groups the one that emerged from the separate analyses of the Department and the Company and the one that emerged from the analysis of the OAG had average equity ratios below 50%. The average for the Company/Department proxy group was 49.77%, and the average for the OAG proxy group was %. 35 These averages confirm the reasonableness of the 50% equity ratio proposed by the OAG and adopted by the Commission. The Company argued that simply being within the range of the equity ratios in the proxy groups was adequate evidence of reasonableness, but the Commission does not agree. Proxy-group averages have much higher probative value than proxy-group ranges; the purpose of a proxy group is to provide a representative average or composite to stand in for the company being studied. Here, that representative average confirms the reasonableness of the 50% equity ratio commended by the ratepayer-protection order, recommended by the OAG, and adopted by the Commission. Further, the Commission attaches less probative value than the Department and the Administrative Law Judge to the 52.60% equity ratio set in the Company s last rate case. While that equity ratio was just and reasonable when set, it is a given that economic conditions are dynamic and that rate-case decisions will be, too. Rate-case decisions are always made on the basis of the best evidence available, but the best evidence available especially on issues sensitive to current economic conditions changes over time. Dated capital-structure decisions are therefore more useful as broad reasonableness checks than as evidence of current capital conditions. 35 ALJ s Report 68; OAG Initial Brief, at

42 Finally, the Commission does not share the Company and the Administrative Law Judge s concern that setting the equity ratio below the one set in the last rate case might signal a lack of regulatory support to the financial community, reducing their confidence in CenterPoint and raising its cost of capital. Ratings agencies, financial analysts, and institutional investors are sophisticated observers of market and regulatory phenomena; the Commission is confident they will perceive today s decision as no less supportive, objective, and fact-driven than those made in the past. 2. Short-Term and Long-Term Debt Ratios The Commission will set the Company s short-term and long-term debt ratios at 7.7% and 42.3% respectively, based on the following considerations: a) The Company s proposed 1.86% short-term-debt ratio is substantially lower than its actual, short-term-debt ratio during the fiscal year preceding the test year and its projected, short-term-debt ratio immediately following the test year, with no explanation in the record for the one-year discrepancy. b) The Company s proposed 1.86% short-term-debt ratio is substantially lower than the average short-term-debt ratios of the two proxy groups. c) The Company s proposed 1.86% short-term-debt ratio is otherwise unsupported in the record. d) The 42.3% long-term-debt ratio is the automatic result of the 50% equity ratio and 7.7% short-term-debt ratio adopted by the Commission. The Company proposed to heavily increase its reliance on long-term as opposed to short-term debt, reducing its short-term-debt ratio from the currently authorized 7.24% to a proposed 1.86%. Since long-term debt is significantly more expensive than short-term debt, this proposal carries significant rate implications. The Company offered two reasons for the proposed shift: (1) long-term debt is a better match for the long-term capital projects that much of its debt will be financing over the test year; and (2) a recent long-term-debt issuance on very favorable terms substantially reduced its need for short-term debt. Neither of these explanations is adequate to support the shift. First, the Company did not attempt to match specific new capital projects or new capital projects in the aggregate with specific needs for new long-term debt. It relied instead on the general claim that large, pending capital investments justified increased reliance on long-term debt. Those pending investments, however, are part of an ongoing, multi-year, infrastructure-investment initiative that was already underway in its last rate case, where the short-term debt ratio was set at 7.24%. Similarly, those investments were being made throughout fiscal-year 2014, when, as discussed in more detail below, short-term-debt levels were substantially higher than 1.86%. The matching principle, therefore, does not justify this marked departure from the existing short-term debt ratio. Second, the Company s large, recent long-term-debt issuance however favorable its terms does not justify a shift from short-term to long-term financing. Long-term debt remains more expensive than short-term debt, and it therefore remains important to set the long-term and short-term debt ratios appropriately. 36

43 Third, the proposed 1.86% short-term-debt ratio is inconsistent with the Company s recent short-term-debt levels. In fiscal-year 2014, the Company had an actual short-term-debt ratio of 7.7% and carried average monthly short-term debt of $58,700,000. For the test year, on the other hand, it proposed to carry average monthly short-term debt of only $17,000,000, with a short-term-debt ratio of 1.86%. 36 This substantial drop in short-term-debt levels is not explained in the record. Further, the Company expects its short-term debt to double the month after the test year ends (October 2016) and to double again by December Again, the record does not explain the one-year, test-year disruption in the Company s short-term debt patterns, nor does it support a finding that that disruption is reasonable and appropriate. Fourth, the proposed 1.86% short-term-debt ratio is deeply inconsistent with industry norms. It is 6.38 percentage points below the average short-term-debt ratio of the utilities in the Company and Department s proxy group and 5.29 percentage points below the average in the OAG s proxy group. It is also 5.38 percentage points below the short-term-debt ratio authorized in the Company s last rate case. In short, the proposed 1.86% short-term-debt ratio is not supported by record evidence, and the best evidence in the record industry norms, the Company s recent short-term-debt history, and the Company s projected post-test-year, short-term debt activity all point to a short-term-debt ratio in the range of the Company s currently authorized short-term-debt (7.24%), industry averages (7.15% 8.24%), and the Company s last fiscal-year average (7.7%). The Commission concludes that the most reasonable number is the 2014 fiscal-year average, 7.7%, since it both directly reflects actual Company experience and is squarely within the range of the other ratios reasonably supported in the record. The Commission will therefore set the short-term-debt ratio at 7.7%, which, when factored in with the 50% equity ratio, results in a long-term-debt ratio of 42.3%. The 42.3% figure, too, is supported in the record, its reasonableness confirmed by the average long-term-debt ratios of the two proxy groups (41.99% and 44.80%) and the Company s currently authorized long-term-debt ratio (40.16%). The resulting capital structure is set forth below: Component Ratio Equity 50.0% Long-Term Debt 42.3% Short-Term Debt 7.7% 36 Department Exceptions, at OAG Initial Brief, at

44 XX. Return on Equity A. Introduction In determining just and reasonable rates, the Commission is required to give due consideration to the public need for adequate, efficient, and reasonable service and to the need of the public utility for revenue sufficient to enable it to meet the cost of furnishing service, including adequate provision for depreciation of its utility property used and useful in rendering service to the public, and to earn a fair and reasonable return upon the investment in such property. 38 One of the critical components of that fair and reasonable return upon investment is the return on common equity, which together with debt finances the utility infrastructure. The Commission must set rates at a level that permits stockholders an opportunity to earn a fair and reasonable return on their investment and permits the utility to continue to attract investment. In short, the Commission must determine a reasonable cost of equity and factor that cost into rates. It would normally begin by examining the price of the utility s stock, but CenterPoint is a division company of CenterPoint Energy Resources, which is a subsidiary of CenterPoint Energy, Inc., and therefore has no publicly traded common stock. Its cost of common equity essential to determining overall rate of return and the final revenue requirement must therefore be inferred from market data for companies that present similar investment risks. B. The Analytical Tools CenterPoint, the Department, and the OAG conducted cost-of-equity studies and based their analysis on groups of utilities they considered similar enough to CenterPoint to serve as proxies in determining the Company s cost of equity. By the end of the case, the Company and the Department had the same seven-company proxy group, and the OAG had a five-company proxy group, with all five companies also appearing in the Company and Department groups. All three parties did thoroughgoing studies using the Discounted Cash Flow (DCF) analytical model, on which this Commission has historically placed its heaviest reliance. All three also conducted studies using the Capital Asset Pricing Model (CAPM), which the Commission has historically used as a secondary, corroborating resource. The Company also conducted a third analysis using the Bond Yield Plus Risk Premium Model (RP), which the Commission has historically relied on less heavily, considering the model prone to producing volatile and unreliable outcomes. The DCF model uses the current dividend yield and the expected growth rate of dividends to determine what rate of return is high enough to induce investment. The model is derived from a formula used by investors to assess the attractiveness of investment opportunities using three inputs dividends, market equity prices, and earnings/dividend growth rates. Its two basic variants are the Constant Growth DCF, the classic version, and the two-growth DCF, designed for situations in which the short-term, projected earnings growth rates may not be expected to 38 Minn. Stat. 216B.16, subd. 6 (emphasis added). 38

45 continue in the long run. The two-growth model uses one growth rate for an initial period, followed by a different growth rate for the long term. The CAPM model estimates the required return on an investment by determining the rate of return on a risk-free, interest-bearing investment; adding a historical risk premium determined by subtracting that risk-free rate of return from the total return on all market equities; and multiplying the remainder by beta, a measure of the investment s volatility compared with the volatility of the market as a whole. The RP model determines the cost of equity by adding to current corporate bond yields a premium reflecting the greater returns realized by equity holders over various historical periods. C. Positions of the Parties 1. The Company The Company proposed a return on equity of 10.3%, based on a multi-factor analysis not directly tied to the outcome of any specific analytical model, but based on the professional judgment of its expert witness. The Company conducted a classic DCF study, which assumes constant growth in earnings and dividends, but adjusted the trading period normally used to determine two inputs stock prices and dividends from the most recent 30-day period to the average of the most recent 30-day, 60-day, and 90-day periods; the Company argued that using multiple trading periods increased accuracy by reducing volatility. Similarly, the Company adjusted the model s growth-estimate input by substituting its expert s retention-growth estimate for one of the three publicly available earnings-growth estimates normally used. The Company also conducted a Multi-Stage DCF study, developing inputs for multiple time periods and extrapolating future financial performance from the results. The Company conducted a CAPM study, using 30-year treasury notes (instead of the more typical 20-year notes) as the risk-free asset the study requires. It conducted an RP study, also using 30-year treasury notes as the baseline asset, and using coefficients that assumed no change in investors expectations or behavior over time. The Company s expert weighed the results of these studies. He also factored in and adjusted for business risks specific to the Company s small size, and market risks specific to current economic conditions (e.g., the perceived likelihood of rising interest rates and the continuing uncertainty regarding future Federal Reserve policy decisions). Both adjustments were upward adjustments. Neither adjustment was given a specific numerical value, but both were included in the constellation of factors the expert weighed in calculating the appropriate cost of equity. Finally, the Company s ultimate recommendation of a 10.3% return on equity included an adjustment for flotation costs, the costs of issuing securities, since those costs result in a utility receiving less than the full sales price for shares issued. 2. The Department The Department proposed a return on equity of 9.63%, the result of applying a two-growth DCF model to the proxy group, using the average (mean) of the three growth-rate projections for each 39

46 company in that group, and adjusting the final number to include flotation costs. 39 The Department argued that relying on the DCF model to calculate return on equity was the most reasonable approach, because that model is the most objective, transparent, and historically reliable analytical tool available. The Department noted that it had also conducted two CAPM analyses on the companies in the proxy group one using the classic CAPM model and the other the Empirical CAPM model, which is tailored for companies with betas lower than one as reasonableness checks. The agency stated that these studies confirmed the general accuracy of its DCF results but yielded returns on equity too low to be seriously considered for CenterPoint. The Department rejected the proposition that the cost of equity should be adjusted to reflect company size, arguing that adjusting for one company-specific characteristic on a stand-alone basis would defeat the purpose and dilute the validity of using a proxy group in DCF modeling. The agency also stated that the revenue-decoupling mechanism adopted by the Company in its last rate case cuts against a size adjustment, since decoupling eliminates much of the volatility risk the size adjustment would be adopted to reflect. Similarly, the Department opposed any adjustment for the uncertainties inherent in current economic conditions, arguing that the stock prices used in the DCF formula fully reflect investors expectations and risk assessments. The Department also challenged the Company s execution of its DCF, CAPM, and RP studies, claiming that the following practices, among others, significantly reduced their reliability: using overly long trading periods to project growth rates in the DCF, substituting a less reliable and more subjective retention growth rate for one of the DCF earnings growth rates, using an unduly subjective and largely unsupported payout ratio in the DCF, using 30-year treasury notes instead of the less risky and more commonly used 20-year notes as the riskless asset in the CAPM, and assuming that both coefficients in the RP analysis would be stable over time and would not change due to investors changing expectations or behaviors as federal fiscal or monetary policies changed. Finally, the Department argued that since the Company did not explain how it weighted or balanced any modelling outcome against any other modelling outcome or against any other factor, its recommended return on equity lacked transparency, rigor, and evidentiary support and should be rejected. 3. The OAG The OAG proposed a return on equity of 9.34%, based on a two-growth DCF study using a different proxy group than the one used by the Company and Department. The Office also opposed adjusting the DCF results upward to include a flotation adjustment. 39 The Company and the Department concurred in setting flotation costs at 3.689%. 40

