BEFORE THE NEW MEXICO PUBLIC REGULATION COMMISSION IN THE MATTER OF THE APPLICATION OF PUBLIC SERVICE COMPANY OF NEW MEXICO FOR REVISION OF ITS RETAIL ELECTRIC RATES PURSUANT TO ADVICE NOTICE NO. 513 PUBLIC SERVICE COMPANY OF NEW MEXICO, Applicant ) ) ) ) ) ) ) ) ) ) ) Case No. 15-00261-UT RECOMMENDED DECISION August 4, 2016 TABLE OF CONTENTS I. STATEMENT OF THE CASE II. BACKGROUND 11 III. PNM’S REQUESTED, AND HEARING EXAMINER’S RECOMMENDED, REVENUE REQUIREMENTS 12 IV. LEGAL STANDARDS FOR RATEMAKING 14 V. STANDARD OF PROOF 16 VI. TEST PERIOD/CWIP/YEAR-END OR 13-MONTH AVERAGE RATE BASE/ESCALATION FACTORS 16 VII. 1 A. Test Period 17 B. Rate Base 18 C. CWIP 21 D. AG’s Proposed 30% Adjustment to Rate Base 23 E. Escalation Factors 1. 1.5% Escalation Factor Applicable to Non-Labor O&M Expenses 2. 1.5% Escalation Factor Applicable to Planned Outage Expenses 24 24 26 REVENUE REQUIREMENT ALLOCATION FOR FERC WHOLESALE 27 A. Background 27 B. City of Aztec 28 C. The Jicarilla Apache Nation 29 VIII. PNM’S TEST PERIOD CAPITAL STRUCTURE 29 IX. PNM’S TEST PERIOD COST OF DEBT 31 X. PNM’S TEST PERIOD COST OF PREFERRED STOCK 31 XI. RETURN ON EQUITY 31 i A. Positions of Parties and Staff 31 B. 2015 EPE Rate Case 32 C. The Constant Growth DCF Method Should be Used to Determine PNM’s Authorized ROE 35 D. Explanation of the DCF Constant Growth Method 36 E. Issues Regarding Calculation of ROE Using the Constant Growth DCF Method 1. Proxy Group 2. Growth Rates 37 37 42 F. Recommended Inputs to Constant Growth DCF Analysis 45 G. Results of Parties’ and Staff’s Constant Growth DCF Analyses 46 H. Proposed Adjustments 47 I. 49 Hearing Examiner’s Recommended ROE XII. PNM’S TEST PERIOD WEIGHTED AVERAGE COST OF CAPITAL 50 XIII. RENEWABLE ENERGY RATE RIDER 50 XIV. FUEL AND PURCHASED POWER COSTS 52 A. Elimination of Fuel Misallocation 52 B. Fuel Handling Expenses and Purchases and Sales of Spinning Reserves 64 XV. C. Revenues from Chemical Pretreatment of coal for SJGS 69 D. WRA’s Risk Sharing Proposal 70 PALO VERDE CAPACITY 71 A. Background/History 71 B. PNM’s Current Interests in PV Units 1 and 2 78 C. PNM’s Responsibilities for O&M, Leasehold Improvements, and Decommissioning Expense 1. Governing Documents 2. Lease Payments 3. O&M 4. Leasehold, Common Plant, Capital Improvements & Depreciation 5. Decommissioning 81 81 82 82 82 83 ii D. Preservation of PRC’s Ratemaking Authority Including Recovery of Plant Improvements and Recovery of Decommissioning Costs 83 E. SUMMARY OF THE PARTIES’ POSITIONS 1. PNM 2. WRA 3. NEE 4. City/County 5. ABCWUA 6. AG 7. NMIEC 8. CFRE 9. Staff 85 85 85 85 85 85 86 86 87 87 F. PNM’s Decisions to Extend the Leases and to Repurchase the 64.1 MW Were Not Prudent 1. Prudence 2. Failure to Consider Alternatives 3. Insufficient Notices to PRC 87 87 88 98 G. PNM Did Not Prove that It Repurchased the 64.1 MW at FMV 100 H. Alleged Double-Counting of Improvements 105 I. 106 Sale/Leaseback Payments J. Hearing Examiner’s Recommendation XVI. BALANCED DRAFT 107 110 A. Background on Installation of Balanced Draft 111 B. Analysis of PNM’s Arguments in Support of Its Claim that Installation of Balanced Draft Was Prudent 116 XVII. PREPAID PENSION ASSET 120 XVIII. LAS VEGAS DECOMMISSIONING 128 XIX. NEW REGULATORY ASSETS AND LIABILITIES 128 A. General 128 B. Alvarado Square Lease Regulatory Asset 131 C. Time Of Use Regulatory Asset 133 D. Rate Case Expense Regulatory Asset 133 E. Department of Energy Refunds Regulatory liability 133 iii F. Credit Card Program Regulatory Asset 134 G. Regulatory Liability to Record Timing Difference Relating to AROs and Coal Mine Reclamation Obligations 135 H. Regulatory Liability to Record Excess Deferred State Income Taxes 140 I. XX. Regulatory Asset for State Net Operating Carryforward Impairment Loss EXPENSES 142 149 A. Depreciation Rates 1. PNM’s Existing Depreciation Rates 2. PNM’s Proposed Depreciation Rates — In General 3. Public Utility Depreciation Concepts 4. Production Plant 5. Transmission and Distribution (T&D) Plant 6. Net Salvage for Account 373 (Streetlighting) 149 150 150 151 152 154 164 B. Amortization Period for Reserve Differences on Amortized Accounts 166 C. Rate Case Expenses 167 D. Nuclear Decommissioning Trust Fund Contributions 170 E. Four Corners Coal Contract 172 F. Payroll Tax Expense 174 G. Non-Qualified Retirement Plan Expense 174 H. Dues and Fees 175 I. 177 Advertising Expense J. Employee Meals and Employee Expenses 178 K. Palo Verde Incentive Compensation 178 L. PNM Employee Incentive Compensation Programs 179 M. ROW Expenses 183 XXI. PNM’S EMBEDDED COST OF SERVICE STUDY: FUNCTIONALIZATION; CLASSIFICATION; ALLOCATION 184 A. Embedded Cost of Service Study 184 B. Classification 1. generation O&M expenses other than fuel and purchased power 2. Fixed Fuel Transportation and Demand PPA Costs 185 186 187 iv 3. Coal Mine Decommissioning Costs 4. Distribution Costs C. Allocation 1. Background 2. Generation Allocation Method 3. Transmission Allocation Method 4. Fuel Costs XXII. RATE DESIGN 188 188 191 191 192 197 199 199 A. Banding 1. PNM’s Proposal 2. Banding and Allocation of Discounts from Interruptible Rate 3. Rate 11B — Water & Sewage 4. Analysis and Hearing Examiner’s Recommendation on Banding 201 201 203 209 213 B. PNM’s Proposed Changes to Its Fixed Customer Charges 217 C. Proposed Residential Class Rates 1. Positions of Parties and Staff 2. Commission Precedent on Residential rate design 3. Analysis and Recommendation 4. Small Power Class 5. Irrigation Class 6. Water & Sewage Class 219 219 220 224 228 229 229 D. Data Gathering Proposals 230 E. Non –Residential Three-Tier Rates 1. Load Factor and Demand Charges in General 2. PNM’s Proposal in General 3. Intervenors Served Under Three-Part Tariffs with Demand Charges 4. Rate 3B/3C — General Power 5. Rate 4B — Large Power 6. Rate 5B — Large Service 7. Rate 15B — Universities 8. Rate 30B — Manufacturing 9. Rate 33B — Station Power 230 230 231 234 234 239 241 241 242 242 F. Proposed Rate 35B: New Higher Load Factor Large Customer Class 242 G. Time of Use Rates 243 H. Streetlighting Tariff 1. Proposed Continuation of Consolidation Adjustment Rider 2. Simplify Tariff 3. Increased Customer Choice 4. Maintenance for Customer-Owned Lights 5. Positions of Staff and Intervenors 247 247 247 248 248 248 I. 251 Bill Comparisons v J. City/County’s Proposal for Separate Rate Classes Or Cost-Based Rates Under Existing Rate Classes 251 K. Rate 16 — Special Charges 253 L. Rate 23 — Special Contract Service 254 M. Economic Development Rider 254 N. Tariff Revisions and Clean-Up 256 O. Elimination of Consolidation Adjustment Rider 256 P. Modifications to Voltage Adjustment Factors 257 Q. Split of Demand-Related Revenue Requirement Between Seasons 257 XXIII. PNM’S PROPOSED REVENUE BALANCING ACCOUNT 258 A. The Efficient Use of Energy Act 258 B. History of Decoupling in New Mexico 259 C. PNM’s Proposed Revenue Balancing Account (RBA) 265 D. Positions of Staff and Intervenors 269 E. Analysis/Recommendation 270 XXIV. PAYMENT CENTERS 274 XXV. FINDINGS OF FACT AND CONCLUSIONS OF LAW 274 XXVI. DECRETAL PARAGRAPHS 275 vi TABLE OF AUTHORITIES This Recommended Decision uses short-form citations to some authorities. The following table shows the full citation for the short-form citations. Short-form Citation AG v. PRC PNM Gas Servs. 2015 EPE Rate Case 2015 PNM RPS Case 2013 PNM Fuel Clause Case 2012 SPS Rate Case San Juan Case 2010 PNM Abandonment Case PNM Renewable Rider Case 2010 PNM Rate Case 2008 PNM Fuel Clause Case 2008 PNM Rate Case 2007 PNM Electric Rate Case 2007 SPS Rate Case 2006 PNM Gas Rate Case Full Citation New Mexico Attorney General v. New Mexico Pub. Regulation Comm’n, 2015-NMSC-032. PNM Gas Servs. v. New Mexico Pub. Util. Comm’n, 2000-NMSC-012, 129 N.M. 1. Case No. 15-00127-UT, Recommended Decision (2-16-16), partially adopted by Final Order Partially Adopting Recommended Decision (6-816) Case No. 15-00166-UT, Recommended Decision (10-20-15), as corrected by Errata Notices (10-23-15 & 10-28-25), partially adopted by Final Order Superseding Vacated Final Order Issued on November 18, 2015 (2-3-16) Case No. 13-00187-UT, Certification of Stipulation (3-11-14), adopted by Final Order (4-23-14) Case No. 12-00350-UT, Recommended Decision (1-23-14), as supplemented by Supplemental Recommended Decision (2-12-14), partially adopted by Final Order Partially Adopting Recommended Decision (3-26-14) Case No. 13-00390-UT, Certification of Stipulation (4-8-15), as modified by Certification of Stipulation 11-16-15, approved by Final Order (12-1615) Case No. 12-00264-UT, Recommended Decision (1-21-11), adopted by Final Order (2-22-11) Case No. 12-00007-UT, Recommended Decision 53-54(6-19-12), as corrected by Errata Notice, adopted by Final Order (8-14-12), as corrected by Errata Notice (8-24-12), as clarified by Order On Rehearing (10-9-12). Case No. 10-00086-UT, Certification of Stipulation (6-21-11), partially adopted by Final Order Partially Approving Certification of Stipulation (728-11) Case No. 08-00092-UT, Final Order (5-22-08) Case No. 08-00273-UT, Final Order Conditionally Approving Stipulation (6-18-09) Case No. 07-00077-UT, Recommended Decision (3-6-08), partially adopted by Final Order Partially Adopting Recommended Decision (424-08) Case No. 07-00319-UT, Corrected Recommended Decision (7-31-08), partially adopted by Final Order Partially Adopting Recommended Decision (9-25-08) Case No. 06-00210-UT, Recommended Decision (5-23-07), partially adopted by Final Order Partially Adopting Recommended Decision (629-07) vii Short-form Citation Attorney General v. PRC Pub. Util. Deprec. Practices Accounting for Public Utils. Bonbright Cost Allocation Manual CWIP ECCOSS FERC FMV FPPCAC O&M NBV PUA PV REA SJGS or SJ FMV NBV Full Citation Attorney General v. New Mexico Pub. Regulation Comm’n, 2011-NMSC034, National Association of Regulatory Commissioners, Public Utility Depreciation Practices (August 1996) Robert L. Hahne & Gregory E. Aliff, Accounting for Public Utils. (2012). James C. Bonbright, Albert L. Danielsen, and David R. Kamerschen, Principles of Pub. Util. Rates (2nd ed. 1988). National Association of Regulatory Commissioners, Electric Util. Cost Allocation Manual (1992) Construction Work in Progress PNM’s Embedded Cost of Service Study Federal Energy Regulatory Commission Fair Market Value Fuel and Purchased Power Cost Adjustment Clause Operation and Maintenance Net Book Value New Mexico Public Utility Act Palo Verde Nuclear Generating Station New Mexico Renewable Energy Act San Juan Generating Station Fair Market Value Net Book Value viii TABLE OF ATTACHMENTS Attachment A B C D E F G H I Description Dollar Impact of Each of Hearing Examiner’s Recommended Changes on PNM’s Proposed Revenue Requirement Test Period Cost of Service Under Hearing Examiner’s Recommendations Sample Proposed Bill Using Method A Comparison of Rate Class Revenue Allocations, Before Banding, Using the 12CP Versus 3S1WCP Methods to Allocate Transmission Demand Costs Revenue Allocations Among Classes Using Hearing Examiner’s Recommended Revenue Requirement and Banding Proposal Class Movement Toward/Away from Unity Under Current and Proposed Rates and Hearing Examiner’s Recommended Rates Bill Comparison for Non-Capped/Non-Exempt Customers Bill Comparison for Exempt Customers Bill Comparison for Large Capped Customers ix Carolyn R. Glick, Hearing Examiner for this case, submits this Recommended Decision to the New Mexico Public Regulation Commission (PRC or Commission) under 1.2.2.29(D)(4) and 1.2.2.37(B) NMAC. The Hearing Examiner recommends that the PRC adopt this Recommended Decision in its Final Order. I. STATEMENT OF THE CASE On August 27, 2015, PNM filed with the Commission its base rate case filing, which included: (i) the Application for Revision of Retail Electric Rates; (ii) the transmittal letter in accordance with 17.9.530 NMAC (“Rule 530”); (iii) Advice Notice No. 513; (iv) proposed rate schedules; and (v) direct testimony and exhibits prepared by numerous PNM witnesses. On September 9, 2015, the Commission issued an Order Suspending Rates and Designating Hearing Examiner that suspended the proposed rates filed by PNM in its Advice Notice No. 513 for a period of nine months commencing on October 1, 2015, and appointed the Undersigned as Hearing Examiner to preside over this matter. The following parties filed Motions for Leave to Intervene: x x x x x x x x x x x x x x x x x 1 The Albuquerque Bernalillo County Water Utility Authority (ABCWUA) The New Mexico Attorney General (AG) Freeport-McMoRan Chino Mines Company (FMI) New Mexico Gas Company (NMGC) The Kroger Company Western Resource Advocates (WRA) The Coalition for Clean Affordable Energy (CCAE) Walmart Stores East, LP and Sam’s East, Inc. (Walmart) The International Brotherhood of Electrical Workers Local Union No. 611, AFL-CIO (Union) The New Mexico Industrial Energy Consumers (NMIEC) The Buckman Direct Diversion Board (BDDB) The Energy Freedom Coalition of America (EFCA) Southwest Generation Operating Company, LLC Bernalillo County AARP1 New Energy Economy (NEE); Clarke Metcalf; AARP later withdrew its intervention. 