47 The OAG urged the Commission to base its decision on the DCF analytical model, arguing that its predictability, replicability, and objectivity made it superior to other analytical models, as illustrated by the fact that all three parties unadjusted DCF results fell within a 40-basis-point range. 40 The OAG also pointed to the Commission s heavy historical reliance on the DCF model and on the close alignment between that model s forward-looking focus and the forward-looking focus of the ratemaking process. The OAG challenged the Company s execution of the DCF, CAPM, and RP analytical models on the same grounds as the Department, citing the use of overly long trading periods to project growth rates, substituting a less reliable and more subjective retention growth rate for an earnings growth rate, using a subjective and inadequately supported payout ratio, using 30-year treasury notes, instead of the less risky and more commonly used 20-year notes; and assuming that both coefficients in the RP analysis would be stable over time and would not reflect investors reactions to changing federal fiscal or monetary policies. The OAG also opposed the Company s proposed upward adjustments for company size and current economic conditions, for the same reasons the Department opposed them, and pointed out that both adjustments had been disallowed in the Company s last rate case. The Office also opposed the Company s ultimate reliance on the professional judgment of its expert, as inadequately explained and insufficiently supported by record evidence. The OAG opposed the Company s and Department s support of a flotation adjustment. The Office pointed out that the Company s parent had not incurred flotation costs since 2010, that none of the other regulated utilities owned by its parent appeared to have flotation costs included in their cost of equity, and that investors, as well as companies issuing securities, incurred issuance costs. Finally, the OAG argued that the five-member proxy group it used as a proxy to determine CenterPoint s cost of equity more accurately reflected the risk profile of CenterPoint than the seven-member group used by the Company and the Department. The OAG s group omitted two companies South Jersey Industries and New Jersey Resource Corporation that appeared in the other group. All three parties used similar screens in choosing companies for their proxy groups. All companies in the three proxy groups share the following characteristics: They are publicly traded on a stock exchange. They are listed on the Compustat Research Insight database. They are assigned to Standard Industrial Classification Code 9424 Natural Gas Distribution. They had at least 60% of total net operating income from natural-gas distribution operations in They have a Standard & Poor s bond rating within the range of BBB to A OAG Initial Brief, at One company in the proxy group, New Jersey Resources Corporation, is a holding company, with no bond rating of its own, but its natural-gas distribution company has a bond rating of A. 41

48 The OAG, however, required that proxy companies have not just 60% of total operating income from natural-gas distribution operations, but 50% of total revenues and 75% of total assets associated with those operations. The Office contended that these two criteria were necessary to screen out companies whose unregulated operations were significantly larger than CenterPoint s and whose risk profiles were therefore significantly different from CenterPoint s. 42 These two criteria disqualified two companies that were in the other parties proxy groups: New Jersey Resource Corporation and South Jersey Industries, Inc. D. The Recommendation of the Administrative Law Judge The Administrative Law Judge recommended a return on equity of 9.63%, concurring with the Department s analysis. He concurred with the Department and the OAG that the DCF model was the first and best resource for determining the cost of equity because of its transparency, historical reliability, replicability, and predominant reliance on publicly available as opposed to subjectively developed inputs. He pointed to the relative convergence of the three parties DCF results unlike their CAPM results as demonstrating the greater objectivity of the DCF model. He found that the Company s and the Department s proxy group was more appropriate than the OAG s group, finding that the two companies the OAG omitted were comparable to CenterPoint, that their inclusion resulted in a larger proxy group less vulnerable to the impact of anomalies specific to individual companies, and that the Commission had approved of their inclusion in proxy groups in previous rate cases. He found the Department s and OAG s use of 30-day trading periods in their DCF analyses superior to the Company s longer trading periods, given the absence of evidence of volatility or anomalies within those 30-day periods, and he found that recent, 30-day trading periods best captured all publicly available information and investor expectations regarding Company performance. He concurred with the Department and the OAG that retention growth rates were not a reasonable substitute for forecasted earnings growth rates in DCF modeling. He concurred with the OAG and the Department that Company proposals to increase the return on equity to reflect Company size and current economic conditions were unreasonable and unsupported by substantial evidence in the record. He concurred with them that the payout ratios assumed in the Company s DCF analysis were unreasonable, inconsistent with industry projections, and unsupported in the record. He also agreed that the Company s use of 30-year treasury notes as the riskless asset in its CAPM analyses was unsupported in this record, unreasonable, and inconsistent with past Minnesota practice. He concurred with the Company and the Department that flotation costs were properly included in the cost of equity, because without them, the Company would be credited with less than the full value of issued stock, diluting its available capital and potentially depriving it of the opportunity to earn its authorized rate of return. 42 Some 90% of CenterPoint s revenue and 98% of its assets are related to regulated natural-gas distribution. Initial Brief, OAG at

49 He found the Department s execution of the DCF model reasonable in every respect, supported by record evidence, and consistent with historical Minnesota practice. He found that the Department had fully supported and demonstrated the reasonableness of its recommendation. Conversely, he found that the Company and the OAG had failed to support and demonstrate the reasonableness of some parts of their positions the Company s 90-day trading periods, the Company s proposed upward adjustments for company size and economic conditions, the payout ratios in the Company s DCF analysis, the OAG s small proxy group, and the OAG s rejection of flotation costs. He recommended adopting the Department s proposed 9.63% return on equity as demonstrably reasonable, supported by the record, and consistent with past Minnesota regulatory practice. E. Commission Action The Commission concurs with nearly all the Administrative Law Judge s closely reasoned findings, conclusions, and recommendations on the cost of equity. The Commission agrees that the Department s cost-of-equity studies are methodologically transparent, analytically sound, ably executed, and supported by substantial evidence in the record. 43 They are the best evidence in the record on the cost of equity, and the Commission concurs with the Administrative Law Judge s acceptance and adoption of them, with one exception: The Commission will not include a flotation adjustment in CenterPoint s cost of equity, since the Company s parent has not issued stock since 2010 and has stated that it has no plans to issue stock in the reasonably foreseeable future. Flotation costs are sometimes added to the cost of equity to reflect the fact that companies do not receive the full price of the stock they issue the amount they receive is less than the face value of the stock due to legal and issuance fees incurred out-of-pocket and underwriting fees withheld from the proceeds paid to the company. Flotation costs in the amount of these fees are sometimes added to the cost of equity to credit the company with the full amount of the issuance. Setting the cost of equity is fact-intensive and record-specific, and the Commission concurs with the OAG that this record does not support the addition of flotation costs to the cost of equity. The record demonstrates that CenterPoint s parent company has not issued stock since 2010 and has in fact issued stock only twice since its restructuring in Further, the parent company has no plans to issue stock in the foreseeable future. 43 The Commission will make a technical correction to ALJ Finding 133, as set forth below: 133. For the second component of the dividend yield DCF calculation, the Department and the OAG relied upon the expected growth rate in dividends provided by three respected and widely-used investment research services: Zack s Investment Research (Zack s); Value Line; and Thomson First Call Consensus (Thomson). Specifically, it they used the three projected earnings growth rates (lowest, average and highest) provided by Zack s, Value Line, and Thomson. 44 OAG Ex. 404, at 24 (Urban Direct). 43

50 The Company offered no evidence on the actual amount or ongoing financial impact of the costs incurred in connection with these earlier stock issuances. It relied instead on the argument that a flotation adjustment was required to prevent capital dilution and ensure a reasonable opportunity for the Company to earn its authorized rate of return. The Commission concurs with the OAG that the Company has not demonstrated the need for an additional $708,366 per year in flotation costs 45 to ensure it the opportunity to earn a fair return on its investment. First, as the OAG points out, the Company is one of several siblings providing natural-gas-distribution service in six states. None of the other states currently incorporate flotation adjustments into returns on equity as part of their rate-case processes, 46 and the sibling companies continue to attract the capital required to provide natural-gas-distribution service meeting regulatory standards. Second, the Commission concurs with the OAG that the stock-dilution argument is effectively countered by the fact that, under current economic conditions, the shares of natural-gasdistribution companies, including CenterPoint s parent company, are selling at prices far exceeding their initial issuance price or book value; the OAG s expert witness testified that these stocks are currently selling at more than 1.5 times their book values. 47 Existing shareholders therefore realize a gain, not a loss, in the book value of their shares when stock is issued. Finally, the Company has stated publicly that it intends to file another rate case within two years of this one. At that point it will have ample opportunity to demonstrate on the record the nature, amount, and financial impact of all costs associated with any completed or planned stock issuance. In the absence of that kind of factual detail, the Commission will disallow a flotation adjustment in this case and set the Company s cost of equity at 9.49%, the figure resulting from the Department s analysis, minus the flotation adjustment. XXI. Cost of Long-Term Debt The Company proposed a 5.38% cost of long-term debt, based on recent long-term debt issuances at favorable rates. The 5.38% interest rate is over 50 points below the interest-rate benchmark in Moody s A-Rated Utility Bond Index. No one challenged the reasonableness of the Company s proposal. The Department recommended adopting it as reasonable, and the Administrative Law Judge agreed. The Commission concurs and will set the cost of long-term debt at 5.38%. 45 Id., at Mississippi does incorporate a type of flotation adjustment as part of an automatic review of return on equity outside the rate-case process; that review can result in a rate case or the sharing of earnings, and is therefore not comparable to this proceeding. OAG Ex. 422, at 3 (Urban Rebuttal). 47 OAG Ex. 404, at 28 (Urban Direct). 44

51 XXII. Cost of Short-Term Debt A. Introduction Since the Company is not a free-standing entity, but a division of a subsidiary of a public-utility holding company, it does not go to the market for its short-term debt. Instead, it receives its short-term financing from its parent company, whose projected, test-year cost of short-term debt is 5.88%. No one proposes using that cost for ratemaking purposes, however, since the ratepayer-protection order discussed in the Capital Structure section above limits CenterPoint s cost of debt to the applicable cost of financing typical of an A rated utility. 48 Instead, the Company set its short-term-debt cost at the rate in its revolving credit agreement with its parent: the London Interbank Offered Rate (LIBOR), a widely used interest-rate benchmark, plus a spread of Using this rate, the Company calculated a composite test-year short-term debt cost of 1.62% The Company and the Department disagreed on the cost the Commission should assign to the Company s short-term debt. The cost of short-term debt set in the last rate case, the Company s 2014 average cost of short-term debt, the parties positions, and the Administrative Law Judge s recommendation are set forth below: Last Rate 2014 Actual Case Average Company Department ALJ 0.36% 0.65% 1.62% 0.65% 1.62% B. Positions of the Parties 1. The Company The Company recommended a short-term debt cost of 1.62%, based on a composite of projected LIBOR rates during the test year and the spread in its revolving credit agreement; this rate represented a 97-basis-point increase over the Company s average, actual short-term-debt cost for calendar year The Company argued that this 97-basis-point increase was reasonable, given general expectations that interest rates would rise during the test year and given similar projections by financial research firms e.g., Blue Chip Financial Forecasts projections that rates for one-year treasury notes would rise by 118 basis points between first quarter 2015 and second quarter 2016 and that LIBOR would increase by 114 basis points during the same period. 2. The Department The Department claimed that the Company had failed to meet its burden of proof to demonstrate that 1.62% was a reasonable cost of short-term debt for an A-rated utility, as required under the ratepayer-protection order. 48 In the Matter of an Inquiry into Possible Effects of the Financial Difficulties at Reliant Energy, Inc. on Reliant Energy Minnegasco and its Customers, Docket No. G-008/CI , Order Requiring Filings to Protect Minnesota Ratepayers, at 12 (April 8, 2003). 45