1 x City of Albuquerque x Athena Christodoulou x Citizens for Fair Rates and the Environment (CFRE) Bernalillo County and the City of Albuquerque participated jointly and are referred to collectively as the City/County. PNM filed an Affidavit of Publication and Mailing, attesting that PNM published the Notice (i) in the Albuquerque Journal on October 22, 2015; (ii) in the Alamogordo Daily News on October 16, 2015; (iii) the Las Cruces Sun-News on October 30, 2015; and (iv) the Union County Leader on October 7, 2015. PNM also attested that it mailed or e-mailed the Notice to its customers by October 30, 2015, and posted the Notice on its website. A hearing to take public comment occurred on April 7, 2016. The public hearing began on April 11, 2016, and ended on April 29, 2016. The following witnesses testified: For PNM: x x x x x x x x x x x x x x x x x x x Gerard Ortiz, PNM’s Vice President of Regulatory Affairs Henry Monroy, PNMR’s Director of Internal Audit & Cost of Service Robert Hevert, Principal in Sussex Economic Advisors Chris Olson, PNM’s Vice President, Generation Aubrey Johnson, PNM’s Vice President of New Mexico Operations Dane Watson, Principal in Alliance Consulting Dr. Ahmad Faruqui, Principal in the Brattle Group Susan Taylor, PNM’s Manager-Utility Margins Elizabeth Eden, PNMR Service Company’s Vice President and Treasurer Jason Peters, PNMR’s Director, General Accounting Erik Buchanan, PNMR’s Assistant Controller for Shared Services Laurie Monfiletto, PNMR’s Vice President, Human Resources Roger Larsen, PNM’s Manager of Marketing & Energy Efficiency Outreach Matthew Harland, PNM’s Director of Income Tax Stella Chan, PNM’s Director of Pricing & Load Research Julio Aguirre, PNM’s Senior Pricing Analyst Daniel Hansen, Vice President at Christensen Associates Energy Consulting Stewart Houck, PNMR’s Senior IT Manager Gene Wickes, Senior Consultant & Actuary at Willis Towers Watson For NMIEC: x Michael Gorman, Managing Principal, Brubaker & Associates, Inc. Recommended Decision Case No. 15-00261-UT 2 x James Dauphinais, Managing Principal, Brubaker & Associates, Inc. x Brian Andrews, Consultant, Brubaker & Associates, Inc. x Nicolas Phillips, Consultant, Brubaker & Associates, Inc. For ABCWUA: x William Dunkel, Principal, William Dunkel and Associates x James Dittmer, Senior Regulatory Consultant, Utilitech, Inc. x Joseph Herz, Vice President, Sawvel and Associates, Inc. For CCAE: x John Howat, Senior Policy Analyst, National Consumer Law Center x Ralph Cavanagh, Energy Program Co-Director, Natural Resources Defense Council x Adam Bickford, Senior Associate, Utilities Program, Southwest Energy Efficiency Project For the AG: x Andrea Crane, President, The Columbia Group, Inc. x Randall Woolridge, Professor of Finance, Pennsylvania State University x Doug Gegax, Director, Center for Public Utilities, New Mexico State University For WRA: x Douglas Howe, Consultant For NEE: x David Van Winkle, Financial/Technical Advisor For the City of Albuquerque/Bernalillo County: x August Ankum, Senior Vice President & Chief Economist, QSI Consulting, Inc. x Tony Gurule, Energy & Sustainability Manager, Department of Municipal Development, City of Albuquerque For Kroger: x Neal Townsend, Principal, Energy Strategies, LLC For Walmart: x Steve Chriss, Senior Manager, Energy Regulatory Analysis, Walmart Stores, Inc. For Staff: x John Reynolds, Staff Economics Bureau Chief x Vincent De Cesare, Staff Economist Recommended Decision Case No. 15-00261-UT 3 x Heidi Pitts, Staff Economist x Charles Gunter, Staff Accounting Bureau Chief x Bruno Carrara, Staff Utility Division Director The Hearing Examiner admitted the following exhibits into evidence: PNM Exhibits: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Direct Testimony of Gerard Ortiz Oct. 9, 2015 Supplemental Testimony of Gerard Ortiz Rebuttal Testimony of Gerard Ortiz February 24, 2016 Supplemental Testimony of Gerard Ortiz March 14, 2016 Supplemental Testimony of Gerard Ortiz March 25, 2016 Surrebuttal Testimony of Gerard Ortiz March 30, 2016 Reply Testimony of Gerard Ortiz Direct Testimony of Sheila Mendez2 Direct Testimony of Leonard Sanchez3 Direct Testimony of Aubrey Johnson Rebuttal Testimony of Aubrey Johnson April 8, 2016 Affidavit of Aubrey Johnson March 14, 2016 Supplemental Testimony of Stewart Houck Direct Testimony of Ahmad Faruqui Rebuttal Testimony of Ahmad Faruqui April 12, 2016 Updated Table Direct Testimony of Erik Buchanan Rebuttal Testimony of Erik Buchanan Direct Testimony of Robert Hevert Rebuttal Testimony of Robert Hevert Direct Testimony of Dane Watson Rebuttal Testimony of Dane Watson June 2003 Depreciation Study Rebuttal Testimony of Gene Wickes Direct Testimony of Daniel Hansen Rebuttal Testimony of Daniel Hansen March 25, 2016 Surrebuttal Testimony of Daniel Hansen Direct Testimony of Susan Taylor October 9, 2015 Supplemental Testimony of Susan Taylor Rebuttal Testimony of Susan Taylor February 4, 2016 Supplemental Testimony of Susan Taylor March 14, 2016 Supplemental Testimony of Susan Taylor March 30, 2016 Reply Testimony of Susan Taylor Direct Testimony of Gail Vavruska-Marcum, adopted by Laurie Monfiletto Rebuttal Testimony of Laurie Monfiletto Direct Testimony of Henry Monroy Rebuttal Testimony of Henry Monroy 2 Ms. Mendez is PNMR Service Company’s Director of IT Program/Portfolio Management & Quality. Ms. Mendez’s Direct Testimony was admitted into evidence via stipulation. She did not testify. 3 Mr. Sanchez is a PNM Associate General Counsel. Mr. Sanchez’s Direct Testimony was admitted into evidence via stipulation. He did not testify. Recommended Decision Case No. 15-00261-UT 4 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 68 69 70 72 73 74 75 76 April 8, 2016 Affidavit of Henry Monroy Direct Testimony of Chris Olson Rebuttal Testimony of Chris Olson Direct Testimony of Elizabeth Eden Rebuttal Testimony of Elizabeth Eden Direct Testimony of Matthew Harland Rebuttal Testimony of Matthew Harland Direct Testimony of Roger Larsen Table titled “Return on 64 MW of Palo Verde Unit 2” Direct Testimony of Jason Peters Rebuttal Testimony of Jason Peters February 22, 2016 Supplemental Testimony of Jason Peters Direct Testimony of Stella Chan Rebuttal Testimony of Stella Chan February 29, 2016 Supplemental Testimony of Stella Chan March 25, 2016 Surrebuttal Testimony of Stella Chan April 14, 2016 Affidavit of Stella Chan Summary of Errors – Stella Chan Testimonies Decision No. C15-1318, Decision Modifying RD, CO PUC (12-9-15) January 14, 2016 Regulatory Focus Direct Testimony of Julio Aguirre October 9, 2015 Supplemental Testimony of Julio Aguirre Rebuttal Testimony of Julio Aguirre February 24, 2016 Supplemental Testimony of Julio Aguirre March 14, 2016 Supplemental Testimony of Julio Aguirre March 30, 2016 Reply Testimony of Julio Aguirre March 31, 2016 Supplemental Testimony of Julio Aguirre April 5, 2016 Supplemental Testimony of Julio Aguirre Summary of Errors – Testimonies of Julio Aguirre NMED Permit No. 0063-M9 Dec. 24, 2015 e-mail from Douglas Howe January 12, 2016 e-mail from Richard Goodyear EIA Document EEI, Electric Utility Automatic Adjustment Clauses (2006) April 15, 2016 Regulatory Focus Utility Comparisons Listed of Pre-Filed Testimony of Chris Olson in Case No. 13-00390-UT NMIEC Exhibits: 2 4 5 6 7 8 9 10 11 12 Pollina Corporate Top 10 Pro-Business States for 2015 Case No. 10-00086-UT, pages 3 & 15 of Robert Hevert’s Direct Testimony PNM Response to Interrogatory NMIEC 8-11 Case No. 15-00296, Att. DAW-2, pp. 41-42 PNM Response to Interrogatories NMIEC 8-7 through 8-9 PNM Response to Interrogatory ABCWUA 19-18 PNM Response to Interrogatory AG 2-30 PNM Response to Interrogatory AG 1-65 Pages from New Mexico Legislature’s Website PNM Response to Interrogatory NMIEC 8-1 Recommended Decision Case No. 15-00261-UT 5 13 14 15 16 17 18 Direct Testimony of Michael Gorman Direct Testimony of Brian Andrews Direct Testimony of James Dauphinais March 23, 2016 Response Testimony of James Dauphinais Direct Testimony of Nicholas Phillips Rebuttal Testimony of Nicholas Phillips Staff Exhibits: 1 2 6 7 8 9 10 11 12 13 14 15 16 Form 10-K for 2015 NRDC’s Third Annual Energy Report Direct Testimony of Elisha Leyba-Tercero, adopted by John Reynolds Direct Testimony of Vincent De Cesare Direct Testimony of John Reynolds Direct Testimony of Heidi Pitts Rebuttal Testimony of Heidi Pitts Direct Testimony of Charles Gunter Rebuttal Testimony of Charles Gunter Direct Testimony of Bruno Carrara March 4, 2016 Affidavit of Bruno Carrara March 8, 2016 Affidavit of Bruno Carrara March 23, 2016 Affidavit of Bruno Carrara WRA Exhibits: 1 2 3 4 5 6 7 8 9 10 11 12 13 April 2, 2014 PNM News Release RAP, Time-Varying and Dynamic Rate Design (2012) PNM Response to Interrogatory ABCWUA 19-12 PNM Response to Interrogatory ABCWUA 19-13 June 11, 2014 letter from Erik Bakken Evercore, San Juan Mine Opportunity (2014) October 1, 2014 e-mail from Jimmie Blotter PNM Response to Interrogatory WRA 3-31 PNM Exhibit JCA-3 Rebuttal Erratas to Douglas Howe’s Testimonies Direct Testimony of Douglas Howe March 23, 2016 Response Testimony of Douglas Howe WRA Response to PNM Interrogatory 1-4 CCAE Exhibits: 1 2 3 4 5 6 7 8 9 The PNM Resources Proposition PNM Resources’ Notice of 2015 Annual Meeting of Shareholders Direct Testimony of John Howat Direct Testimony of Ralph Cavanagh Rebuttal Testimony of Ralph Cavanagh Direct Testimony of Adam Bickford Rebuttal Testimony of Adam Bickford PNM Response to Interrogatory CCAE 3-9 PNM Response to Interrogatory CCAE 1-15 Recommended Decision Case No. 15-00261-UT 6 ABCWUA Exhibits: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 PNM Response to Interrogatories ABCWUA 17-18 & 17-19 PNM Response to Interrogatory ABCWUA 19-1 PNM Response to Interrogatory ABCWUA 19-2 Direct Testimony of Dane Watson, Alaska Regulatory