52 The Department emphasized that the 1.125% spread over LIBOR in the short-term-debt formula exceeded the Company s actual, average cost of short-term debt in calendar year The agency pointed out that the LIBOR rate was less than half the Company s proposed short-term-debt rate on the day the Department filed its direct testimony. And the Department stated that in the past it has supported short-term-debt rates that were much closer to the LIBOR rate than the Company s proposal in this case. In the absence of better evidence on the reasonable cost of short-term debt during the test year, the Department recommended using the Company s end-of-2014 actual, average short-term-debt cost, 0.65%. C. The Recommendation of the Administrative Law Judge The Administrative Law Judge found that the Company s proposed cost of short-term debt was reasonable and recommended adopting it. He concurred with the Company that its use of a projected test year justified reliance on projected short-term interest rates and noted that the Company s proposed cost of short-term debt represented a smaller increase over 2014 actual costs than the projected increases in LIBOR and one-year-treasury-note rates for the same time period. D. Commission Action The Commission will set the cost of short-term debt at 0.65%, the most recent actual cost of short-term debt in the record, finding that the Company has failed to demonstrate by a preponderance of evidence the reasonableness of its proposed 1.62% cost of short-term debt. As explained in the earlier discussion of capital structure, the ratepayer-protection order of 2003 is the starting point and touchstone in setting the cost of debt. That order provides as follows: 7. Without releasing Minnegasco from commitments and requirements made and imposed in this and related dockets, Minnegasco shall adhere to the following commitments: a. on its Minnesota jurisdictional books and for regulatory purposes, Minnegasco shall recognize capitalization structure and applicable cost of financing typical of an A rated utility; and b. Minnegasco shall maintain approximately a 50/50 debt equity ratio, with each debt instrument reflecting the costs associated with that of an A-rated utility at the time that the debt instrument is booked In the Matter of an Inquiry into Possible Effects of the Financial Difficulties at Reliant Energy, Inc. on Reliant Energy Minnegasco and its Customers, Docket No. G-008/CI , Order Requiring Filings to Protect Minnesota Ratepayers, at 12 (April 8, 2003). 46

53 The Company s 1.62% proposed cost of short-term debt is more than double its actual cost of short-term debt for calendar-year 2014 (0.65%). It is more than double the cost of short-term-debt set in its last rate case (0.36%), and more than double the six-month LIBOR rate as of date the Department filed its direct testimony. That date was already nearly two months into the test year, suggesting that the Company s proposed rate would be excessive for some eight months of the test year. The Company based its proposal on projected increases in interest rates, particularly in the LIBOR, since its short-term debt cost under its revolving credit agreement with its parent company puts its cost of short-term debt at LIBOR plus points. The Company did not, however, provide evidence of the reasonableness of that agreement s spread over LIBOR, which in itself exceeds the Company s actual cost of short-term debt in Nor did the Company provide evidence of the average cost of short-term debt for A-rated utilities, the standard it must meet under the ratepayer-protection order. These evidentiary gaps raise significant doubts about the reasonableness of the proposed 1.62% rate, and under Minn. Stat. 216B.03, all doubts as to reasonableness must be resolved in favor of the ratepayer. The Commission will therefore set the short-term-debt rate at the same level as year end 2014, 0.65%, the rate supported by the most recent and most reliable evidence in the record. XXIII. Final Capital Structure and Overall Cost of Capital The final capital structure and overall cost of capital resulting from the decisions made in this order are set forth below, rounded to the second decimal place: Component Ratio Cost Weighted Cost Long-Term Debt 42.30% 5.38% 2.28% Short-Term Debt 7.70% 0.65% 0.05% Common Equity 50.00% 9.49% 4.75% Total % 7.07% 47

54 XXIV. Class-Cost-of-Service Study A. Background CLASS COST OF SERVICE STUDY ISSUES As required by rule, the Company s rate-case filing included a class-cost-of-service study. 50 The purpose of a class-cost-of-service study is to determine, as accurately as possible, the costs of serving each customer class. While these costs cannot be determined with precision, it is critical that the cost study make both its underlying assumptions and the cost figures they yield as accurate and transparent as possible, because the Commission puts substantial weight on cost causation in determining what portion of the total revenue requirement each customer class should pay. The OAG and the Department challenged five aspects of the Company s cost study: (1) its classification of distribution-system costs based solely on a minimum-size study, (2) its interclass allocation of sales expenses by number of customer locations, (3) its interclass allocation of regulatory-commission expenses by number of customer locations, (4) its failure to update the cost of its distribution mains to present dollars, and (5) its classification of income taxes as a customer cost. Each challenge is addressed below. B. Distribution System 1. Summary An important purpose of a class-cost-of-service study is to classify the costs of the Company s distribution system consisting primarily of its gas mains as customer costs, commodity costs, or capacity costs. Customer costs are the costs of connecting all customers to the distribution system. Commodity costs are costs that vary directly with the amount of gas a customer consumes. And capacity costs are costs that vary based on how much a class contributes to the system s peak demand. The classification of costs has important rate consequences because each type of cost is allocated to the customer classes in a different way. For example, customer costs are generally allocated to each class residential, small business, large industrial, and so on based on the number of customers in that class. Since 91.7% of CenterPoint s customers are in the residential class, that class pays approximately 92% of customer costs. Because the distribution system is used jointly by all customer classes, its component costs cannot be classified with precision. Instead, these component costs are estimated using a classification method a system for classifying costs based on economic and engineering theory. In classifying distribution costs, the Commission considers the classification methods in the record and determines which method or methods are the most reasonable. 50 Minn. R (C). 48

55 The parties in this case have presented cost studies based on four classification methods: the minimum system method, the alternative minimum system method, the basic customer method, and the peak and average method. Each method is based on a different theory of how distribution-system costs are caused and results in a different mix of customer, commodity, and capacity-related distribution costs. In the sections that follow, the Commission reviews each method and the parties arguments for and against using it to classify distribution-system costs in this case. Ultimately, the Commission finds merit in each method; it will therefore consider the results of all the methods in apportioning CenterPoint s distribution-system costs among the customer classes. 2. The Classification Methods a. Minimum System CenterPoint recommended using the minimum-system method to classify the costs of its distribution system. The minimum-system method or theory holds that costs related to the size of the distribution system are caused by the need to meet peak demand and should be classified as a capacity cost. Some of the costs, however, are related purely to connecting customers to the system. These costs would still exist even in the case of a minimum system, a hypothetical system with little or no capacity to deliver natural gas. They are classified as customer costs and divided among the customer classes according to the number of members in each class. The two most common methods of calculating the cost of a minimum system are a zero-intercept study and a minimum-size study. A zero-intercept study uses regression analysis to estimate the cost of a hypothetical distribution system with zero-inch mains. A minimum-size study calculates the cost of a system with equipment of minimum practical size. Minimum practical size means the smallest size that fairly represents what is actually installed within a utility s distribution system. In this case, CenterPoint proposed to use a minimum system composed of two-inch mains. Two-inch mains make up 52.98% of the Company s distribution system in terms of linear footage and 25.71% in terms of cost. CenterPoint asserted that a system constructed of two-inch pipe would have essentially no capacity. The Company s minimum-size study resulted in a residential-class revenue deficiency of $61 million out of a total revenue deficiency of $54 million The residential class s deficiency was greater than the total deficiency because, under the Company s study, certain other classes were deemed to be paying more than their cost of service at current rates. 49

56 Minimum-Size Study Results 52 Customer Class Deficiency (Surplus) Residential $60,936,228 C&I A $6,156,106 C&I B $2,629,033 C&I C $(6,220,649) Large Firm Sales + Transport $(2,497,220) Sm Dual Fuel A $(4,779,181) Sm Dual Fuel B $(2,954,391) Lg Dual Fuel $1,023,866 Sm Dual Fuel A Transport $(1,142,610) Sm Dual Fuel B Transport $(1,082,332) Lg Dual Fuel Transport $2,036,488 The Department noted that the Commission had approved CenterPoint s use of a two-inch minimum-size study in the Company s last rate case. 53 It supported the Company s application of the minimum-system method with an adjustment to the historical mains-cost data used. 54 The OAG argued that the minimum-system method is grounded on the questionable premise that any part of the distribution system is customer-related. It argued that the method depends on several assumptions such as the minimum practical size of distribution equipment that can be manipulated in a way that will help a utility maximize its profits. And it argued that the Commission should hesitate to classify costs based on a single theoretical method where that method will have a substantial impact on the residential class. The OAG acknowledged that the Commission had approved CenterPoint s use of the minimum-system method in previous cases, but it argued that the Commission should consider other methods, discussed below, that were not developed in past cases. The Suburban Rate Authority agreed that CenterPoint s minimum-system study should not be the sole basis for the revenue apportionment in this case for the reasons given by the OAG. The SRA also pointed out that CenterPoint had failed to produce the residential subdivision model on which it relied in selecting a two-inch main size for its minimum-system study. 52 As required by the Commission s order in its last rate case, CenterPoint also filed minimum-system studies based on zero-inch and one-inch mains. See In the Matter of an Application by CenterPoint Energy Resources Corp. d/b/a CenterPoint Energy Minnesota Gas For Authority to Increase Natural Gas Rates in Minnesota, Docket No. G-008/GR , Findings of Fact, Conclusions, and Order, at 37 (June 9, 2014). However, no party supported using zero-inch or one-inch mains in this case. 53 See id. 54 The Department s recommended adjustment is discussed in section E, below. 50

57 b. Alternative Minimum System If the Commission uses the minimum-system method to classify distribution costs, the OAG recommended an alternative method of calculating the cost of the minimum system. The OAG referred to its method as the alternative minimum system method. The OAG selected 1.25-inch mains as best representing the minimum practical size of distribution equipment. The OAG reasoned that mains of 1.25 inches and smaller make up 10% of CenterPoint s distribution system, and that since 2-inch mains make up more than 50% of the utility s system, they cannot be considered to have no capacity. The OAG took the average installed unit cost of 1.25-inch mains, subtracted the materials cost of those same mains, and multiplied the result by the total length of all the mains in CenterPoint s distribution system. It reasoned that removing the materials cost of the mains would approximate the cost of a hypothetical system with zero-inch mains, arriving at the same result as the zero-intercept method by different means. 55 The OAG s alternative-minimum-system method would reduce the costs assigned to the residential class by approximately $15 million. CenterPoint argued that the OAG s alternative-minimum-system method is not a method recognized by this Commission, other state utility commissions, or the utility industry generally. And it argued that the OAG s calculations understate the customer component of the distribution system because they use cost data from the period of to represent the cost of mains that were installed over the system s 109-year history. The Department did not support the OAG s alternative-minimum-system method, arguing that the method falls short of recognizing the costs of materials that are needed for a customer to be able to receive service from CenterPoint. And the Department noted that the Commission had rejected the OAG s method in another utility s rate case. 56 c. Basic Customer The OAG also advanced the basic customer method. The OAG stated that this method is included in the Gas Distribution Rate Design Manual authored by the National Association of Regulatory Utility Commissioners ( NARUC Gas Manual ) and is used by a number of state utility commissions. 55 CenterPoint filed a zero-intercept study in this case. However, the OAG argued that its analysis was so flawed that it was unusable for allocating costs. 56 See In the Matter of the Application of Dakota Electric Association for Authority to Increase Rates for Electric Service in Minnesota, Docket No. E-111/GR , Findings of Fact, Conclusions, and Order, at (June 8, 2015). In the Dakota case, the OAG referred to its method as the zero intercept proxy method. 51

58 The basic-customer method or theory reasons that the distribution system is designed and constructed to meet peak demand, not to serve any particular customer. Thus, only costs that can be traced back to a specific customer such as the costs of service lines, meters, billing, and collection are classified as customer-related. Other costs, including the cost of the distribution system, are classified as capacity-related. Using the basic-customer method would reduce the costs assigned to the residential class by approximately $31 million. CenterPoint criticized the basic-customer method as being driven by economic theory rather than engineering principles. The Company stated that it had surveyed other jurisdictions and found that regulators rely far less on the basic-customer approach than suggested by the OAG and that the number of states using the minimum-system approach has grown significantly in recent years. The Department also did not support the basic-customer approach advocated by the OAG, concluding that the OAG had not provided a reasonable basis for departing from the usual practice of using the minimum-system method. d. Peak and Average Finally, the OAG advanced the peak-and-average method. The OAG stated that this method is described in NARUC s Electric Manual and is used by a number of state utility commissions. The peak-and-average theory reasons that some part of the distribution system is built to supply a base level of gas at all times represented by the system s load factor. This part of the system is classified as a commodity cost because it is related to how much gas is sold. The rest of the costs of the system are deemed to be related to the need to serve peak demand and are classified as capacity-related. Using the peak-and-average method would reduce the costs assigned to the residential class by approximately $63 million. CenterPoint both criticized the OAG s application of the peak-and-average method and questioned whether the method was as widely accepted as the OAG had represented. And the Department, as with the basic-customer method, asserted that the OAG had not provided a reasonable basis for departing from the usual practice of using the minimum-system method. 3. The Recommendation of the Administrative Law Judge The Administrative Law Judge recommended that the Commission use the minimum-size method to classify the costs of CenterPoint s distribution system. He found that the record supported the Company s conclusion that two-inch mains represent the smallest practical size of the distribution system. And, relying in part on past Commission practice, he did not recommend using the alternative-minimum-system, basic-customer, or peak-and-average methods. 52