Commission PNM Response to Interrogatories ABCWUA 2-3 through 2-5 (5 pages) PNM Response to Interrogatory ABCWUA 19-3 PNM Response to Interrogatories ABCWUA 2-3 through 2-5 (4 pages) PNM Response to Interrogatory Staff 1-1 PNM Response to Interrogatory ABCWUA 19-14 PNM Response to Interrogatories ABCWUA 17-3 & 17-5 PNM Response to Interrogatory AG 1-04 Depreciation Study Update PNM Response to Interrogatory ABCWUA 19-17 PNM Response to Interrogatory AG 1-09 PNM Response to Interrogatory ABCWUA 18-3 PNM Response to ABCWUA Interrogatories 17-6, 17-7 & 17-9 through 17-11 PNM Response to ABCWUA Interrogatory 4-1 Direct Testimony of William Dunkel April 4, 2016 Surrebuttal Testimony of William Dunkel Direct Testimony of James Dittmer Corrections to Direct Testimony of James Dittmer Direct Testimony of Joseph Herz Corrections to Direct Testimony of Joseph Herz Rebuttal Testimony of Joseph Herz AG Exhibits: 2 3 5 6 7 8 9 PNM Resources’ Notice of 2016 Meeting of Shareholders, selected pages FASB No. 87 Errata to Direct Testimony of Andrea Crane Second Errata to Direct Testimony of Andrea Crane Direct Testimony of Andrea Crane Direct Testimony of J. Randall Woolridge Direct Testimony of Doug Gegax Kroger Exhibits: 1 2 3 Direct Testimony of Neal Townsend March 23, 2016 Supplemental Testimony of Neal Townsend Kroger’s Response to NMPRC Bench Request 1 Walmart Exhibits: 1 Direct Testimony of Steve Chriss COA/County Exhibits: Recommended Decision Case No. 15-00261-UT 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 PNM Response to Interrogatory CABQ-BERNCO 3-5 Direct Testimony of August Ankum Rebuttal Testimony of August Ankum PNM Response to Interrogatory CABQ-BERNCO 2-4 PNM Response to Interrogatory CABQ-BERNCO 2-7 PNM Response to Interrogatory CABQ-BERNCO 2-8 PNM Response to Interrogatory CABQ-BERNCO 2-23 PNM Response to Interrogatory CABQ-BERNCO 2-25 PNM Response to Interrogatory CABQ-BERNCO 2-26 PNM Response to Interrogatory CABQ-BERNCO 2-27 PNM Response to Interrogatory CABQ-BERNCO 2-30 PNM Response to Interrogatory CABQ-BERNCO 3-2 PNM Response to Interrogatory CABQ-BERNCO 3-3 NARUC Electric Utility Cost Allocation Manual Rebuttal Testimony of Tony Gurule Acuity Controls printout NEE Exhibits: 1 2 4 6 7 9 11 12 13 14 15 16 17 18 19 20 21 22 PNM Response to Interrogatories ABCWUA 17-7 & 17-8 PNM Response to Interrogatory CCAE 3-1 Pages from PNM Resources’ “plug in here” presentation PNM Response to Interrogatory NEE 1-8 PNM Response to Interrogatory NEE 5-2 PNM Response to Interrogatory NEE 1-12 PNM Response to Interrogatories NEE 1-27 & 1-28 PNM Response to Interrogatory NEE 1-4 PNM Response to Interrogatory Staff 8-6 Form 10-K, B-87 PNM Response to Interrogatory AG 2-1 PNM Response to Interrogatory ABCWUA 17-14 PNM Response to Interrogatory NEE 1-13 Pages 21-22 of PNM’s Response to Interrogatory NEE 1-2(2) PNM Response to Interrogatory NEE 2-3 January 29, 2016 Direct Testimony of David Van Winkle Errata to January 29, 2016 Direct Testimony of David Van Winkle March 18, 2016 Direct Testimony of David Van Winkle Commission Exhibits: 1 2 3 PNM’s March 25, 2016 FPPCAC Report Fuel Revenue Numbers Calculation James Bonbright, Principles of Public Utility Rates, pp. 494-511 During the hearing, the Hearing Examiner discussed on the record the likely need for a committee of Technical Assistants to provide her with revised cost of service revenue requirement calculations, rate base, rate design allocation, and revenue reconciliations to reflect Recommended Decision Case No. 15-00261-UT 8 the decisions to be made after briefing. Tr. (4-29-16) 3608-09. No party opposed the arrangement. In accordance with this arrangement, the following Technical Assistants signed and filed Confidentiality and Non-Disclosure Agreements: Scott Vogt, Julio Aguirre, Matthew Harland, Kyle Sanders, James Dittmer, Henry Monroy, Charles Gunter, and Vincent DeCesare. On May 18, 2016, the PRC issued an Order Reopening Proceeding, which extended the suspension period through August 31, 2016, and reopened the evidentiary portion of this case for the limited purposes of taking additional evidence on: 1. The net book value of the 64.1 MW of PV2 that PNM purchased in 2016 2. The net book value of the underlying assets of the renewed PV leases. The PRC ordered the filing of additional testimony and scheduled a supplemental hearing to begin on June 27, 2016. The supplemental hearing began on June 27, 2016, and ended on June 29, 2016. The following witnesses testified: For PNM: x x x x x Gerard Ortiz Henry Monroy Matthew Harland Elisabeth Eden Jason Peters For the AG: x Andrea Crane The Hearing Examiner admitted the following exhibits into evidence: PNM Exhibits: 77 78 79 80 81 82 83 May 25, 2016 Supplemental Testimony of Gerard Ortiz June 20, 2016 Supplemental Testimony of Gerard Ortiz May 25, 2016 Supplemental Testimony of Henry Monroy Initial Order, Case Nos. 13-00199-UT & 13-00200-UT E-mail filed on October 29, 2013, with attachment Final Order Terminating Proceeding & Closing Docket, Case No. 13-00199-UT May 25, 2016 Supplemental Testimony of Matthew Harland Recommended Decision Case No. 15-00261-UT 9 84 85 86 87 88 89 90 92 May 25, 2016 Supplemental Testimony of Elisabeth Eden June 10, 2016 Supplemental Testimony of Elisabeth Eden June 20, 2016 Supplemental Testimony of Elisabeth Eden June 24, 2016 Affidavit of Elisabeth Eden PNM Response to Hearing Examiner Bench Request May 25, 2016 Supplemental Testimony of Jason Peters June 20, 2016 Supplemental Testimony of Jason Peters PNM Supplement to Oral Testimony in Response to June 28, 2016 Bench Directive Staff Exhibits: 17 18 19 20 21 September 18, 2015 Sale Agreement November 20, 2015 Sale Agreement November 20, 2015 Sale Agreement FERC Docket No. EC15-107-000, Order Authorizing Disposition of Jurisdictional Facilities March 31, 2015 Letter from Ryan Anderson to Kimberly Bose ABCWUA Exhibits: 25 26 27 PNM Response to Interrogatory ABCWUA 20-7 Exhibit to PNM Response to Interrogatory ABCWUA 20-7 PNM Response to Interrogatory ABCWUA 21-3 AG Exhibits: 10 June 14, 2016 Supplemental Testimony of Andrea Crane NEE Exhibits: 23 24 27 28 30 31 Selected pages of John Reed’s Direct Testimony in Support of Stipulation in Case No. 13-00390-UT Selected pages of transcript in Case No. 13-00390-UT PNM Response to Interrogatory NEE 8-12 E-mail string PNM Response to Interrogatory NEE 9-3 PNM Response to Interrogatory NEE 8-1 Commission Exhibits: 4 5 Calculation Sheet Issues for Supplemental Brief On May 23, 2016, Staff and the following parties filed Initial Posthearing Briefs, addressing issues not related to the supplemental hearing: PNM, Walmart, Kroger, AG, NMIEC, ABCWUA, CCAE, WRA, NEE, and City/County. Recommended Decision Case No. 15-00261-UT 10 On June 3, 2016, Staff and the following parties filed Response Briefs: PNM, Walmart, Kroger, AG, NMIEC, ABCWUA, CCAE, WRA, NEE, City/County, and CFRE. On July 11, 2016, the following parties filed Supplemental Briefs relating to the supplemental hearing: PNM, ABCWUA, NEE, City/County, NMIEC, AG, and WRA. II. BACKGROUND PNM has about 500,000 retail customers in New Mexico. Generally, the New Mexico transmission system has been designed to transport electricity from base load coal and nuclearfired resources in and around Four Corners, eastern Arizona and Phoenix, to the large load centers in central and south-central New Mexico – namely, the Albuquerque, Santa Fe and El Paso metropolitan areas. Aubrey Direct 1, 6. PNM is a subsidiary of PNM Resources, Inc. (PNMR). Buchanan Direct 1. PNM’s current retail service rates were approved in the 2010 PNM Rate Case.4 They took effect on August 21, 2011. Monroy Direct 8. The Final Order in the 2010 PNM Rate Case eliminated previously existing separate tariffs for PNM North, PNM’s service territory before its acquisition of Texas-New Mexico Power Company (TNP), and PNM South, PNM’s service territory previously served by TNP. PNM North serves approximately 460,000 customers and consists of the metropolitan area of Albuquerque and the cities of Santa Fe, Las Vegas, Deming and Clayton and surrounding areas. PNM South serves approximately 50,000 customers and primarily consists of the cities of Silver City, Lordsburg, Alamogordo, Tularosa, and Ruidoso.5 The PRC approved a Stipulation subject to the Stipulating Parties’ agreement to amend the Stipulation consistent with the PRC’s Final Order. On August 11, 2011, an Amended Stipulation to Conform to Commission Order was filed. 5 Citations to pre-filed testimony are to the witness’ last name followed by an abbreviated reference to the title of the testimony and the page number. Citations to the transcript of the hearing are to “Tr. (date) page number.” 4 Recommended Decision Case No. 15-00261-UT 11 In the 2010 PNM Rate Case, PNM submitted an “Illustrative Cost of Service” to support the reasonableness of the rates proposed in the Stipulation. However, the Signatories to the Stipulation did not agree on any single cost of service study to derive the proposed rates. Certification of Stipulation 33. Nevertheless, the PRC treated the Illustrative Cost of Service as the cost of service supporting the stipulated rates because (i) no other detailed cost of service was submitted to support the stipulated rates; and (ii) challenges to the revenue requirement were to elements of the Illustrative Cost of Service. Certification of Stipulation 59. PNM claims that it faces a $123.5 million revenue deficiency arising primarily from: x Investments it has made since its last rate case that are not included in its rate base, which accounts for about 77% of the deficiency; and x Declining energy sales, which account for about 25% of the deficiency. Ortiz Direct 4-5. The table below shows PNM’s New Mexico Jurisdiction kWh sales for calendar years 2012-2014: Calendar Year NM Jurisdiction Sales (in kWh) 2012 2013 2014 8,760,908,453 10,279,188,064 10,204,991,649 From December 31, 2013 to December 31, 2014, PNM’s New Mexico jurisdictional sales decreased by 74,196,415 kWh. Exh. GTO-2 to Ortiz Direct. III. PNM’S REQUESTED, AND HEARING EXAMINER’S RECOMMENDED, REVENUE REQUIREMENTS A utility’s revenue requirement is the amount of money required to fund the utility’s operations, including all expenses, depreciation and the opportunity to earn a fair rate of return. The table below compares (i) PNM’s existing fuel and non-fuel revenues; (ii) PNM’s requested Test Period fuel and non-fuel revenues; and (iii) the Hearing Examiner’s Recommended Decision Case No. 15-00261-UT 12 recommended Test Period fuel and non-fuel revenues. It excludes revenues collected by PNM through its Renewable Energy Rider. Existing Non-Fuel Revenues at Current Rates Proposed Test Period Non-Fuel Revenue Requirement Proposed Increase in Test Period Non-Fuel Revenue Requirement Existing Fuel Revenues Proposed Test Period Fuel Revenue Requirement Proposed Decrease in Fuel Revenues from Current Total Proposed Test Period Revenues/Revenue Requirement PNM $642,095,920 Hearing Examiner $642,095,920 $763,582,007 $683,370,135 $121,486,088 = 18.9% $41,274,215 = 6.4% $215,861,263 $174,786,171 $215,861,263 $155,630,458 ($41,075,092) = (19%) ($60,230,805) = (27%) $938,368,178 $839,000,593 The difference in the amount in the “Decrease in Fuel Revenues from Current” row between PNM and the Hearing Examiner results from the Hearing Examiner’s recommendation that recovery of PNM’s procurement costs related to the New Mexico Wind Energy Center be moved from PNM’s fuel and purchased power cost adjustment clause to PNM’s Renewable Energy Rider. See § XIV(A). Attachment A to this Recommended Decision shows the dollar and percentage impact of each of the Hearing Examiner’s recommended changes to PNM’s proposed cost of service. Attachment B to this Recommended Decision shows the Test Period Cost of Service under the Hearing Examiner’s recommendations. For comparison purposes, PNM’s proposed Test Period Cost of Service is Exhibit HEM-3, COS TEST, pages 1-19 to Mr. Monroy’s Direct Testimony. Recommended Decision Case No. 15-00261-UT 13 Staff, ABCWUA, NMIEC, and the AG recommended particular revenue requirements. However, it would be confusing and misleading to compare Staff and Intervenor proposed Test Period revenues to those of PNM and the Hearing Examiner because Staff and Intervenor recommended amounts are based on differing PNM proposed revenue requirements. For example, some Intervenors’ recommended revenue requirements include fuel costs and others do not. Some incorporate actual, known fuel savings from PNM’s new San Juan Coal Contract and others do not. Very generally, Staff, ABCWUA, NMIEC and the AG recommend smaller non-fuel revenue increases than PNM requests. Walmart does not recommend a particular revenue requirement, but urges the PRC to consider customer impact when setting a revenue requirement. If the PRC reduces PNM’s requested revenue requirement, Walmart recommends that 50% of the amount of the decrease be allocated to customer classes that bear a subsidy burden and 50% of the amount be allocated equally on a percentage basis among all customer classes. Chriss Direct 3, 5. IV. LEGAL STANDARDS FOR RATEMAKING6 Under the PUA, the PRC has “general and exclusive power and jurisdiction to regulate and supervise every public utility in respect to its rates and service regulation and in respect to its securities . . . .” The PUA requires that public utility rates be just and reasonable. Section 628-1 offers no guidance to the Commission for achieving this goal, nor does it specify procedures.” However, another section of the PUA does establish “very specific” procedures for setting rates, including requiring utilities to bear the burden of proof to show that an increase in rates is just and reasonable. An applicant for a rate increase must support its request with substantial evidence. ͸ This Section is taken from the Corrected Recommended Decision of the Hearing Examiner in the 2007 SPS Rate Case. This Section omits citations supporting the discussion, which are included in the discussion in the 2007 SPS Rate Case Corrected Recommended Decision. Recommended Decision Case No. 15-00261-UT 14 To set a just and reasonable rate, the Commission must balance the investor’s interest against the ratepayer’s interest.” As the Supreme Court has concluded, “Neither [interest] is paramount . . . . we cannot focus solely on investor interests.” In reaching its ultimate decision, the Commission is not tied down either by the PUA or by case law to considering only a single factor in establishing rates. Instead, “[t]he Commission is vested with considerable discretion in determining whether a rate to be received and charged is just and reasonable.” The Supreme Court has “consistently construed the [Public Utility Act] broadly rather than to limit the Commission to any one particular method [of ratemaking]; the touchstone is the reasonableness of the ultimate decision.” When setting rates, it is the end result reached, not the method employed, which is controlling. The Commission is not bound by the opinions of experts so long as its decision is supported by sufficient evidence. The Supreme Court has stated that a regulatory commission has “discretion on public policy issues involved with regard to apportionment” of rates, and that “determining the level of subsidies, if any, is a Commission function.” Correlative to the “end result” test in determining whether a utility regulatory commission’s decision was just and reasonable is the “zone of reasonableness” test adopted by the New Mexico Supreme Court and the federal courts after the announcement of that test in Federal Power Commission v. Natural Gas Pipeline Co. As stated by the New Mexico Supreme Court, “There is a zone of reasonableness between confiscation and extortion in which the Commission’s jurisdiction to make rates should be confined.” Justice Frost later concluded that there “is a significant zone of reasonableness, then, between utility confiscation and ratepayer extortion.” As explained by Judge Bazelon of the D.C. Circuit: So long as the public interest—i.e., that of investors and consumers—is safeguarded, it seems that the Commission may formulate its own standards. But there are limits inherent in the statutory mandate that rates be ‘reasonable, just, and nondiscriminatory.’ Among those are the minimal requirements for Recommended Decision Case No. 15-00261-UT 15 protection of investors outlined in the Hope case. And from the earliest cases, the end of public utility regulation has been recognized to be protection of consumers from exorbitant rates. Thus, there is a zone of reasonableness within which rates may properly fall. It is bounded at one end by the investor interest against confiscation and at the other by the consumer interest against exorbitant rates. “Ultimately, the Commission must ensure that rates are neither unreasonably high so as to unjustly burden ratepayers with excessive rates nor unreasonably low so as to constitute a taking of property without just compensation or a violation of due process by preventing the utility from earning a reasonable rate of return on its investment.” In this case, the Commission follows these principles to fairly balance the interests of ratepayers and investors and to set just and reasonable rates in a zone of reasonableness. V. STANDARD OF PROOF The standard of proof in administrative adjudications is, unless expressly provided otherwise, the preponderance of the evidence. Case No. 12-00131-UT, Recommended Decision 16 (11-7-12), adopted in relevant part by Final Order (12-11-12). Preponderance of the evidence means the greater weight of the evidence. Campbell v. Campbell, 1957-NMSC-001, ¶ 24, 62 N.M. 330. It is evidence that, when weighed with that opposed to it, has more convincing force. It has superior evidentiary weight that, though not sufficient to free the mind wholly from all reasonable doubt, is still sufficient to incline a fair and impartial mind to one side of the issue rather than the other. Black’s Law Dictionary 547 (2nd pocket ed. 2001). VI. TEST PERIOD/CWIP/YEAR-END OR 13-MONTH AVERAGE RATE BASE/ESCALATION FACTORS PNM’s selected Test Period and its requests for approvals of Construction Work in Progress (CWIP), a year-end rate base and escalation factors are discussed together because they are interrelated: they all relate to at what period in time a utility’s investments and costs are to be measured. Recommended Decision Case No. 15-00261-UT 16 A. TEST PERIOD Before its amendment in 2009, the PUA did not address use of a historical Test Period or future Test Period. ,Q WKH /HJLVODWXUH SDVVHG DQG WKH *RYHUQRU VLJQHG LQWR ODZ 6HQDWH %LOO ZKLFK added the following provision to the PUA: The Commission shall set rates based on a Test Period that the commission determines best reflects the conditions to be experienced during the period when the rates determined by the commission take effect. If a future Test Period is proposed, the commission shall give due consideration that the future Test Period may best reflect those conditions. NMSA 1978, § 62-6-14(D). Senate Bill 477 defined “future Test Period” as “a twelve-month period beginning no later than the date a proposed rate change is expected to take effect.” NMSA 1978, † 3 6HQDWH %LOO DV DPHQGHG DOVR DGGHG WKH IROORZLQJ 6XEVHFWLRQV ' DQG ( WR † ' 7KH FRPPLVVLRQ VKDOO VHW UDWHV EDVHG RQ D 7HVW 3HULRG WKDW WKH FRPPLVVLRQ GHWHUPLQHV EHVW UHIOHFWV WKH FRQGLWLRQV WR EH H[SHULHQFHG GXULQJ WKH SHULRG ZKHQ WKH UDWHV GHWHUPLQHG E\ WKH FRPPLVVLRQ WDNH HIIHFW ,I D IXWXUH 7HVW 3HULRG LV SURSRVHG WKH FRPPLVVLRQ VKDOO JLYH GXH FRQVLGHUDWLRQ WKDW WKH IXWXUH 7HVW 3HULRG PD\ EHVW UHIOHFW WKRVH FRQGLWLRQV ( 8SRQ D UHTXHVW WR LQFOXGH FRQVWUXFWLRQ ZRUN LQ SURJUHVV LQ WKH UDWH EDVH WKH FRPPLVVLRQ VKDOO JUDQW WKH UHTXHVW RQO\ XSRQ D ILQGLQJ WKDW D SURMHFW¶V FRVWV DUH UHDVRQDEOH 7KH FRPPLVVLRQ VKDOO QRW LQFOXGH WKH DVVRFLDWHG DOORZDQFH IRU IXQGV XVHG GXULQJ FRQVWUXFWLRQ LQ LQFRPH 7KH SURMHFWV IRU ZKLFK WKH FRPPLVVLRQ VKDOO JUDQW D UHTXHVW LQFOXGH HQYLURQPHQWDO LPSURYHPHQW SURMHFWV DQG JHQHUDWLRQ DQG WUDQVPLVVLRQ LQYHVWPHQWV IRU ZKLFK WKH XWLOLW\ KDV REWDLQHG D FHUWLILFDWH RI SXEOLF FRQYHQLHQFH DQG QHFHVVLW\ SURYLGHG WKDW WKH SURMHFWV DUH DQWLFLSDWHG WR EH LQ VHUYLFH QR ODWHU WKDQ ILYH PRQWKV DIWHU WKH HQG RI D XWLOLW\¶V 7HVW 3HULRG EXW LQ QR HYHQW ODWHU WKDQ WZHQW\ IRXU PRQWKV DIWHU WKH ILOLQJ GDWH RI D XWLOLW\¶V UDWH SURFHHGLQJ PNM’s revenue requirement is based on a Future Test Period of October 1, 2015, through September 30, 2016. PNM’s Base Period begins April 1, 2014 and ending March 31, 2015. No Intervenor or Staff challenges PNM’s Test Period selection, and the PRC’s approval of that selection is not precedent. See Piedra, Inc. v. State of New Mexico, 2008-NMCA-089, ¶ 32, 144 Recommended Decision Case No. 15-00261-UT 17 N.M. 382 (rejecting claim that cited case controlled on issue of PRC’s authority to abandon and convey roads and highways where issue was neither analyzed nor decided). B. RATE BASE Between June 30, 2010 — the rate base cutoff date used in PNM’s Illustrative Cost of Service in Case No. 10-00086-UT — and the end of the Adjusted Base Period — March 31, 2015 — PNM increased its rate base by about $265 million. The projected net increase to rate base from March 31, 2015, through September 30, 2016, is $390,815,445. Ortiz Direct 5; Sch. A-4. The projected net increase to rate base from June 30, 2010, and projected through September 30, 2016, including CWIP, is $655 million. PNM has or expects to place 460 capital projects in service between March 31, 2015, and February 28, 2017. PNM’s Initial Posthearing Brief 1. PNM refers to the period April 1, 2015, through September 30, 2016, as the “Capital Investment Period.” April 1, 2015 through September 30, 2015 is the “Linkage Period,” which is the period of time between the end of the Base Period and the beginning of the Test Period. PNM refers to the period April 1, 2015 through February 28, 2017, as the “Capital Clearing Period.” It refers to October 1, 2016 through February 28, 2016 as the “CWIP Period.” Johnson Direct 1-2. PNM’s Generation capital investments during the Capital Investment Period are about $496 million. The investments fall into three major categories: 1. Capital projects for new generation plant (La Luz and 40 MW Solar) are $133.1 million; 2. Capital projects at existing generation plants are $199.4 million; and 3. Capital to purchase 64 MW of the PV2 leases is $163.5 million. Olson Direct 20. PNM’s Transmission and Distribution (T&D) capital investments during the Capital Investment Period Test Period are $247,653,475. Johnson Direct 30-31. Recommended Decision Case No. 15-00261-UT 18 PNM requests approval of a year-end rate base, which means that its rate base is based on net plant in-service balances as of the end of the Test Period, or September 30, 2016, with minor exceptions.7 PNM’s proposed rate base is $2,458,087,084. Sch. A-4 An alternative approach, advocated by ABCWUA and NMIEC, is to calculate the average of the monthly net plant-in-service balances during the Test Period based on a 13-month convention. Ditter Direct 8-11; Gorman Direct 70-74. The 13-month convention uses the 13month average of the future Test Period. 