59 4. Commission Action The Commission does not concur with the Administrative Law Judge that the strengths of the minimum-system method are so superior to those of the other three analytical models developed in the record that they justify relying exclusively on a minimum-system analysis to classify and allocate distribution-system costs. While the Company s minimum-system cost studies are similar to studies approved by the Commission in the past, this record is uniquely comprehensive in its examination of alternative approaches. It therefore offers an opportunity to delve more deeply into these complex issues and potentially refine the Commission s approach. The Commission finds that the class-cost-of-service studies presented by the parties in this case are a useful guide to revenue apportionment and rate design and will consider all the classification methods in making a revenue-apportionment decision. The OAG showed that there are several methods, including the minimum-system, basiccustomer, and peak-and-average methods, for classifying and allocating distribution-system costs. The Commission finds merit in each theory. The results of the class-cost-of-service studies vary significantly depending on the method used. Out of the total Company deficiency of $54 million, the two-inch minimum-system study sponsored by the Company assigns a revenue deficiency of roughly $61 million to the residential class, whereas the peak-and-average method advanced by the OAG shows a revenue sufficiency of roughly $2 million attributable to this class. The other residential-class revenue-deficiency estimates fall within this range. The basic-customer approach shows a residential-class revenue deficiency of roughly $30 million, and the Department s adjustments to the Company s minimum-system model result in a residential-class revenue deficiency of $52 million. The OAG s adjustments to the Company s two-inch minimum-system model show a revenue deficiency attributable to the residential class of roughly $46 million. While the Company s minimum-system cost studies are in line with previous cost studies approved by the Commission, the Commission is still in the process of exploring many complex issues regarding the allocation of distribution costs. The Commission is persuaded, on valid theoretical grounds, that the minimum-system studies over-allocate distribution costs to the customer component. Other class-cost-of-service studies suggest that the over-allocation to the customer component may be significant. The results from the minimum-system studies could not be compared with, or validated against, the alternative zero-intercept study filed by the Company in this case. However, the other class-cost-of-service studies in the record in this case serve to suggest a range of plausible cost allocations to various customer classes. As the Commission continues to explore class-cost-of-service issues in the Company s next rate case, the Commission will require the Company to file a minimum-system study based on a statistically significant zero-intercept study, in addition to the two-inch pipe study it has traditionally used. 53

60 The Company should also file a basic-customer cost study and a peak-and-average cost study. These studies will yield a range of guidance upon which an apportionment may be established. Although the basic-customer and peak-and-average methods are not commonly used in Minnesota, of the 43 states that use an embedded cost study, as Minnesota does, 20 states use one of these two methods. The Commission will also require CenterPoint to provide a substantive explanation and justification of its classification and allocation methods when it files its class-cost-of-service studies in its next rate case. The Commission declines to adopt several ALJ findings that either are not essential to, or conflict with, its findings and conclusions above. The specific ALJ findings not adopted are listed in the ordering paragraphs at the end of this order. The Commission will also adopt certain supplemental findings advocated by the OAG in its exceptions that are supported by the record and that support the Commission s decision. These supplemental findings are also listed in the ordering paragraphs. The Commission will adopt the remainder of the ALJ s findings to the extent that they are consistent with the foregoing discussion. C. Sales Expenses 1. The Issue CenterPoint s class-cost-of-service study allocates sales expenses by number of customer locations. The Company argued that sales are made to all customers and, therefore, the expenses associated with the sales are properly allocated by customer location as a proxy for direct assignment. The Company also argued that this treatment would allow it to maintain consistent practice across the states in which it operates. The Department argued that CenterPoint s proposal to allocate sales expenses based on the number of customer locations contradicted the Commission s decision in the Company s previous rate case. There, the Commission ordered CenterPoint to allocate its sales expenses based on each class s share of the overall revenue requirement. The Commission was persuaded by the NARUC Electric Manual s recommendation that sales expenses, as a general expense, should be allocated on the basis of actual data on cost causation, if available, and otherwise on the basis of the overall revenue responsibility of each class. 2. The Recommendation of the Administrative Law Judge The Administrative Law Judge concluded that (a) CenterPoint had not demonstrated that it targets its sales activity at every customer in equal measure, (b) the guidance of the NARUC Electric Manual as to appropriate allocation of these expenses was persuasive, notwithstanding the differences between delivering natural gas and electricity, and (c) it was more reasonable to allocate sales costs based on actual data or, failing that, overall revenue responsibility than it was to apportion the costs based on the number of customer locations. 54

61 3. Commission Action The Commission concurs with the Administrative Law Judge and adopts his findings, conclusions, and recommendation. As in its last rate case, CenterPoint did not demonstrate that it targets its sales activities at every customer equally, and, in the absence of actual data on cost causation, these general expenses should be allocated generally. D. Regulatory-Commission Expenses 1. The Issue CenterPoint has one account devoted to regulatory-commission expenses, FERC Account 928. The Company proposed to allocate regulatory-commission expenses by the number of customer locations in each class. It argued that this treatment would be consistent with its practice in other jurisdictions. The Department argued that CenterPoint s proposal to allocate regulatory-commission expenses by customer location contradicted the Commission s decision in the Company s last rate case. There, the Commission decided that these costs should be allocated using a general allocator, Total Production and Distribution O&M Expense less Gas Cost. The Department argued that regulatory-commission costs are general, not directly customer related, and therefore should be assigned using a general allocator. 2. The Recommendation of the Administrative Law Judge The Administrative Law Judge concluded that regulatory-commission expenses should be allocated based upon Total Production and Distribution O&M Expense less Gas Costs. 3. Commission Action The Commission concurs with the Administrative Law Judge and adopts his findings, conclusions, and recommendation. Regulatory-commission costs are general, are not directly customer related, are not linked to the number of customers in a class, and should be allocated as he recommends. E. Treatment of Historical Cost Data on Distribution Mains 1. The Issue While the Department agreed with CenterPoint s use of a two-inch minimum system to classify its distribution system, the Department recommended repricing the Company s distribution mains at their relevant Handy Whitman indexed costs, rather than at their original costs. The average unit cost of mains is used in both a minimum-size study and a zero-intercept study to determine the proportion of the distribution-system costs that should be classified as customer related and capacity related. According to the Department, applying a price index to the original cost of the mains would normalize the investment data and ensure consistent valuation. 55

62 CenterPoint agreed that normalizing historical cost data was appropriate for a zero-intercept study because the study regresses the cost data. However, the Company argued that it was not necessary to manipulate the cost data in a minimum-size study. It asserted that neither this Commission nor any commission in the other states where it operates require that cost data be normalized in a minimum-size study. 2. The Recommendation of the Administrative Law Judge The Administrative Law Judge was persuaded by the Department s argument that normalization was required for an apples to apples cost comparison, finding limited support for CenterPoint s claim that no other jurisdiction in which the Company operates requires normalized cost data for a minimum-size study. He therefore recommended that all distribution mains be priced at their relevant Handy Whitman adjusted cost. 3. Commission Action The Commission concurs with the Administrative Law Judge and adopts his findings, conclusions, and recommendation. Given that the Company s mains have been installed at various times over a more-than-100-year period, logic suggests that, for both a zero-intercept study and a minimum-size study, the original cost data should be adjusted to current dollars so that the system can be valued accurately. F. Income Taxes 1. The Issue CenterPoint s class-cost-of-service study classifies all income taxes as customer costs. The Department and the OAG objected to classifying income tax as a customer cost, since income taxes are not related to the number of customers CenterPoint has. They argued that income taxes should be classified as customer, commodity, and capacity costs according to the percentage of rate-base costs in each of those categories. CenterPoint argued that its proposal was consistent with the Commission s order in its previous rate case, which did not require any particular classification of income taxes, but required the Company to allocate income taxes based on the percentage of rate-base costs attributable to each customer class The Recommendation of the Administrative Law Judge The Administrative Law Judge concluded that the Commission s previous order implied the proper way to classify income taxes: in proportion to the customer, capacity, and commodity components of rate base. 57 See Docket No. G-008/GR , Findings of Fact, Conclusions, and Order, at (June 9, 2014). 56

63 3. Commission Action The Commission concurs with the Administrative Law Judge and adopts his findings, conclusions, and recommendation. CenterPoint s income taxes are caused by the investments it makes to provide natural gas service, represented by its rate base. Accordingly, in the Company s last rate case, the Commission directed the Company to allocate income taxes according to the percentage of rate-base costs attributable to each customer class. The proportion of rate base attributable to each class cannot be determined without first classifying each rate-base component as a customer, commodity, or capacity cost. The Commission therefore agrees with the ALJ that allocating income taxes to the customer classes according to each class s share of rate base requires that income tax first be divided into customer, commodity, and capacity components in proportion to the percentage of rate-base costs in each of those categories. XXV. Class Revenue Apportionment A. Introduction RATE DESIGN ISSUES In every rate case the new revenue requirement must be apportioned among the customer classes, raising the issue of whether to adjust the interclass revenue responsibility built into the existing rate structure. In this case, CenterPoint proposed an apportionment that would move each customer class closer to paying its cost of service as estimated by the Company s class-cost-of-service study. To accomplish this result, the Company proposed to Increase rates for the Commercial/Industrial (C/I) C, Small Volume Dual Fuel (SVDF) A and SVDF B classes only by the change in their Conservation Cost Recovery Charge (CCRC) rate; Increase rates for the C/I B class and the Large Volume (LV) customers to eliminate those classes revenue deficiencies; Increase rates for the Commercial A class to eliminate half of its revenue deficiency; and Apportion the remaining deficiency to the Residential class. CenterPoint s recommended revenue apportionment would result in the following rate increases for each customer class under its proposed revenue requirement: 57

64 Customer Class B. Positions of the Parties Rate Increase Residential 9.2% Commercial A 17.2% C/I B 7.0% C/I C 0.1% SVDF A 0.2% SVDF B 0.3% Large Volume 1.4% Total System 6.4% 1. The Department The Department recommended setting the Residential rate increase at 7.5%, setting the Commercial A increase at 10%, and spreading the remaining revenue requirement proportionally among the other classes. The Department argued that this apportionment would appropriately balance the rate-design goals of reducing interclass subsidies and minimizing the potential for rate shock. 2. The OAG The OAG developed its revenue apportionment after considering the results of multiple class-cost-of-service studies including the Company s minimum-system study, its own alternative-minimum-system study, the basic-customer method, and the peak-and-average method. The OAG s methodology can be summarized as follows: Classes that had a revenue deficiency in each of the studies Commercial A, C&I B, Large Volume Dual Fuel (LVDF) Sales, and LVDF Transport received the largest proportional increases in their revenue apportionment. The SVDF B Sales class had a revenue sufficiency under every method, so the OAG recommended minimizing the revenue-apportionment increase for that class. Classes that had a revenue deficiency under some methods and a revenue sufficiency under others received rate increases within two percentage points of the 6.4% total system increase the one exception being LV Firm Sales and Transport. Finally, the OAG reviewed three classes assigned relatively large rate increases Commercial A, LV Firm Sales and Transport, and LVDF Transport to ensure that there was a good reason for the increase. The OAG reasoned that LV Firm Sales and Transport s rate increase was justified because it was based on assigning the class the same percentage of revenues as in CenterPoint s last rate case, while the increase to the other two classes was justified because they had large revenue deficiencies under every cost study. The OAG s recommended revenue apportionment would result in the following rate increases for each customer class under CenterPoint s proposed revenue requirement: 58

65 Customer Class 58 Rate Increase Residential 6.2% Commercial A 11.6% C/I B 8.13% C/I C 4.93% SVDF A Sales 4.31% SVDF A Transport 6.44% SVDF B Sales 3.55% SVDF B Transport 8.44% LV Firm Sales and Transport 16.26% LVDF Sales 9.0% LVDF Transport 17.78% 3. The Company CenterPoint maintained that its apportionment recommendation would reasonably move class revenue responsibilities closer to cost while maintaining a substantial subsidy benefitting the Residential class. The Company argued that the Department s recommendation would unnecessarily subsidize the Residential class and was so close to the Company s recommendation that it would do little to prevent alleged rate shock. And it argued that the OAG s proposal presented a significant break from past practice and was based on flawed cost-of-service studies. C. The Recommendation of the Administrative Law Judge The Administrative Law Judge found that the cost-of-service studies submitted by CenterPoint and the Department were the most faithful to the Commission s past practice and that CenterPoint s revenue-apportionment recommendation reasonably moved class revenue responsibilities closer to cost while minimizing rate shock to the Residential and Commercial A classes. The ALJ found that the record did not support the reasonableness of the OAG s apportionment recommendation, which relied on alternative cost studies, and that the OAG s recommendation would increase the interclass subsidy favoring the Residential class. For these reasons, the ALJ recommended that the Commission approve CenterPoint s proposed revenue allocation. 58 While CenterPoint s revenue-apportionment recommendation aggregated certain SVDF and large-volume classes, the OAG s recommendation broke out each class, resulting in a longer list of customer classes. 59