17.1.3.16(C)(1) NMAC. The difference between these approaches is significant. Use of a 13-month average rate base reduces PNM’s requested revenue requirement by about $82.7 million. Monroy Rebuttal 8. Accounting for Public Utilities explains: When using forecasted data, most often the rate base investment is averaged over a 12-month period and related to the results of operations for the period. This averaging concept produces a matching of the rate base investment with the revenues generated by the investment and the costs incurred in the process. Accounting for Pub. Utils, § 7.04.8 Using a year-end rate base assumes, for ratemaking purposes, that all investments are made at the beginning of the Test Period. Under the well-recognized matching principle, revenues, expenses and investment must be measured on a comparable basis. Therefore, “If a year-end rate base is utilized, the company’s test-year revenues and the associated variable operating expenses should be adjusted to reflect the year-end number of customers and the trend in customer usage . . . at year-end or expected in the near future.” Columbia Gas, Case No. 7545, 1981 WL 721478, under “Rate Base” (MD PSC 11-9-1981). 7 The exceptions relate to the working capital accounts included in rate base, which are based on 13month average balances. Monroy Direct 6. 8 Except in extraordinary circumstances, FERC requires an averaging method to estimate plant in service Test Period plant balances. It infrequently approves end of year plant in service balances as reasonable. In re Pub. Serv. Co. of Indiana, 19 P.U.R.4th 150 (F.P.C. 1977), 57 F.P.C. 1173, 1186-87. Recommended Decision Case No. 15-00261-UT 19 ABCWUA witness Dittmer argues against PNM’s use of a year-end rate base because PNM did not make annualizing adjustments to revenues and expenses to reflect the year-end rate base. Mr. Dittmer advocates that PNM’s rate base be calculated using a 13-month average of Plant in Service, Accumulated Depreciation, and plant-related Accumulated Deferred Income Tax balances. Dittmer Direct 9. PNM admits that it did not make annualizing adjustments, but defends this lack of synchronization on the ground that doing so would have increased its revenue requirement. In support of this assertion, PNM says that annualizing its depreciation expense would increase its depreciation expense by more than $3 million, while annualizing its sales to reflect an estimated year-end customer count would only increase test-year revenues by $1 million. Monroy Rebuttal 10. PNM’s late identification, in Rebuttal Testimony, of the result of annualizing a single expense against a single billing determinant, falls short. A utility proposing a year-end rate base should make a credible attempt to identify expense savings and revenue producing effects for major plant additions. As the Idaho Public Utilities Commission explained: It is unfair to ratepayers to assume that the investment in these plants will not increase Company revenues or decrease Company expenses in the future. Further, it is unreasonable to expect the Commission to allow full recovery of plant investment as if the plant had been in operation the full year without a corresponding adjustment to revenues and expenses. Application of Idaho Power, 233 P.U.R.4th 107, under “Adjustments to Rate Base, Annualizing Plant Adjustments” (Idaho P.U.C. 5-25-04). Relatedly, PNM has not shown that it has invested substantially in non-revenue producing plant, which can justify using a year-end rate base. E.g, Columbia Gas, Case No. 8687, Order No. 72213, 1995 WL 848257, § III(A) (MD PSC 10-6-95). Lastly, but not unimportantly, PNM objected to the AG’s request that PNM update its actual plant balances through January 31, 2016. Monroy Rebuttal 21 (“PNM does not agree that the Company should be obligated to update its future Test Period cost of service in this regard.”) Recommended Decision Case No. 15-00261-UT 20 Verification of PNM”s projected plant balances with actual data further into PNM’s future Test Period could have supported using a year-end rate base. Using a 13-month average rate base properly matches Test Period revenues and expenses that occur throughout the Test Period with the level of rate base also occurring throughout the Test Period. PNM’s rate base should be calculated using a 13-month average of Plant in Service, Accumulated Depreciation, and plant-related Accumulated Deferred Income Tax balances. Mr. Dittmer said that because he advocates using an average year rate base, then a matching adjustment should be made and lease payments that PNM made during the Test Period before it purchased the 64.1 MW of PV2 should be included in the cost of service. Dittmer Direct 15-16. The lease payments made on the 64.1 MW before it was purchased should be removed in calculating the Test Period cost of service because termination of those leases is a known and measurable change — PNM has purchased those leases and no longer makes the lease payments.9 Removal of these lease payments is an attritional adjustment. Attritional adjustments may be made to Test Period data for known and measurable changes in events or conditions that will affect future cost of revenue levels. Attritional adjustments include annualization adjustments to eliminate from the Test Period conditions that will that no longer exist. Accounting for Pub. Utils., § 7.05. C. CWIP Carrying charges of debt interest and reasonable equity return on construction capital may be recovered through rates by including CWIP in rate base. A utility recovers its carrying charges currently from ratepayers through the return component of its rates rather than adding them to the cost of construction for recovery when the plant is in service. The return on CWIP is recorded as income on a current basis, and actual cash payments are made by the ratepayers 9 See § XV. Recommended Decision Case No. 15-00261-UT 21 currently. Including CWIP in rate base compensates utility for its advance commitment of capital. Bonbright 246-47. PNM includes in rate base the CWIP balance as of September 30, 2016, for any project that has an estimated clearing from October 2016 through February 2017, but excluding revenue-producing CWIP.10 PNM includes the lesser of the CWIP balance as of September 30, 2016, or the estimated clearings for the capital project through February 2017, in the amount of $78.2 million. Monroy Direct 32; Buchanan Direct 8. PNM witness Olson identified two environmental projects included in PNM’s CWIP request: Plant Common Evaporation Pond Heightening ($786,371); and Plant Common Shumway Arroyo Slurry Wall and Water Containment ($2,887,153). Tr. (4-18-16) 1564-65; Exh. CMO-2, pp. 2-3 to Olson Direct. ABCWUA opposes PNM’s entire CWIP request, arguing first that “PNM makes no compelling argument as to why its CWIP request is reasonable in this proceeding.” Second, ABCWUA argues that PNM’s reliance on PRC cases approving CWIP requests, issued before the PUA’s amendment in 2009 and possibly influenced by concerns over regulatory lag, “are off point and irrelevant” in this case where PNM has used “a fully forecasted Test Period.” Third, ABCWUA argues that post-Test Period adjustments such as CWIP can be viewed as selective, one-sided, and lacking consideration of actual and potential offsets. Dittmer Direct 14-15. The AG also opposes PNM’s entire CWIP request because CWIP does not represent facilities that are used and useful and requiring current ratepayers to pay for such plant violates the principle of intergenerational equity. Crane Direct 29-30. ABCWUA’s and the AG’s arguments are not compelling. 10 CWIP is cleared to plant in service at forecasted completion dates. Buchanan Direct 23. Recommended Decision Case No. 15-00261-UT 22 Section ( GRHV QRW UHTXLUH 310 WR VKRZ WKDW LWV &:,3 UHTXHVW LV UHDVRQDEOH EXW WKDW ³D SURMHFW¶V FRVWV DUH UHDVRQDEOH ´ 310 ZLWQHVVHV 2OVRQ -RKQVRQ DQG 0HQGH] GLVFXVVHG WKH FDSLWDO SURMHFWV LQFOXGHG LQ 310¶V &:,3 UHTXHVW DQG VSRQVRUHG EXGJHW GRFXPHQWV IRU WKHVH SURMHFWV 1R ,QWHUYHQRU RU 6WDII FKDOOHQJHG WKH UHDVRQDEOHQHVV RI WKH SURMHFWV¶ FRVWV 0RQUR\ 5HEXWWDO 6HFRQG WKH /HJLVODWXUH E\ DGGLQJ ODQJXDJH WR WKH 38$ DW WKH VDPH WLPH WKDW LW IRU WKH ILUVW WLPH DGGHG D GHILQLWLRQ RI IXWXUH 7HVW 3HULRG LQWHQGHG WKDW &:,3 EH DOORZHG ZKHQ D XWLOLW\ XVHV D IXWXUH 7HVW 3HULRG 7KH /HJLVODWXUH DGGHG &:,3 WR WKH OLVW RI LWHPV WKDW WKH 35& ³VKDOO JLYH GXH FRQVLGHUDWLRQ WR´ ZKHQ IL[LQJ D XWLOLW\¶V UDWHV 106$ † $ 7KLUG &:,3 KDV DOZD\V EHHQ VXEMHFW WR EHLQJ YLHZHG DV RQH VLGHG DQG FUHDWLQJ LQWHUJHQHUDWLRQDO LQHTXLW\ DQG WKH 35& KDV DSSURYHG LWV LQFOXVLRQ LQ UDWH EDVH RYHU WKHVH REMHFWLRQV &DVH 1R 87 5HFRPPHQGHG 'HFLVLRQ 310¶V UHTXHVW IRU &:,3 VKRXOG EH DSSURYHG D. AG’S PROPOSED 30% ADJUSTMENT TO RATE BASE 310¶V UDWH EDVH LQFOXGHV PLOOLRQ LQ QHW SODQW LQ VHUYLFH DGGLWLRQV IURP WKH HQG RI WKH %DVH 3HULRG WKURXJK WKH HQG RI WKH 7HVW 3HULRG 310¶V VXSSRUWLQJ WHVWLPRQLHV VKRZHG DFWXDO FRVWV WKURXJK WKH HQG RI WKH %DVH 3HULRG ² 0DUFK ² DQG PRQWKO\ SURMHFWLRQV WKURXJK 6HSWHPEHU ,Q D GLVFRYHU\ UHVSRQVH 310 SURYLGHG DFWXDO PRQWKO\ SODQW DGGLWLRQV WKURXJK 1RYHPEHU 7KURXJK 1RYHPEHU 310¶V DFWXDO SODQW DGGLWLRQV ZHUH RI LWV IRUHFDVWHG SODQW DGGLWLRQV &UDQH 'LUHFW 11 11 PNM provided gross plant values through December 2015, but not accumulated depreciation and ADIT associated with such plant through December 2015. Tr. (4-27-16) 3033 (Crane). The SCNR and balanced draft capital projects were complete at the time that PNM filed its Application. Tr. (4-27-16) 3034-35 (Crane). The AG has no concerns about cost overruns for these projects. Tr. (4-27-16) 3034-35 (Crane). Recommended Decision Case No. 15-00261-UT 23 %HFDXVH RI WKLV GLVFUHSDQF\ EHWZHHQ DFWXDO DQG SURMHFWHG SODQW DGGLWLRQV WKH $* UHFRPPHQGV UHGXFLQJ 310¶V 7HVW 3HULRG UDWH EDVH E\ DSSO\LQJ D UHGXFWLRQ WR WKH FKDQJH LQ QHW WUDQVPLVVLRQ DQG GLVWULEXWLRQ SODQW IURP WKH %DVH 3HULRG WR WKH HQG RI WKH 7HVW 3HULRG 7KH $* GRHV QRW UHFRPPHQG D VLPLODU UHGXFWLRQ IRU JHQHUDWLRQ SODQW &UDQH 'LUHFW The AG’s recommendation should be rejected because it is not sufficiently cost-based. See Attorney General v. PRC, 2011-NMSC-034, ¶¶ 12, 18 (rates under PUA must be cost-based). $GGLWLRQDOO\ 310¶V UHIXVDO WR SURYLGH XSGDWHG SODQW EDODQFHV LV D IDFWRU LQ UHMHFWLQJ 310¶V \HDU HQG UDWH EDVH E. ESCALATION FACTORS 1. 1.5% ESCALATION FACTOR APPLICABLE TO NON-LABOR O&M EXPENSES For a number of O&M expenses, PNM’s Test Period amounts are based on componentspecific information. Examples of O&M expenses that were developed considering specific facts include: x O&M labor, based on actual or expected union contract changes x Self-insured medical and dental expense for active and retired employees, based on actuarial studies by Towers Watson; and x Urea, based on commodity cost and expected use data. Dittmer 17. For Test Period O&M expense amounts not based on expense-specific information, PNM escalated Adjusted Base Period amounts by 1.5% for one annual period. Monroy Direct 68; Tr. (4-15-16) 1308, 1317 (Monroy). PNM’s Base Period is April 1, 2014 to March 31, 2015. Monroy Direct 6. The Adjusted Base Period is the Base Period adjusted for annualizations, normalizations and known and measurable changes and regulatory requirements that occur within the Base Period. 17.1.3.7(A) NMAC. PNM witness Monroy explained that PNM chose a 1.5% escalator because it reflects a conservative estimate of the increase of O&M expenses between the Base Period and Test Period Recommended Decision Case No. 15-00261-UT 24 based on the trend indicated by the consumer price index (CPI), which measures the change over time in the prices of consumer goods. Monroy Direct 69. According to Mr. Monroy, between the end of the Base Period —March 31, 2015 — and through March 31, 2016, the CPI rose more than 1.5%. Tr. (4-15-16) 1280 (Monroy). ABCWUA opposes PNM’s application of the 1.5% escalation factor, arguing that PNM has not justified using it and that using it is not reasonable. Dittmer Direct 17. ABCWUA relies on PNM’s 2014 and 2015 Annual Operating Plans. PNM’s 2014 Annual Operating Plan assumed a (0.5%) non-labor O&M expense escalation factor. PNM’s 2015 Annual Operating Plan assumed a 0% non-labor O&M expense escalation factor. ABCWUA also relies on a goal of PNM’s 2015 Business Unit Annual Incentive Plan, which is to achieve actual O&M of 1% below the O&M level in the Annual Operating Plan. Dittmer Direct 18; Exhs. 4 & 8 to Dittmer Direct. ABCWUA dismisses PNM’s reliance on its 2016 Annual Operating Plan, issued in December 2015, which assumes a 1.5% non-labor O&M expense escalation factor, because it is inconsistent with PNM’s 2014 and 2015 expectations and goals. Also, ABCWUA witness Dittmer says it is reasonable to at least question PNM’s assumption of a 1.5% expense escalation rate “that precisely matches PNM’s rate case assumption” in a document issued just before the deadline for filing Staff/Intervenor testimony. Dittmer Direct 20. Staff also opposes PNM’s application of the 1.5% escalation factor and recommends applying no escalation factor to reflect actual month-to-month changes in the CPI in 2015. For the following months in 2015, the changes in the CPI from the preceding month were: (.07%) in January 2015; 0.4% in May 2015; and 0.2% in October 2015. De Cesare Direct 11 & Exh. VDC-2. PNM’s application of a 1.5% escalation factor to non-labor O&M expenses should be approved. It is justified by the actual increase in the CPI between the end of the Base Period and the mid-Test Period, which is greater than 1.5%. See PNM Gas Servs., 2000-NMSC-012, ¶ 87, Recommended Decision Case No. 15-00261-UT 25 129 N.M. 1 (“[C]ommon sense requires that the latest available economic information should be utilized in order to insure that the projected figures bear a meaningful relation to future as well as past and present fiscal realities.”). 2. 1.5% ESCALATION FACTOR APPLICABLE TO PLANNED OUTAGE EXPENSES For Test Period planned outage expenses, PNM used a six-year inflation-adjusted average for the years 2009-2014, because planned outage expenses can vary significantly from year-to-year. Monroy Direct 74. For example, if the scheduled production expense at a production plant was $1,000 in 2009 — the first year of the six-year period — PNM compounded the $1,000 using a 1.5% escalation rate for six annual periods to derive a $1,093 amount used in calculating the six-year average. Dittmer Direct 41-42 & n.11. ABCWUA recommends using a 0.5% escalation factor rather than a 1.5% escalation factor in calculating the six-year average. In support its rejection of a 1.5% escalation factor, ABCWUA first says that PNM’s 2014 and 2015 Annual Operating Plans, which assumed flat and negative escalation rates non-labor O&M expenses, would apply to non-labor scheduled production maintenance expenses for four plants owned or partially owned and operated by PNM. Second, ABCWUA relies on a 2011 preliminary study that projected non-fuel O&M savings from installation of balanced draft at San Juan Units 1 and 4. Third, ABCWUA relies on PNM’s PV Business Plan, which shows that PV non-fuel expenses have been relatively flat in years 2011 through 2014 and were projected to be relatively flat in 2015 and 2016. Dittmer Direct 42-45. The basis for ABCWUA’s recommended 0.5% escalation factor is two-pronged: first, Mr. Dittmer did not review PNM documents that might indicate projected inflation factors for the first two years of the six-year average period, so he lacked evidence to conclude that assuming a 0% escalation rate over all six years is reasonable. Second, Mr. Dittmer said that he doesn’t “take significant exception to employment of an assumed 1.5 percent escalation rate to be Recommended Decision Case No. 15-00261-UT 26 applied to the two first years of data,” which equates to using a 0.5% escalation rate for six annual periods, which he uses for expediency. Dittmer Direct 46. The AG recommends eliminating the 1.5% escalation factor, arguing that the purpose of using a six-year average is to account for year-to-year fluctuations and that any increases related to inflation are already incorporated in the actual annual costs. The AG also argues that PNM has not demonstrated that outage costs are impacted by inflationary trends. Crane Direct 71. PNM’s use of a six-year inflation-adjusted average for the years 2009-2014 to determine Test Period planned outage expenses should be approved. It is not appropriate to assume that the outage expense going back to 2009 will cost the same in the Test Period in today’s dollars. The escalation factor appropriately reflects the time value of money, is supported by annual CPI increases, and is not speculative. Cf. Gas Co. of New Mexico, 35 P.U.R.4th 106, 129 (N.M.P.S.C. 1980) (denying attrition factor that was “purely speculative;” approval of year-end test adjusted designed to consider known and actual effects of inflation). Monroy Rebuttal 51-52 & Exh. HEM-5. VII. REVENUE REQUIREMENT ALLOCATION FOR FERC WHOLESALE A. BACKGROUND PNM develops generation demand, generation energy and transmission demand allocation factors to allocate its revenue requirement between PNM retail and FERC wholesale generation jurisdictions. Chan Direct 13-14. In developing the revenue requirement allocation for FERC wholesale generation in the Test Period, PNM considered current and projected loads of its FERC wholesale generation customers, including: Navopache Electric Cooperative (“NEC”); the City of Gallup; the City of Aztec; and the Jicarilla Apache Nation. Recommended Decision Case No. 15-00261-UT 27 The impact of PNM excluding wholesale billing determinants from its revenue requirement calculation is that more of PNM’s revenue requirement is recovered from retail customers through rates approved in this case. ABCWUA’s Initial Posthearing Brief 30. B. CITY OF AZTEC PNM participated in the City of Aztec’s Request for Proposal process but was not selected to be the future supplier for the City of Aztec. As such, PNM’s wholesale contract with the City of Aztec expired on June 30, 2016, in advance of the end of Test Period on September 30, 2016. PNM reflected expiration of the City of Aztec contract by removing all of the City of Aztec’s loads when it developed jurisdictional demand and energy allocation factors. Monroy Direct 11-12. In other words, PNM excluded the City of Aztec’s loads in the calculation of its demand and energy allocators for the Test Period to reflect the fact that the wholesale contract with the City of Aztec terminated on June 30, 2016, and therefore the City would not be a customer during any part of the rate effective period. Monroy Rebuttal 71. ABCWUA Witness Dittmer argued that the City of Aztec load should be reflected in the Test Period revenue requirement allocators in the months during the Test Period that the City received PNM service — October 1, 2015 through June 30, 2016. Dittmer Direct 35. The City of Aztec load should be removed in calculating the Test Period jurisdictional allocators because removal of the City’s load is a known and measurable change — the City is no longer a PNM wholesale customer. Removal of the City’s load is an attritional adjustment. Attritional adjustments may be made to Test Period data for known and measurable changes in events or conditions that will affect future cost of revenue levels. Attritional adjustments include annualization adjustments to eliminate from the Test Period conditions that will that no longer exist. Accounting for Pub. Utils., § 7.05. Mr. Dittmer said that because he advocates using an average year rate base, then a matching adjustment should be made and the City of Aztec load should be reflected in PNM’s Test Period for the time during the Test Period that the Recommended Decision Case No. 15-00261-UT 28 City was a PNM wholesale customer. However, Mr. Dittmer knew of no New Mexico precedent that links an attritional adjustment to use of an average or year-end rate base. Tr. (4-25-16) 2769-70. C. THE JICARILLA APACHE NATION The Jicarilla Apache Nation (“JAN”) filed public comments with the PRC stating its intent to provide notice of contract termination and to seek alternative providers effective during the future Test Period. Because PNM was unsure whether JAN would continue to be a wholesale customer, PNM removed all of the JAN loads from its calculation of forecasted generation energy and demand allocators. However, to compensate all remaining customers in the event the JAN contract is not terminated, PNM proposed to reflect revenues received from JAN through the FPPCAC for as long as JAN remains a wholesale customer. Monroy Direct 12; Taylor Direct 16. ABCWUA argued that PNM presented insufficient evidence that JAN would cease taking wholesale service from PNM, and recommended that the demand and energy allocators reflect that JAN remains a wholesale customer for the Test Period. Dittmer Direct 35. The JAN load should be included in PNM’s calculation of forecasted generation energy and demand allocators because it is not known that JAN will cease receiving service from PNM. “A known and measurable change must be known with reasonable certainty and capable of being objectively quantified.” Case No. 2262, Recommended Decision (3-8-90); see also Accounting for Pub. Utils., § 7.05 (“Many commissions have permitted adjustments for conditions that come into being subsequent to the test year, but only when they are known with an almost absolute finality[.]”). VIII. PNM’S TEST PERIOD CAPITAL STRUCTURE Recommended Decision Case No. 15-00261-UT 29 Capital structure is the relationship between a company’s debt and equity. It influences overall cost of capital because capital is more expensive than debt. PNM’s proposed capital structure is 50% long term debt, 0.39% preferred stock, and 49.61% common equity. Hevert Direct 72. This proposed capital structure is based on PNM’s projected capital structure at the end of the Test Period, reflecting projected debt issuances and refinancing expected to occur in the Test Period. Eden Direct 20. No Intervenors oppose this capital structure. Staff, however, proposes a different capital structure of 53% long-term debt, 0.39% preferred stock, and 46.61% common equity. Staff witness Pitts based her recommendation on three sources of data: 1. 2. 