66 D. Commission Action The Commission disagrees with several of the ALJ s findings, as well as his recommendation to approve CenterPoint s proposed revenue allocation. The Commission will instead adopt the OAG s recommended class revenue apportionment based on multiple cost studies, as shown in the following table: Customer Class OAG s Recommended Revenue Apportionment 59 Revenue Apportionment Current Revenue Apportionment Percent Change Residential 64.90% 65.60% -1.07% C&I A 2.70% 2.51% 7.65% C&I B 4.10% 3.85% 6.48% C&I C 15.05% 15.02% 0.19% Large Firm Sales + Transport 2.92% 2.92% 0.02% Sm Dual Fuel A 2.89% 2.89% 0.11% Sm Dual Fuel B 1.50% 1.52% -1.34% Lg Dual Fuel 1.12% 0.84% 33.17% Sm Dual Fuel - A Transport Sm Dual Fuel - B Transport Lg Dual Fuel Transport 0.52% 0.57% -8.35% 0.50% 0.53% -6.30% 3.80% 3.75% 1.30% The Commission has already found that it is appropriate, in this case, to consider the results of multiple class-cost-of-service studies in making a revenue-apportionment decision. Because most costs are caused jointly by all customer classes, cost studies can only estimate what portion is attributable to each class based on economic and engineering theory. And where, as here, competing theories produce widely varying cost allocations, it is most reasonable to consider the results of all the studies to avoid unduly burdening one or more classes. The OAG has presented a clear, transparent, and rational methodology for reaching a revenue apportionment based on multiple class-cost-of-service studies. The OAG derived its apportionment recommendation by identifying patterns in the class revenue deficiencies or sufficiencies across the several cost studies in the record and recommending adjustments based on those patterns. 59 Nelson Direct Testimony, at

67 Finally, an additional benefit of the OAG s apportionment is that it hews more closely to CenterPoint s currently approved revenue apportionment than does the Company s recommended apportionment. This results in a gradual change in class apportionments, which will better promote stability of rates and minimize the chance of rate shock. For all these reasons, the Commission will adopt the OAG s recommended revenue apportionment; the Commission declines to adopt the ALJ s findings on this issue and will instead adopt proposed findings from the OAG s exceptions to the ALJ s Report. XXVI. Customer Charges A. Introduction The customer charge is a fixed monthly charge assessed without regard to usage levels. It is designed to help recover fixed customer-related costs such as the cost of meters, service lines, meter reading, and billing. CenterPoint proposed to move the customer charges for certain classes closer to the customer cost estimated in its class-cost-of-service study. The Company proposed the following customer-charge increases: Customer Class Current Charge Proposed Charge Increase Residential $9.50 $11.75 $2.25 Commercial A $15.00 $17.25 $2.25 C/I B $21.00 $26.25 $5.25 LV Sales $ $1, $ LV Transport $ $1, $ B. Positions of the Parties 1. The Company CenterPoint argued that increasing these customer charges would move them closer to cost, reduce intraclass subsidies, and send more accurate price signals to customers. The Company also argued that the increase would stabilize customer bills and require smaller rate adjustments under its revenue-decoupling mechanism. Revenue decoupling is a regulatory tool designed to separate a utility s revenue from changes in energy sales. CenterPoint has full revenue decoupling, meaning that when its sales depart from forecasts for any reason, its rates will be adjusted to ensure that the Company eventually recovers the revenues it would have earned at the forecasted level of sales. According to CenterPoint, the adjustments needed would be smaller with a larger fixed charge since it would result in less revenue being collected through the per-therm, or volumetric, charge. 61

68 Finally, CenterPoint asserted that the Commission had consistently approved increases to the customer charges to move them closer to the fixed cost of service. And it argued that its proposed increase to the Residential charge was generally consistent with trends across the country and in the region, based on a survey published by the American Gas Association (AGA) in May The Department The Department generally agreed that the Residential and Commercial A customer charges should be brought closer to cost as estimated in the class-cost-of-service study. However, the Department recommended a smaller increase of $1 per month, raising the Residential charge to $10.50 and the Commercial A charge to $ It argued that a $1 increase would appropriately balance the competing rate-design goals of encouraging conservation and minimizing intraclass subsidies. Intraclass subsidies arise when certain customers pay more than the cost to serve them, subsidizing other customers within the same class who pay less than their cost of service. For example, when the customer charge for a given class is not set high enough to recover the class s fixed cost of service, the volumetric charge must be increased to capture fixed costs not covered by the customer charge. In this situation, high-usage customers will end up subsidizing lower-usage customers, paying not only their own fixed costs and usage costs but also part of lower-usage customers fixed costs. Finally, the Department argued that a rate design with more recovery through the customer charge and less recovery through the volumetric charge would benefit low-income customers in the high-usage category. And it argued that its proposal was the most balanced because it would result in a similar percentage increase in bills for Residential ratepayers using various amounts of energy. 3. The OAG The OAG recommended reducing the Residential customer charge to $8.00 and the Commercial A customer charge to $ The OAG argued that CenterPoint had overestimated the fixed costs that should be included in the customer charge. According to the OAG, CenterPoint s recommendation assumes that the customer charge should reflect every customer-related cost from the class-cost-of-service study. However, the OAG maintained that customer costs estimated in a class-cost-of-service study are not equivalent to the fixed costs that should be reflected in a customer charge. According to the OAG, economic theory and academic literature establish that the customer charge should reflect only the marginal cost of adding one more customer to the system i.e., the costs of a service line, a regulator, a meter, meter reading, and account administration. It calculated that the monthly cost of adding a Residential customer to the system was only $5.33 and that the monthly cost of adding a Commercial A customer was $ The OAG thus argued that moving the Residential and Commercial A customer charges closer to cost would actually require decreasing them from their current levels. 62

69 Finally, the OAG argued that one of primary purposes of a fixed customer charge promoting a utility s revenue stability is obviated where, as here, the utility has implemented full decoupling. And it argued that increasing the customer charges would reduce customers incentive to conserve energy, contrary to the legislative mandate to set rates that encourage energy conservation to the maximum reasonable extent Fresh Energy Fresh Energy recommended maintaining the Residential customer charge at its current level. Fresh Energy argued that the Company had not met its burden to establish that increasing the customer charge was just and reasonable. It agreed with the OAG that a customer charge should reflect only the marginal cost of adding a customer to the system; that increasing the customer charge would reduce customers incentive to conserve; and that increasing the charge was unnecessary to ensure revenue stability in light of CenterPoint s implementation of full decoupling. Fresh Energy also argued that CenterPoint and the Department had focused on one kind of intraclass subsidy higher-usage customers subsidizing lower-usage customers without acknowledging any other intraclass subsidies that exist or considering how an increase in the customer charge would affect existing subsidies, or whether it might create new subsidies. 5. SRA Like the OAG and Fresh Energy, the SRA opposed increasing the Residential customer charge. The SRA acknowledged that the Commission had approved previous proposals to increase CenterPoint s Residential customer charge but noted that in other recent rate cases, the Commission had given increased scrutiny to such proposals. For example, in Xcel Energy s last rate case, the Commission declined to raise residential and small-general-service customer charges, finding that doing so would give too much weight to the customer cost calculated in Xcel s class-cost-of-service study and not enough weight to affordability and energy conservation. 61 The SRA questioned the Department s assertion that increasing the customer charge would not discourage conservation, arguing that it was reasonable to presume that the greater the fixed portion of a customer s bill, the lower the customer s incentive to invest in conservation. It maintained that neither CenterPoint nor the Department had introduced any evidence such as customer surveys rebutting that presumption. Absent such evidence, the SRA argued that the Commission should err on the side of encouraging conservation. 60 Minn. Stat. 216B In the Matter of the Application of Northern States Power Company for Authority to Increase Rates for Electric Service in the State of Minnesota, Docket No. E-002/GR , Findings of Fact, Conclusions, and Order, at 88 (May 8, 2015). 63

70 C. The Recommendation of the Administrative Law Judge The Administrative Law Judge recommended that the Commission approve CenterPoint s proposed customer-charge increases. He found that increasing the charges would move them closer to cost, reduce intraclass subsidies, be consistent with past practice, and place CenterPoint s rates among its regional peers, without discouraging energy conservation. D. Commission Action The Commission disagrees with the Administrative Law Judge s conclusion that CenterPoint s customer-charge proposal best balances the competing rate-design goals in this case. For the reasons discussed below, the Commission will require the Company to maintain the current customer charges for all its customer classes. The Commission adopts the ALJ s other findings to the extent that they are consistent with its decision. Every rate made by a public utility must be just and reasonable; must not be unreasonably preferential, unreasonably prejudicial, or discriminatory. 62 And, the Commission must set rates to encourage conservation and renewable energy use to the maximum reasonable extent. 63 Any doubt as to the reasonableness of a utility s rates must be resolved in favor of the consumer. 64 The Commission concludes that increasing CenterPoint s customer charges would place too little emphasis on the need to set rates to encourage conservation. This is particularly true since the Company has a full-decoupling mechanism. 65 One of the benefits of customer charges in the absence of decoupling is that they stabilize the utility s revenue, guaranteeing a minimum amount of recovery each month regardless of customers usage. However, decoupling already guarantees that CenterPoint will not fail to recover its revenue requirement due to lower-than-predicted sales. Further, a major goal of revenue decoupling is to align a utility s interests with the public s interest in energy efficiency. Increasing the customer charge undermines this goal by incrementally reducing customers incentive to conserve energy since, with a higher customer charge and relatively lower volumetric charge, they are less able to control the size of their bills by using less energy. Keeping CenterPoint s customer charges at current levels will maintain the existing incentive to conserve without affecting the Company s revenue stability. Keeping the current charges is also reasonable in light of evidence adduced by the OAG that casts doubt on the usefulness of the class-cost-of-service study for determining the appropriate customer-charge amount. The OAG argued that the customer component in the Company s cost study was not an accurate measure of the cost of adding a customer to the system. 62 Minn. Stat. 216B Id. 64 Id. 65 Importantly, while CenterPoint argues that increasing its customer charges is consistent with the practice of utilities surveyed by the AGA, the Company has not shown that these utilities used revenue decoupling and therefore that they were similarly situated to CenterPoint. 64

71 The Company and the Department both argued that increasing the customer charges would reduce intraclass subsidies. However, this conclusion is based on the premise that the charges are currently set below cost a premise on which the OAG has cast significant doubt. Moreover, low-usage/high-usage subsidies are not the only subsidies that can occur within a class. For example, customers with more expensive connections based on geographical location, length of service line, etc. will always be subsidized by customers with less expensive connections. Some level of intraclass subsidy is unavoidable; the task is to balance the potential for and magnitude of a subsidy against other rate-design considerations. Here, the Commission finds that that balance favors the existing customer-charge levels. For all these reasons, the Commission will require CenterPoint to maintain the existing customer charges for all classes. XXVII. Future Rate Case Requirements A. Calendar-Year Test Year 1. Introduction In a rate case, the Commission uses a utility s reasonable costs from a historical or projected test year to calculate the utility s revenue requirement. The Commission s rules define a test year as the 12-month period selected by the utility for the purpose of expressing its need for a change in rates. 66 In this case as in its other recent rate cases, CenterPoint proposed a 12-month test year ending on September 30 a split test year rather than a calendar-year test year. However, the Company s witnesses did not consistently use a split test year in presenting financial information. This made reviewing the Company s financial data more difficult for other parties, who had to make extra efforts to ensure that the data were comparable. 2. Positions of the Parties The OAG and the Department recommended that the Commission require CenterPoint to file, in its next rate case, a test year that coincides with the calendar year. Because Minn. R , supb. 17, permits a utility to choose its test year, these parties recommended that the Commission vary the rule under either Minn. Stat (Department) 67 or Minn. R (OAG). 68 Alternatively, the OAG recommended that the Commission require CenterPoint to file all its financial information according to the test year selected in its next rate case. 66 Minn. R , subp Minn. Stat allows an agency to vary a rule if it finds that (1) application of the rule to the petitioner would result in hardship or injustice, (2) a variance would be consistent with the public interest, and (3) a variance would not prejudice the substantial legal or economic rights of any person or entity. 68 Minn. R allows the Commission to vary a rule if (1) enforcement of the rule would impose an excessive burden on the applicant or others affected by the rule, (2) granting the variance would not adversely affect the public interest, and (3) granting the variance would not conflict with standards imposed by law. 65