3. PNM’s planned capital improvements Various PNM financial filings, including its 2015 SEC Form 10-K PNM’s 2015 Annual Informational Financing Filing (AIFF) Pitts Direct 23-24. Staff particularly relies on a statement in a Moody’s Credit Opinion of PNM, dated June 23, 2015, that says that as of March 31, 2015, PNM’s debt to total capitalization ratio was about 54%, and PNM could borrow up to $100 million from its parent. Staff’s Initial Posthearing Brief 30; Exh. EAE-3, p.4, to Eden Direct. Staff argues that its proposed capital structure is more beneficial to ratepayers because financing capital projects through debt rather than equity is less expensive. Pitts Direct 31. Staff also argues that it is disingenuous on PNM’s part to exclude term loan debt from its capital structure merely because it is classified as shortterm debt under the PUA. Staff’s Initial Posthearing Brief 31. Staff’s recommendation is opposed by PNM. Dr. Pitts made a similar argument in the 2015 EPE Rate Case in support of reducing the equity portion of EPE’s proposed capital structure. In that case, like in this case, Dr. Pitts relied on financial data to support Staff’s position. The PRC rejected Staff’s argument, stating: The information that the Commission has previously relied upon has been the actual percentages of debt and equity for regulatory purposes at the end of the Test Recommended Decision Case No. 15-00261-UT 30 Period. Based upon the record, the Hearing Examiner finds no compelling evidence to recommend that the Commission change its policy as to this issue. 2015 EPE Rate Case, Recommended Decision 64-70. In this case, as in the 2015 EPE Rate Case, Staff presents no compelling evidence that the PRC change its policy, and Staff’s argument should be rejected here as well. IX. PNM’S TEST PERIOD COST OF DEBT PNM’s proposed Test Period 5.87% cost of debt is unopposed and should be approved. Eden Direct 22. X. PNM’S TEST PERIOD COST OF PREFERRED STOCK PNM’s proposed Test Period 4.62% cost of preferred stock is unopposed and should be approved. Eden Direct 22. XI. RETURN ON EQUITY A. POSITIONS OF PARTIES AND STAFF The table below shows the recommended ROEs among those parties and Staff who recommend an ROE: Party PNM NMIEC ABCWUA Staff AG CFRE NEE Witness Robert Hevert Michael Gorman Did not call witness on ROE. Recommends NMIEC’s position Heidi Pitts Randall Woolridge Did not call witness on ROE. Recommends the AG’s position Did not call witness on ROE. Recommends the AG’s position PNM-41 at 21:8-11. Gorman Direct 3. 14 Dittmer Direct 53-54. 15 Pitts Direct 59. 16 Woolridge Direct 5. 17 CFRE’s Posthearing Response Brief 20. 18 NEE’s Initial Posthearing Brief 41. 12 13 Recommended Decision Case No. 15-00261-UT 31 Proposed ROE 10.5%12 9.35%13 9.35%14 9.1%15 9.0%16 9.0%17 9.0%18 Walmart does not recommend a particular ROE, but its witness expressed concern that PNM’s proposed ROE is excessive, especially in light of: 1. The customer impact of PNM’s proposed revenue requirement 2. PNM’s use of a future Test Period 3. PNM’s proposed decoupling mechanism 4. ROEs recently approved by this Commission and other state commissions. Chriss Direct 9. In PNM’s last rate case, the PRC authorized a 10% ROE. PNM’s return on equity in 2014 was 7.24%. Ortiz Direct 7. B. 2015 EPE RATE CASE The legal standards for determining the ROE and a description of the models used for determining the ROE have been extensively discussed in PRC cases. The legal standards and models have not materially changed, if at all, and it is not necessary to repeat them here. Those looking for these discussions are directed to the 2007 SPS Rate Case, Recommended Decision 22-73, and the 2015 EPE Rate Case, Recommended Decision 36-61. Very recently, on June 8, 2016, in the 2015 EPE Rate Case, the PRC approved a 9.48% ROE for El Paso Electric Company (EPE). Due particularly to the currentness of that decision, the policies adopted by the PRC in that case should be followed in this case absent sufficient justification for departure. Cf. 2007 PNM Electric Rate Case, Recommended Decision 59 (stating that “the Commission’s decision in the PNM Gas rate case steers this one with respect to the determination of PNM’s return on equity.”); 2007 SPS Rate Case, Recommended Decision 57 (giving due consideration to using a two-stage DCF model but finding no compelling reason to depart from using Constant Growth DCF model). In the 2015 EPE Rate Case, x EPE sought approval of a 9.95% ROE x The AG recommended a 8.95% ROE x Staff recommended a 9.225% ROE. Recommended Decision Case No. 15-00261-UT 32 The Hearing Examiner found that the evidence showed a “zone of reasonableness” between 9.225% and 9.995%, and she recommended a 9.6% ROE. Recommended Decision 61. The PRC approved a 9.48% ROE. 2015 EPE Rate Case 33, ¶ 68. While the PRC said it is not bound to use a single method, it relied on the DCF Model, saying it has “‘relied primarily on the Constant Growth DCF model to determine utilities’ ROEs.’” Id. at 33, ¶ 68 (quoting 2012 SPS Rate Case, Recommended Decision 64). Without discussion of alternatives, the PRC followed its decision in the 2012 SPS Rate Case to use a 360-day average period to determine the stock price used to calculate the dividend yield and to use a full year’s dividend growth to determine the dividend yield. Id. at 33, ¶ 68. The PRC said that EPE witness Hadaway most closely followed the Constant Growth DCF Model. However, Dr. Hadaway’s recommendations combined the results of his DCF and CAPM analyses. The PRC observed that it had rejected a similar approach in the 2007 PNM Electric Rate Case that combined the results of DCF and CAPM analyses, and found, based on the evidence, that “it should do the same here, and continue its primary reliance upon the Constant Growth DCF model.” Id. at 34, ¶ 69. The PRC used Dr. Hadaway’s DCF analysis as a starting point. From there, it eliminated from Mr. Hadaway’s analysis three companies that were engaged in merger activity. This change resulted in a “reasonable range of 8.95% to 10.0%.” The PRC said, “The midpoint of that range is 9.48%, which the Commission finds is a reasonable ROE that should be approved in this case.” The PRC noted that the 9.48% ROE was (1) between Staff’s initially recommended 9.225% ROE and the Hearing Examiner’s recommended 9.6% ROE (which Staff later accepted as within the range of reasonableness) and therefore consistent with Staff’s recommendations; and (2) consistent with the AG’s position that the ROE should be less than 9.6%. Id. at 35-36, ¶¶ 72-73. The 360-day average period to determine the stock price in the DCF analysis was April 2014 to March 2015. Id. at 36, ¶ 73 (incorporating data from Exh. SCH-6); SCH-6, p.4. Recommended Decision Case No. 15-00261-UT 33 The PRC said that EPE’s argument that the Hearing Examiner’s recommended 9.6% ROE — and by extension the approved 9.48% ROE — was too low “flie[d] in the face of evidence submitted by its own witness.” The PRC explained that the 9.48% ROE was based entirely on Dr. Hadaway’s DCF analysis but corrected in a manner that even he admitted was appropriate. The PRC rejected EPE’s reliance on the PRC’s approval of a 9.96% ROE for SPS in March 2014 and approval of a 10% ROE for PNM in July 2011, explaining that authorized ROEs for vertically integrated utilities fell from an average annual rate of 10.24% in 2011 to 9.65% in 2015. The average ROE for all such utilities during the third quarter of 2015 — the latest period shown in Dr. Hadaway’s testimony — was 9.4%. Id. at 36-37, ¶ 74. The PRC also rejected EPE’s argument that its ROE should be higher than SPS’s 9.96% ROE because of the increase in interest rates between 2014 and 2015, saying it was undermined by the downward trend in approved ROEs for other utilities during the same period. The PRC said that although increasing interest rates “can under certain circumstances put upward pressure on approved utility ROEs, that clearly has not occurred during the 2014 to 2015 period relied upon by EPE.” Id. at 37, ¶ 75. Lastly, the PRC said that EPE’s argument that it deserved a higher ROE because it faced greater business risks than the average of the proxy companies used by Dr. Hadaway “also misses the mark.” The PRC said that if, as EPE contended, its business risk was greater than those of the companies included in Dr. Hadaway’s proxy group, EPE should have asked Dr. Hadaway to use a proxy group whose size, generation portfolio, and other business characteristics matched those of EPE. Moreover, the PRC said that because EPE made no attempt to quantify the risk premium associated with the greater risks that it allegedly faced, EPE failed to provide any evidentiary support for its claim. Id. at 37-38, ¶ 76. Recommended Decision Case No. 15-00261-UT 34 C. THE CONSTANT GROWTH DCF METHOD SHOULD BE USED TO DETERMINE PNM’S AUTHORIZED ROE The DCF Constant Growth Method should be used to determine PNM’s ROE because (i) just two months ago, the PRC found, based on the evidence, that it should continue its primary reliance upon the Constant Growth DCF Method; and (ii) no party or Staff showed by a preponderance of the evidence that the evidence in this case warrants departure from this method. Therefore, the results of other methods used by witnesses to determine ROE are not discussed in this Recommended Decision. Mr. Hevert’s objection to using the Constant Growth DCF Method is not convincing. Mr. Hevert objects to using this Method because he says that “the period over which my analyses were performed included market data that were highly unusual and inconsistent with that model’s fundamental assumptions.” The data that he says is inconsistent with the Model’s fundamental assumptions are: x His Proxy Group’s average Price/Earnings (P/E) ratio recently has significantly exceeded its long-term average. x His Proxy Group’s P/E multiple exceeded the overall market P/E ratio. Mr. Hevert said that in early 2015, investors put money into the utilities sector because interest rates were low and investors were “reaching for yield.” He said that the utilities sector typically yields about 4% compared to about 2.6% for Treasuries. At that time, the P/E ratios well exceeded the market. In the spring of 2015, when investors believed that interests were rising, the utilities sector was devalued, lost about 10% to 13% of its value, and underperformed compared to the market. In January and February 2016, the utilities sector has overperformed compared to the market. Tr. (4-13-16) 574-76, 597, 605-06 (Hevert). Mr. Hevert said that using the Constant Growth DCF Model assumes that utility companies will continue to trade at a premium to the market in perpetuity — that their P/E ratios will remain the same — even though that is highly improbable. Tr. (4-13-16) 608-10. In light of that data, Mr. Hevert recommends considering the results of his analyses using the Recommended Decision Case No. 15-00261-UT 35 Multi-Stage DCF approach, the Capital Asset Pricing Model (CAPM), and the Bond Yield Plus Risk Premium (Risk Premium) approach and giving less weight to the Constant Growth DCF Method. He says, “To the extent any weight is given to the DCF estimates, the full range of results, in particular, the mean high estimates, should be considered.” Hevert Direct 4-6. Dr. Woolridge persuasively rebutted Mr. Hevert’s argument. Dr. Woolridge agrees that the P/E ratios of utility stocks have increased. However, Dr. Woolridge explained that the higher valuation of utilities is justified because cost recovery mechanisms have reduced the risk of the utility industry which has led to higher P/E multiples. As utilities increasingly secure more up-front assurance for cost recovery in their rate proceedings, we think regulators will increasingly view the sector as less risky. The combination of low capital costs, high equity market valuation multiples (which are better than or on par with the broader market despite the regulated utilities' low risk profile), and a transparent assurance of cost recovery tend to support the case for lower authorized returns, although because utilities will argue they should rise, or at least stay unchanged. Woolridge Direct 77 (quoting Moody’s Investors Service, Lower Authorized Equity Returns Will Not Hurt Near-Term Credit Profiles (3-10-15)). In fact, the PRC has authorized three adjustment clauses for PNM: (1) its FPPCAC; (2) its Renewable Rider; and (3) its Energy Efficiency Rider. Therefore, Mr. Hevert’s suggestion that the Constant Growth DCF results may provide low results due to the relatively high P/E multiples of utilities is incorrect. As indicated by Moody’s, the lower risk of utilities has led to higher valuation levels and P/E multiples. Woolridge Direct 78. D. EXPLANATION OF THE DCF CONSTANT GROWTH METHOD The DCF method estimates an equity return from a proxy group by adding estimated dividend yields to investors’ expected long-term dividend growth rate: Cost of equity = expected dividend yield + expected dividend growth. 