72 CenterPoint opposed a requirement to file a calendar-year test year, arguing that it would prevent the Company from having interim rates in effect by the beginning of the heating season, thereby delaying recovery of its reasonable and necessary expenses and capital investments. CenterPoint also argued that it was the only party with standing to request a variance from the Commission s rules. 3. The Recommendation of the Administrative Law Judge The Administrative Law Judge concluded that Minn. Stat does not allow a party other than a utility to request a variance from Minn. R , subp. 17. He recommended that the Commission direct the parties to use the period between rate cases to search for a middle ground, suggesting that they discuss whether there may be methods of transitioning CenterPoint to using a calendar-year test year in a way that will not threaten the Company s recovery of its reasonable expenses and capital investments. 4. Commission Action The Commission finds that Minn. R , rather than Minn. Stat , governs the Commission s variance determinations. 69 However, the Commission concludes that a variance is not the most prudent course of action at this time. Instead, the Commission will require CenterPoint, in future rate cases, to present all testimony and supporting schedules consistently, using the same 12-month period as its selected test year. The Company should provide transparent and reconcilable schedules if its selected test year differs from the financial, operating, and budgeting reporting periods used in practice. If the Company fails to comply, the Commission may reject the rate-case filing as incomplete. CenterPoint should also continue to comply with any existing rules regarding the format of information submitted as part of the Company s rate case. Requiring CenterPoint to present information consistent with its test year will significantly ease the burden of the Department and the OAG in reviewing the Company s filings, without impairing the Company s ability to choose its test year. B. Sales Forecast and Weather Normalization 1. The Issue An accurate sales forecast for the test year is critical to calculating a utility s revenue requirement. Because weather has a substantial effect on sales, a sales forecast is normalized to reflect normal weather conditions using historical weather data. In CenterPoint s last rate case, the Commission approved the Company s use of a 10-year weather-normalization period. 70 However, the Commission stated that its approval did not reflect unequivocal support for the practice and required CenterPoint to provide, in its next rate case, a comprehensive examination of the predictive power, volatility, and impact on the test year and future revenues of using 10, 15, and 20-year weather data in its sales forecast. 69 The Commission adopts ALJ finding 600 modified accordingly. The Commission also adopts ALJ finding 598 modified to reflect that the OAG urged the Commission to use the variance procedures in Minn. R Docket No. G-008/13-316, Findings of Fact, Conclusions, and Order, at 29 (June 9, 2014). 66

73 CenterPoint provided the required information in this case; however, it argued that it should not be required to provide a discussion of 20-year weather data in future rate cases. The Department compared forecasts based on 10-, 15-, and 20-year weather data and concluded that, while the 10-year data yielded better predictions in some circumstances, they were also more volatile than the forecasts based on 15- and 20-year data. The Department recommended that CenterPoint continue to examine the predictive power and volatility of using 10-year, 15-year and 20-year weather data in future rate cases. The Administrative Law Judge concluded that the parties would be best served in future rate cases by focusing their attention on whether a 10-year or 15-year normalization period should be used. 2. Commission Action The Commission disagrees with the Administrative Law Judge and will require CenterPoint to continue providing a comprehensive examination of the predictive power, volatility, and impact on test-year and future revenues of using 10-, 15-, and 20-year weather data. While the parties agree on the reasonableness of using 10-year weather-normalization data for the sales forecast in this rate case, it does not necessarily follow that 15- or 20-year data will never be appropriate. In particular, the Department points out that using 10-year data yields more volatile estimates. Given the uncertainty as to the best method for estimating normal weather, the Commission will maintain the existing requirement that the Company examine 10-, 15-, and 20-year weather data in future rate cases. 71 C. Interim-Rate-Design Methodology While a rate case is pending, a utility is permitted to charge interim rates based on its proposed test-year cost of capital, rate base, and expenses, but without any change to its existing rate design. 72 In this case, as in past rate cases, CenterPoint included gas sales in determining the interim-rate surcharge for each customer class. However, in response to an information request, the Company acknowledged that the cost of gas is not part of its existing rate design. The Commission will direct CenterPoint Energy, in future rate cases, to change its existing interim-rate methodology to preserve its existing rate design, by removing cost-of-gas billings when determining and applying the interim-rate increase. This will ensure that future interim rates are consistent with Minn. Stat. 216B.03, subd. 3(b), which requires that interim rates reflect a utility s existing rate design. D. Formatting Financial Information Finally, the Commission observes that some Company witnesses presented financial data in company wide format including costs from all states in which the utility operates while others presented data in Minnesota jurisdictional format including only costs allocated to its Minnesota ratepayers. 71 The Commission adopts ALJ finding 604, as well as findings modified to reflect the Department s position, as set forth in the ordering paragraphs. 72 Minn. Stat. 216B.16, subd. 3(b). 67

74 To facilitate review of CenterPoint s finances in future rate cases, the Commission will require the Company to present financial information (including witnesses testimony) using the following format: $1,000,000 total Company ($750,000 MN). A similar format will also be acceptable as long as the Minnesota-jurisdictional amount is clear. FINANCIAL SCHEDULES AND COMPLIANCE XXVIII. Overall Financial Schedules A. Gross Revenue Deficiency The above Commission findings and conclusions result in a Minnesota jurisdictional total gross revenue deficiency of $27,541,000, as shown below: Revenue Requirement Summary Test Year Ending September 30, 2016 ($000s) Average Net Rate Base $893,113 Rate of Return 7.07% Required Operating Income $63,143 Operating Income $46,996 Operating Income Deficiency $16,147 Gross Revenue Conversion Revenue Deficiency $27,541 B. Rate Base Summary Based on the above findings, the Commission concludes that the appropriate rate base for the test year is $893,113,000, as shown below: Rate Base Summary Test Year Ending September 30, 2016 ($000s) Utility Plant in Service: Intangible $927 Production $21,204 Underground Storage $22,759 68

75 Other Storage $16,965 Distribution $1,669,048 General $203,331 Total Utility Plant in Service $1,934,234 Accumulated Reserve: Intangible $366 Production $18,961 Underground Storage $20,758 Other Storage $17,624 Distribution $723,600 General $110,794 Total Accumulated Reserve $892,104 Net Utility Plant in Service: Intangible $561 Production $2,242 Underground Storage $2,001 Other Storage ($659) Distribution $945,448 General $92,537 Total Net Utility Plant in Service $1,042,130 Gas Stored Underground-Noncurrent $177 Customer Advances for Construction ($214) Accumulated Deferred Income Taxes ($186,604) Working Capital: Materials and Supplies $11,286 Gas Stored Underground-Current $30,913 Liquefied Natural Gas Stored $1,660 Liquefied Petroleum (Propane) Gas $5,919 Prepayments $1,259 Other Deferred Debits & Credits ($13,724) Other Cash Working Capital $311 Total Working Capital $37,624 Average Net Rate Base $893,113 C. Operating Income Summary Based on the above findings, the Commission concludes that the appropriate Minnesota jurisdictional operating income for the test year under present rates is $46,996,000, as shown below: 69

76 Statement of Operating Income Test Year Ending September 30, 2016 ($000s) Operating Revenue Sales of Gas Residential $484,683 Commercial & Industrial $207,893 Total Firm $692,576 Dual Fuel $66,133 Transportation $26,107 Other $1,877 Less: Franchise Fees $0 Total $786,693 Late Payment Charges $2,858 Other Operating Revenue $0 Total Operating Revenue $789,551 Operating Expenses Operation and Maintenance Cost of Gas Purchases $444,488 Production $1,312 Other Gas Supply $806 Underground Storage $896 Other Storage $727 Distribution $39,215 Customer Accounts $35,269 Customer Service & Informational $35,415 Sales $449 Administrative & General $39,359 Total Operation $597,936 Maintenance Expenses $21,480 Total Operation & Maintenance $619,416 Depreciation and Amortization $72,236 Federal & State Income Taxes $12,660 Deferred Income Taxes $5,941 Investment Tax Credit Adjustment $0 Other Taxes $32,302 Total Operating Expenses $742,555 Operating Income Before AFUDC $46,996 Allowance for Funds Used During Construction $0 Utility Operating Income $46,996 70

77 XXIX. Compliance Filing Required The Commission will require the Company to make a compliance filing within 30 days of the date of this order showing the final rate effects of the decisions made here and proposing a plan for refunding the difference between the amounts it collected in interim rates and the amounts it is authorized to collect in final rates. The Commission will establish a brief comment period to give interested persons a chance to review and comment on the filing, apart from the proposed customer notice. ORDER CenterPoint Energy Minnesota Gas is entitled to increase Minnesota-jurisdictional revenues by $27,541,000 to produce jurisdictional total gross revenue of $817,092,000 for the test year ending September 30, The Commission accepts, adopts, and incorporates the findings, conclusions, and recommendations of the Administrative Law Judge, except as set forth herein. Recovery of non-qualified pension costs is denied. Recovery of long-term incentive plan costs is denied. Corresponding test-year expenses are reduced by a total of $1,444,660 the sum of the Company s expenses ($241,945) and the Service Company and Board of Directors Expenses assigned to the Company ($1,204,715). Recovery of the expense items related to 2013 rate-case expenses is denied. Recovery of American Gas Association membership dues is denied. Recovery of the cost of the United Way donation is denied. Recovery of the Company s property-tax expense is approved, excluding the 2% referendum contingency. CenterPoint Energy s test-year short term incentive plan cost recovery shall be capped at 25% of employees base pay. In future rate cases, if the short term incentive goals do not place greater weight on customer-service components, the Commission may lower the cap or deny recovery of the short-term incentive plan costs. The Service Company s short-term incentive plan test year amount shall be adjusted to reflect a 100% payout target. The Service Company s short-term incentive plan costs allocated to CPE shall be capped at 25% of employees base pay. CenterPoint Energy shall refund to ratepayers all incentive compensation amounts approved by the Commission and included in base rates that are not paid out to employees under the program. The Company s short-term incentive compensation plan refund shall include the amounts of unpaid Service Company incentive compensation built into rates unless it shows that corporate allocations to Minnesota jurisdictional utility operations are the reason for lower payout. 71

78 Any short-term incentive expense adjustment shall be calculated in favor of consumers, where numeric discrepancies exist. CenterPoint Energy shall continue filing an annual report on incentive compensation within 30 days after incentive compensation is normally scheduled for payout. The report must include: a. a description of the incentive compensation plan; b. the accounting of amounts of unpaid incentive compensation built into rates to be returned to ratepayers; c. an evaluation of the incentive compensation plan s success in meeting its stated goals, including the payout ratio; d. a proposal for refund, if applicable; e. identification of each performance indicator and its associated scorecard information, such as the measure, the goal for various attainment levels (threshold, target, maximum), its funding weight and the actual result achieved; and to report the overall plan payout percentage attained relative to the target goal of 100%; and f. a separate reporting of the regulated portion of the Service Company incentive plan amount actually paid as compared to the amount included in base rates. 15. The Commission adopts the following modified ALJ Finding 418: The Department recommended that the entire Service Company STI amount be denied because CPE did not meet its burden of proof to justify its proposed recovery. Additionally, Alternatively, the Department recommended that if any STI from Service Company is included in corporate allocations, it should similarly be capped at 25%. DOC Initial Brief at The Commission finds that CenterPoint s purchase of its Nicollet Mall headquarters was prudent. CenterPoint shall apply a cost allocation of 48% Regulated and 52% Non-Regulated for the headquarters building. CenterPoint may therefore include $12,025,047 in rate base. The Commission adopts the following findings as modified: 339. The Administrative Law Judge disagrees. The hearing record establishes that CPE did undertake net present value analyses, albeit incomplete and not part of the initial rate case filing, which compared the benefits of obtaining a follow-on lease as opposed to purchase of a headquarters space. The analyses considered all costs (both capital and expense) of each option over a period of 40 years, before recommending the purchase of 505 Nicollet. The net present value analysis purportedly reviewed a series of key factors underlying the decision, including, capital expenditures, CPE s return on capital invested, depreciation, deferred taxes, cost of capital, and operating expenses. [FN omitted] 72

79 As Ms. Campbell testified, CPE s Net Present Value analysis was not provided in its initial filing, and was only provided in response to DOC information requests. The analysis provided in response to DOC IRs 137 & 171 did not included the rate of return for DOC 137; further both analyses provide inadequate support for assuming: 1) increasing rent expense from $12.97 per square foot to $26.37 per square foot and 2) a one-year increase in operating and maintenance expenses from $1.233 in 2016 million to $3.436 million in 2017, a 178 percent increase. Both analyses raise concerns as to reasonableness of Mr. Nesvig s conclusions The Company s analyses showed the net present value of the costs to purchase the building of $32.4 million, which was lower than the projected $40.5 million net present value cost of renewing the LaSalle lease over the same period. [FN omitted] The NPV analysis referred by the CPE, however, was in response to DOC IR No. 137 and is incorrect because the Company did not include its rate of return costs in the calculations, despite asking ratepayers to pay these costs Following the Department s recommendation, CPE performed a revenue requirements analysis that used a 50-year building life and 10 percent salvage value for the building as assumptions, alongside other operating expenses and capital related inputs. According to the Company, Tthis analysis showed a lower net present value of the revenue requirements for purchase of 505 Nicollet, over a 50-year period $46.7 million as compared to the projected $64.9 million net present value revenue requirement for leasing space during this same time period. [FN omitted] 342. In total, the Company prepared three net present value analyses, and it claimed that each of these concluded that the purchase of 505 Nicollet was more beneficial to CPE s ratepayers than continuing to lease a similar set of spaces from others. [FN omitted] 343. Mr. Nesvig testified persuasively that, over the course of either a 40-year period or a 50-year period, leasing would be more costly and risky to ratepayers than the purchase of the 505 Nicollet property. Lease rates can reasonably be expected to increase over time, increasing the annual cost of leasing arrangements over five decades. By contrast, he claimed that if CPE owned its headquarters building, the Company s rate base will continue to decline over time, thereby lowering the annual cost of service. [FN omitted] 73 DOC Ex. 519 at (Campbell Surrebuttal). 74 Tr. Vol. 4 at (Campbell). 73

80 [New Finding]. The Commission does not find the Company s lease versus cost analysis to be complete or compelling. As Ms. Campbell explained the deficiencies in CPE s analysis, as follows: As discussed in my Direct and Surrebuttal Testimony, 75 the Company did not provide an adequate cost/benefit analysis of lease versus own; it did not provide any cost effective analysis of leasing compared to owning the headquarters in its initial filing and only focused on changes in expense that were misleading and did not reflect the total impact of the option to own (did not include return on rate base or reasonable depreciation assumptions). Only until DOC issued information request 137 did the Company provide such an analysis albeit incomplete in that it did not include rate of return on rate base and, in response to DOC information request nos. 137 and 171, provided inadequate support for assuming increasing rent expense from $12.97 per square foot to $26.37 per square foot, and a one-year increase in operating and maintenance expenses from $1.233 million in 2016 to $3.436 million in 2017, a 178 percent increase. Both analyses raise concerns as to reasonableness of Mr. Nesvig s conclusions For much of the proceeding it was not clear how much headquarters space CPE needed for its Minnesota-jurisdictional operations and how much space was needed by non-regulated entities and work done for other jurisdictions. This lack of clarity was due to the Company s inclusion of very little information in its initial filing or in responses to information requests to show that its proposed headquarters costs were reasonable for ratemaking purposes together with inconsistent and at time confusing discovery responses. 77 Specifically, the Company indicated in a response to an Information Request that 38 percent of this space was attributable to other jurisdictions. [FN omitted] 345. Company witness Mr. Nesvig testified in Surrebuttal Testimony that the 38 percent figure is the amount that CPE allocates to Service Company operations. [FN omitted] 75 Ex. 516 at (Campbell Direct); DOC Ex. 519 at (Campbell Surrebuttal). 76 DOC Ex. 519 at (Campbell Surrebuttal). 77 See, e.g., Tr. Vol. 4 at 8-15, (Campbell); Ex. 516 at 54-58, (Campbell Direct); Ex. 519 at (Campbell Surrebuttal). 74

81 346. Importantly, however, The Company asserts that the costs apportioned to Service Company (such as the costs for office space used by the Service Company s call center) are allocated back to CPE s Minnesota-regulated operations when the work that the Service Company employees perform relates to the utility s Minnesota customers. [FN omitted] 347. The Company claims that Service Company departments directly bill the costs of their services to specific users whenever this is possible. When it is not possible, the costs are allocated using familiar cost-causation methods. [FN omitted] [New Finding]. The Department s investigation raised reasonable doubt as to the Company s claim in Surrebuttal Testimony that 38 percent of the Minnesota regulated square footage, which correlates to the rate base assigned to Minnesota regulated, is being allocated by CenterPoint to Other Jurisdictions/Service Company. Such an allocation was not shown by CPE to actually have occurred, nor did the Company show that those costs are than allocated back to Minnesota Regulated. 78 [New Finding] No Company witness provided any basis to show that it is reasonable for CenterPoint to use one allocator for expense and a different higher allocator for rate base for the same asset which is the new headquarters. The Commission is persuaded by the testimony of Ms. Campbell that it is CPE s use of two different allocators for the same asset (the headquarters) is unreasonable. 79 [New Finding]. CenterPoint did not reconcile record inconsistencies of its own making. The Company s 78% allocation of rate base headquarters to MN Regulated is inconsistent with its responses to DOC IRs 137B and 168, and is inconsistent with its own allocation to headquarters expense. The 78 percent rate base allocation also is inconsistent with CPE s own expense allocation for its previous headquarters lease agreement Notwithstanding his use of the 38 percent figure, with respect to the headquarters Mr. Nesvig testified that the majority of the Service Company s headquarters cost returns to the Minnesota-regulated operations as a result of the work performed by the Service Company employees. Because the majority of the work that is performed by Service Company employees in the Minneapolis headquarters benefits Minnesota ratepayers, Mr. Nesvig stated that it is allocated back to regulated operations. [FN 78 Tr. Vol. 4 at (Campbell). 79 Tr. Vol. 4 at 8, 11 (Campbell); Ex. 516 at and NAC-8 (Campbell Direct); Ex. 519 at (Campbell Surrebuttal). 80 Tr. Vol. 4 at 8-15, (Campbell). 75

82 omitted] The Commission, however, is not persuaded by CPE s claim of that all such costs are allocated back to regulated. Additionally, as noted by the Department, the Company s first attempt to justify its high cost allocation to rate base for the headquarters was in Mr. Nesvig s Surrebuttal Testimony, but even then the Company did not show that all of the call center/service Company cost should be allocated to MN Regulated, since there are several other jurisdictions and the call centers are unified call centers that serve customers in other states. 81 Again, the Company did not address why it considers it reasonable for the rate base allocator to be different, and higher, than the expense allocator for the headquarters The Company s position is that the labeling of the figures on Mr. Nesvig s workpaper as Other Jurisdictions is regrettable, but the labeling does not change the underlying cost detail. Notwithstanding the Service Company s 38 percent share of headquarters costs, it is not true that this same percentage is incurred for, and provide benefits to, customers outside of Minnesota. A true and appropriate distribution of costs requires that they are allocated to the regulated and non-regulated entities that benefit from these expenditures. [FN omitted] [New Finding]. The Commission does not find the Company s testimony regarding 38 percent of headquarters costs allocated to the Service Company to be persuasive. The record simply does not show what happened to the missing 38 percent of headquarters costs that CPE suggests would have been allocated to Other Jurisdictions but were not. The Commission is persuaded by the testimony of Ms. Campbell that, even if such an allocation were made, it is doubtful that all of this 38 percent of costs (costs that might have been allocated to Other Jurisdictions) are reasonably allocated to Minnesota regulated customers. 82 Further, such a claimed allocation does not explain why the expense and rate base allocators should be different for rates in this matter, or why the expense allocator used for the new headquarters and for its previous headquarters lease would not be reasonable to use for rate base in this case For this reason, the $7,362,539 reduction to rate base urged by the Department is reasonablewould result in unjust, unreasonable and confiscatory rates. The Commission does not agree that Tthe best reading of the hearing record is that the reference to Other Jurisdictions was merely inartful perhaps even colossally so 81 Tr. Vol. 4 at 12-15, (Campbell). 82 Tr. Vol. 4 at (Campbell). 83 Tr. Vol. 4 at 10-15, (Campbell); Ex. 519 at (Campbell Surrebuttal). 76

83 given the sums involved but it should not result in exclusion of costs that are fairly attributable to providing gas service in Minnesota. [FN omitted] Instead, the Commission finds that CenterPoint did not demonstrate the reasonableness of its rate base allocation for its new headquarters or of the resulting depreciation adjustments On balance, Doubt must go to ratepayers. For this reason, the record demonstrates that CPE s facility decision-making process was sound and resulted in the best option for ratepayers. Iif CPE removes costs associated with Meet Minneapolis and uses both a 50-year life and 10 percent building salvage value for rate-setting purposes, as it has agreed to do, together with the two additional adjustments recommended by the Ms. Campbell, 84 the expenses and capital-related items for 505 Nicollet are properly recoverable in rates. [FN omitted] The necessary additional two adjustments are as follows: a. Reduce rate base assigned to Minnesota Regulated from the percent recommended by CPE to the 48 percent as determined and recommended in Ms. Campbell s Surrebuttal Testimony, which is consistent with the Company s own allocation of operating expenses for CPE s new headquarters (and its previous headquarter lease), resulting in a reduction to rate base of ($7,362,539). 85 b. Given CPE s conflicting information request responses as to charging Minnesota Regulated $838, and $413, depreciation expense, the Department s recommended depreciation expense reduction of $490, to capture the difference in these depreciation expense amounts is reasonable. [alternative to b. If the Commission does not agree with the Department s depreciation expense reduction of $490,338, then at a minimum the Department recommends a reduction in depreciation expense due the Company s calculated and agreed-to reduction in depreciation expense of $274,022 for the headquarters based on the 84 The Report at Proposed Finding 276 adopts a third recommendation of Ms. Campbell s to reflects the adjustments recommended by OAG witness Ms. Lee and PUC information request no. 413, as agreed to by the Company, to remove the Meet Minneapolis tenant costs from the headquarters. See Ex. 519 at (Campbell Surrebuttal). 85 Ex. 519 at (Campbell Surrebuttal). 86 CPE s response to DOC IR No. 168, Ex. 519 at NAC-S-2 (Campbell Surrebuttal). 87 Ex. 516 at NAC-8 (Campbell Public Direct). 88 $838,533 less $413,825 and adjustments for 50 year remaining life and positive 10 percent salvage rate as shown in DOC table Regulated Headquarters Rate Base and Depreciation, Ex. 519 at (Campbell Surrebuttal). 77

84 Department s assumption changes of 50 year remaining life and positive 10 percent salvage rate. 89 ] CenterPoint shall be required to depreciate Nicollet Mall over 50 years, with a positive 10% salvage value for ratemaking purposes. CenterPoint shall accordingly reduce its Depreciation Expense by $490,338. The Commission approves recovery of O&M expenses associated with CenterPoint s Nicollet Mall headquarters, including those attributable to the PRIME project. CenterPoint shall exclude all rate base and O&M costs related to the Meet Minnesota lease. The Commission approves recovery of CenterPoint s net $2 million environmental costs over the next two years. Recovery shall be limited to this amount and the Company will be required to track and return to ratepayers any excess. CenterPoint shall calculate plant retirements using the OAG s category-by-category methodology. In addition to removing the appropriate Plant in Service and Accumulated Depreciation amounts, CenterPoint s calculation should also include adjustments to any other applicable ancillary accounts, such as Depreciation Expense, Income Taxes, etc. CenterPoint s Net Distribution balance and test year rate base shall be reduced by $2,268,003, resulting in a test year depreciation expense reduction of $42,493. The Commission approves CenterPoint s Cash Working Capital methodology. The Company s final Cash Working Capital calculation shall reflect decisions made in this rate case. The Commission approves CenterPoint s Payroll Inflation Factor of 5.32% and General Inflation Factor of 3.96%. The Commission approves recovery of the executive compensation expense amount, minus Mr. Carroll s compensation costs for amounts in excess of his Minnesota jurisdictional share of compensation for participating on the independent board of directors. The Commission adopts the following finding in lieu of ALJ finding 437: 437. CenterPoint has already recovered its 2013 rate case expenses through its 2013 rate case. Therefore, recovery of $11,736 in 2013 rate case expenses in this proceeding is inappropriate. 28. The Commission declines to adopt the ALJ s findings 493, 494, 498, 501, 502, 511, 516, 521, 522, 546, and Ex. 519 at (Campbell Surrebuttal). 78

85 29. The Commission adopts ALJ finding 492 modified as follows: 492. The investment in these mains is caused by the requirement to install facilities that connect all customers to the common distribution network and install facilities with sufficient capacity to meet the demands of all customers. Because the distribution system is a jointly-caused cost, the costs cannot be assigned directly to any single class or group of customers. The costs must be assigned using a classification method. A classification method is a system for assigning costs based on economic and engineering theory. 30. The Commission adopts the following findings from the OAG s exceptions to the ALJ s Report with the modifications indicated: 493. The parties in this proceeding have suggested three classification methods. One method that the Commission has applied in the past is the minimum system theory. The minimum system approach reasons that the costs of the capacity in the distribution system are caused by peak demand, and should be classified as a demand cost. Some of the costs, however, are related to connecting a customer to the minimum system, a distribution system with little or no capacity. Because this minimum system has no capacity, the cost to connect a customer should be the same. Because the cost is the same, it would vary on the basis of the number of customers in a class and should be classified as a customer cost. The cost of the minimum system can be estimated in several ways, including the minimum size study and the zero intercept study. The OAG has suggested that the cost of the minimum system could be more accurately estimated using a method it calls the alternative minimum size study ( AMSM ) The basic customer theory reasons that the distribution system is installed to ensure that it can satisfy customer demand at the system s peak. Because it is built to serve peak demand, not one customer individually, the basic customer classifies the distribution system as a demand cost The peak-and-average theory reasons that some part of the distribution system is built to provide the regular amount of energy that is used all of the time represented by the system s load factor. These costs are classified as commodity costs because they are related to how much energy is sold. The rest of the costs of the system are installed to serve peak demand, and so are classified as demand costs. 3 79

86 CenterPoint and the Department recommend that the Commission use the minimum system theory, estimated using a 2 inch minimum size method, to classify and allocate the distribution system. The OAG suggests that the Commission consider the results of each of the three theories. The OAG s suggestion is premised on a concern that the results of the CCOSS vary wildly depending on this single modeling decision. Using the Company s 2-inch minimum system theory establishes a revenue deficiency of $60 million for the residential class; but using the peak-and-average method shows a revenue sufficiency of $2 million a swing of $62 million in cost assignment. The OAG suggests that it would not be reasonable to base a swing of such magnitude on a selection of one arbitrary method over another, so instead recommends that the Commission consider each of the three theories As an alternative to the minimum size method, the OAG suggested that the cost of the minimum system should be estimated using the AMSM. The OAG noted that both the minimum size and zero intercept methods are problematic for measuring the cost of the minimum system. The AMSM is, like the minimum size and zero-intercept methods, just another way to estimate the cost of the minimum system. It is based on the same principles, and it attempts to estimate the same result. The fundamental problem with minimum size method is that it includes the cost of the capacity in its size in the minimum system. The zero intercept method tries to correct this problem by removing the cost related to the size of the main. To say it a different way, the difference between the methods is that the zero-intercept methods removes the cost related to the size of the main from the equation. The AMSM does the same thing by applying a different technique based on readily available data CenterPoint and the Department criticized the AMSM because it excludes the material cost necessary to make service available to customers. The OAG responded by noting that this argument is not sound because it ignores the fact that the AMSM is estimating the same thing as the zero-intercept study: a distribution system where the cost of the main itself has been removed. As the OAG reasoned, disagreeing with the theory behind the AMSM, then, is the same thing as disagreeing with the theory behind the zero-intercept method The Commission agrees that the AMSM is a reasonable method to calculate the cost of the minimum system in this case. It improves upon the results of the minimum size method, and is more reliable than the zero intercept study conducted by CenterPoint. 80

87 The OAG responded to Dr. Ouanes s argument about engineering principles. Mr. Nelson testified that the basic customer method is based on engineering principles. It is based on the engineering principle that the distribution system must be built to satisfy the demand of the system on its peak and design day. If the system does not meet the requirements of peak or design day, then the system must be upgraded to meet those requirements. As a matter of engineering principle, the distribution system is built to satisfy peak demand and because of that it should be classified as demand. Mr. Nelson also testified that, in fact, it is the minimum system that is not based on engineering principles. The minimum system does not actually exist, and is an imaginary system that depends on imaginary engineering and operational realities that the Company controls The OAG also responded to CenterPoint s argument that the basic customer method is used less frequently today than in the past. First, the OAG identified several flaws in the survey that CenterPoint conducted to support its argument. And, even with those flaws in the survey, the OAG noted that the Company admitted that almost half of the jurisdictions in the country use the basic customer or peak and average methods. The OAG also noted that, of the jurisdictions that continued to use the minimum system approach, nearly half relied on the zero intercept method rather than the minimum size method. Taken in that light, the OAG noted, CenterPoint s minimum size method in some ways represents a minority position when compared to the rest of the country. But, according to the OAG, the more important conclusion was that each approach suggested in this case has been recognized as reasonable by multiple regulatory commissions. The OAG suggested that it would not be reasonable to shift $60 million in cost assignment to the residential class by choosing one classification method among several reasonable methods The OAG has persuasively demonstrated that tthe basic customer method is supported by reasonable economic theory, and has been used by many other regulatory commissions. The Commission determines that it is reasonable to include consideration of the basic customer method when setting rates in this case. 81

88 521. The peak-and-average method, like the basic customer method, reasons that the distribution system is built to satisfy the peak demand of the system. But the peak-and-average method also recognizes that some parts of the system are designed to supply the regular amount of energy that is required all of the time represented by the system s load factor. These costs are classified as commodity costs because they are related to how much energy is sold. The rest of the costs are required to serve peak demand, rather than regular load, and so are classified as demand costs. After this classification, the costs are allocated on non-coincident peak demand, because that is representative of the maximum amount that each class could contribute to peak costs The OAG produced evidence demonstrating that the economic theory behind the peak-and-average method is sound. The OAG also demonstrated that many other regulatory commissions in the country make use of the peak and average method. In addition, the OAG noted that the peak-and-average method is presented in the NARUC Electric Manual The Commission is persuaded that the peak-and-average method is a reasonable theory that is used by other regulatory bodies around the country, and that it is reasonable to consider the results of the OAG s peak-and-average method is setting rates in this case. 1 Ex. 406, at 36 (Nelson Direct). 2 NARUC Gas Manual, at 22 23; Ex. 406, at 33 (Nelson Direct); OAG Initial Brief, at NARUC Electric Manual, at 34; Ex. 406, at 33 (Nelson Direct); OAG Initial Brief, at Ex. 406, at 73:13 17 (Nelson Direct). 5 OAG Initial Brief, at Ex. 428, at 19:15 16 (Nelson Surrebuttal). 7 OAG Initial Brief, at Ex. 406, at 78:19 80:2 (Nelson Direct); Ex. 425, at 26 (Nelson Surrebuttal). 9 NARUC Electric Manual, at The Commission finds that the OAG questioned whether the Company s class-cost-of-service-study witness had testified consistently in prior rate cases when he represented different interests. The Commission declines to adopt the ALJ s findings and instead adopts proposed findings from the OAG s exceptions to the ALJ s Report: 82

89 564. The OAG s CCOSS is the only CCOSS that considers multiple methods of classification for the distribution system. The OAG derived its recommendation by seeking patterns in the revenue deficiencies or sufficiencies across multiple methods, and recommending changes based on those patterns The OAG s proposed revenue apportionment has other benefits as well. Specifically, the OAG proposed a rational methodology predicated on observing patterns in multiple CCOSS results, and using those patterns to inform revenue apportionment decisions. In addition, the OAG demonstrated that CenterPoint s revenue had a widely varying impact on different classes, while the OAG s revenue apportionment was more stable The Commission finds the OAG s recommendation to consider multiple CCOSSs persuasive, and will adopt the OAG s revenue apportionment. 1 OAG Initial Brief, at The Commission adopts the remainder of the ALJ s findings on the classification of distribution-system costs to the extent that they are consistent with this order. CenterPoint shall maintain its existing customer charges for all classes. The Commission adopts ALJ finding 600 modified to reflect that Minn. R governs the Commission s variance determinations. The Commission adopts ALJ finding 598, adding the following language: The OAG urged the Commission to use the variance procedures in Minn. R The Commission adopts the ALJ s finding 604, as well as findings modified to reflect the Department s position: While the 10-year weather normal is the most appropriate measure in this proceeding, the Commission is not persuaded that the 20-year weather normals should no longer be examined in future rate cases. The Commission reiterates the existing requirement for the Company to provide a comprehensive examination of the predictive power, volatility, and impact on the test year and future revenues of using 10, 15, and 20-year weather data in its sales forecast. 83

90 38. In future rate cases, CenterPoint shall a. employ an interim-rate methodology that preserves its existing rate design by removing cost-of-gas billings when determining and applying the interim-rate increase; b. present financial information in the following format: $1,000,000 total Company ($750,000 MN). A similar format will also be acceptable as long as the Minnesota-jurisdictional amount is clear; c. include and identify in the initial filing of future rate cases the incentive plan program documents when the incentive plan s costs are included in the test year for cost recovery; d. use a 15-year average of the Producer Price Index when developing its general inflation factor; e. present all testimony and supporting schedules consistently, using the same 12-month time period as CenterPoint s selected test-year period; f. provide transparent and reconcilable schedules in the event its selected test-year period differs from the financial, operating and budgeting reporting periods used in practice if the Company fails to comply, the Commission may reject the rate case filing as incomplete; and g. continue to comply with any existing rules and prior Commission orders governing the format of information submitted as part of the Company s rate case In the initial filing of its next rate case, CenterPoint shall a. provide a depreciation analysis of account (General Plant: Structures and Improvements, Nonleasehold); b. discuss how it proposes to handle any environmental costs over-recovery; c. quantify the plant retirement impact using both the FERC-account and the category-by-category methods; d. file a minimum-system study based on a statistically significant zero-intercept study; a basic-customer cost study; and a peak-and-average cost study, in addition to the two-inch pipe study it has traditionally used; and e. provide a substantive explanation and justification of its classification and allocation methods. The Commission adopts ALJ findings 400 and 596, modified to state the witness s name as Nancy Campbell. 84

91 41. The Commission adopts Finding 133 of the Administrative Law Judge s Report, as modified below: 133. For the second component of the dividend yield DCF calculation, the Department and the OAG relied upon the expected growth rate in dividends provided by three respected and widely-used investment research services: Zack s Investment Research (Zack s); Value Line; and Thomson First Call Consensus (Thomson). Specifically, it they used the three projected earnings growth rates (lowest, average and highest) provided by Zack s, Value Line, and Thomson CenterPoint s proposed New Market Development Agreement (as amended) is approved. CenterPoint s proposed Minimum Volume Agreement is approved. Within 30 days of the date of this order, CenterPoint shall make the following compliance filings: a. Revised schedules of rates and charges reflecting the revenue requirement and the rate-design decisions herein, along with the proposed effective date, and including the following information: i. Breakdown of Total Operating Revenues by type; ii. Schedules showing all billing determinants for the retail sales (and sale for resale) of natural gas. These schedules shall include but not be limited to: 1. Total revenue by customer class; 2. Total number of customers, the customer charge, and total customer-charge revenue by customer class; and 3. For each customer class, the total number of commodity- and demand-related billing units, the per unit commodity and demand cost of gas, the non-gas margin, and the total commodity- and demand-related sales revenues. iii. Revised tariff sheets incorporating authorized rate-design decisions; iv. Proposed customer notices explaining the final rates, the monthly basic service charges, and any and all changes to rate design and customer billing. b. A revised base cost of gas, supporting schedules, and revised fuel-adjustment tariffs to be in effect on the date final rates are implemented. c. A summary listing of all other rate riders and charges in effect, and continuing, after the date final rates are implemented. d. A schedule detailing the CIP tracker balance at the beginning of interim rates, the revenues (CCRC and CIP Adjustment Factor) and costs recorded during the period of interim rates, and the CIP tracker balance at the time final rates become effective. e. Because final authorized rates are lower than interim rates, a proposal to make refunds of interim rates, including interest to affected customers. 85

92 Persons wishing to comment on the compliance filing(s) shall do so within 30 days of the date it is filed. Comments are not invited on the proposed customer notice. 45. This order shall become effective immediately. BY ORDER OF THE COMMISSION Daniel P. Wolf Executive Secretary This document can be made available in alternative formats (e.g., large print or audio) by calling (voice). Persons with hearing loss or speech disabilities may call us through their preferred Telecommunications Relay Service. 86