2007 SPS Rate Case, Final Order 2, ¶ 7. Recommended Decision Case No. 15-00261-UT 36 The first step in the Constant Growth DCF analysis is to determine the current dividend yield for each of the proxy utilities. “Simply stated, dividend yield is determined by dividing the utility’s dividend by its current stock price.” Id. at 3, ¶ 9. A subject of controversy in determining dividend yield is over what period of time to average stock prices. As stated above, in the 2015 EPE Rate Case, the PRC, without discussion of alternatives, followed its decision in the 2012 SPS Rate Case to use a 360-day average period to determine the stock price. A 360-day average period to determine the stock price should be used in this case because there has been no justification for departure. Another subject of controversy has been how to adjust the dividend yield to reflect projected dividends. An adjustment is appropriate because utility companies tend to increase their quarterly dividends at different times throughout the year, and it is reasonable to assume that dividend increases will be evenly distributed over calendar quarters. An adjustment ensures that the expected dividend yield is, on average, representative of the Test Period and does not overstate dividends to be paid during that time. Hevert Direct 30. As stated above, in the 2015 EPE Rate Case, the PRC, without discussion of alternatives, applied a one-year adjustment to the dividend yield to reflect a projection of dividends for the coming year. A oneyear adjustment to the dividend yield should be used in this case because there has been no justification for departure. Yet another subject of controversy is the projected growth rate to be applied to the calculated dividend yield. 2015 EPE Rate Case, Recommended Decision 41. This issue, which is a subject of controversy in this case, is discussed below. E. ISSUES REGARDING CALCULATION OF ROE USING THE CONSTANT GROWTH DCF METHOD 1. PROXY GROUP The components of the DCF equation are not generally directly available for an individual utility because most investor-owned utilities are subsidiaries of larger companies and Recommended Decision Case No. 15-00261-UT 37 thus are not publicly traded. PNM itself is a vertically-integrated, electric-only utility. It is a subsidiary of the holding company, PNM Resources, Inc. (PNMR). It is PNMR’s stock that is publically traded, not PNM’s, and PNMR operates in both New Mexico and Texas. Pitts Direct 34. Therefore, the normal practice is to use proxy companies, or a population of publicly traded companies with significant utility business that are considered similar enough to the utility in question to be used as benchmarks in determining what investors will expect out of the utility in question. The composition of the proxy group used in a DCF analysis can be a subject of dispute. The PRC addressed this issue in detail in the 2007 SPS Rate Case, in which SPS, the AG, Occidental Permian LTD (“OPL”) and Staff all presented one or more DCF analyses. All of these parties used proxy data because SPS is a wholly-owned subsidiary of Xcel Energy and does not issue its own stock. However, there was much disagreement as to the composition of the proxy group. SPS, OPL, and Staff started with the 60 companies that Value Line classifies as electric utilities, but then each used different screens or criteria to exclude companies with different risk profiles than SPS. As a result, each of these parties developed different proxy groups. Additionally, SPS, in its Rebuttal Testimony, used a proxy group consisting of all of the utilities included in Staff’s, OPL’s, and SPS’s proposed proxy groups except for one utility, referred to as the “Combined Proxy Group.” The Attorney General used SPS’s proxy group. 2007 SPS Rate Case, Final Order 3, ¶¶ 8-9. The Hearing Examiner found that the most representative proxy group was the Combined Proxy Group, which included 28 utilities. The Hearing Examiner rejected SPS’s proposed proxy group of seven utilities because of infirmities inherent in SPS’s filtering process. Those infirmities included (i) SPS’s elimination of utilities whose regulated electric revenues and net income were less than 90% of total regulated revenues; and (ii) SPS’s elimination of PNM Resources because of its involvement in mergers and acquisitions activities. Also, the Hearing Recommended Decision Case No. 15-00261-UT 38 Examiner found that the small size of SPS’s proxy group rendered it vulnerable to anomalous events associated with any one of the companies in the narrow group. 2007 SPS Rate Case, Recommended Decision 61-66. In its Final Order, the Commission made three changes to the Combined Proxy Group recommended by the Hearing Examiner. First, the Commission eliminated PPL from the proxy group because its regulated electric delivery operations provided only 20% of consolidated net income and less than 50% of consolidated revenue. The Commission found no support for SPS’s screen that excluded utilities whose regulated electric revenues and net income were less than 90% of total regulated revenues, but found it reasonable to require that at least a majority of a utility’s net income be derived from regulated electric services. Second, the Commission eliminated Energy East and PNM Resources from the proxy group because all parties agreed that the proxy group should not include a utility involved in significant merger or acquisition activity. 2007 SPS Rate Case, Final Order 11, ¶ 31. The PRC concluded its discussion of proxy groups by emphasizing that there is not just one correct risk-comparable proxy group that should be used in any DCF analysis and that a number of different samples may be considered reasonable. It elaborated: More importantly, it is not necessary that each utility in the group share the exact same or even substantially identical risk characteristics as the utility in question. Rather it is more important that this group as a whole be risk comparable to that utility. This is because the results for the proxy group stem from an average of that group and will not be distorted if more or less risky members of the group cancel each other out. Final Order 12-13, ¶ 34. Another relevant source for considerations in selecting a proxy group is FERC Opinion 531, issued in 2014, in which FERC announced changes to its DCF method. Under Opinion 531, FERC guidelines include: x Companies that are included in the Electric Utility Industry groups compiled by Value Line x Electric utilities covered by at least two industry analysts Recommended Decision Case No. 15-00261-UT 39 x Electric utilities that are in a “comparable risk band” x Electric utilities that paid common dividends over the last six months and have not announced a dividend cut since that time x Electric utilities with no ongoing involvement in a merger or acquisition x Electric utilities with a published 5-year consensus earnings growth forecast from Institutional Brokers Estimate System Coakley v. Bangor Hydro-Elec. Co., Opinion No. 531, 147 FERC ¶ 61,234 (2014). Mr. Hevert’s proxy group (the Hevert Proxy Group) consists of 27 companies. He began with Value Line’s universe of 46 electric utilities and then applied the following screening criteria: x He excluded companies that do not consistently pay quarterly cash dividends. x He excluded companies that were not covered by at least two utility industry equity analysts. x He excluded companies that do not have investment grade senior unsecured bond and/or corporate credit ratings from S&P. x He excluded companies whose regulated operating income over the three most recently reported fiscal years was less than 60% of the respective totals for that company. x He excluded companies whose regulated electric operating income over the three most recently reported fiscal years represented less than 60% of total regulated operating income. x He excluded companies that he knew to be party to a merger or other significant transaction. Hevert Direct 18-19. Mr. Hevert submitted an updated DCF analysis in his Rebuttal Testimony. In this updated analysis, he removed five companies from his proxy group because of announced mergers or transactions. The companies he excluded are Black Hills, Duke Energy, Empire District, Southern Company, and TECO. Hevert Rebuttal 159-60. Mr. Hevert did not include PNM Resources in his proxy group “to avoid the circular logic that would otherwise occur[.]” Hevert Direct 19. Dr. Woolridge created two proxy groups. One of his proxy groups is the proxy group that Mr. Hevert used in his Direct Testimony, but excluding the five companies excluded by Mr. Hevert in his Rebuttal Testimony. Woolridge Direct 24, 26 n.15. Recommended Decision Case No. 15-00261-UT 40 Dr. Woolridge’s second proxy group — which he calls his Electric Proxy Group — has 27 companies. The qualifications for Dr. Woolridge’s Electric Proxy Group are: 1. 2. 3. 4. 5. 6. At least 50% of revenues from regulated electric operations as reported by AUS Utilities Report Listed as an Electric Utility by Value Line Investment Survey and listed as an Electric Utility or Combination Electric & Gas Utility in AUS Utilities Report An investment grade issuer credit rating by Moody’s and Standard & Poor’s (“S&P”) Has paid a cash dividend in the past six months, with no cuts or omissions Not involved in an acquisition of another utility, the target of an acquisition, or in the sale or spin-off of utility assets, in the past six months, and Analysts’ long-term earnings per share (“EPS”) growth rate forecasts available from Yahoo, Reuters, and/or Zacks Woolridge Direct 24-25. NMIEC witness Gorman used the Hevert Proxy group, but excluded three of the five companies that Mr. Hevert excluded in his Rebuttal Testimony. The companies he excluded are Black Hills Corporation, TECO Energy, and Southern Company. Gorman Direct 18. Dr. Pitts created two proxy groups to present results of two desired qualities of proxy groups that are difficult to represent in a single group. She explained that one of these desired qualities — having a group that is large enough to provide statistically relevant results — is difficult to achieve in combination with the other desired quality, which is: [F]orm[ing] a proxy group with an exact risk exposure given that utilities can be subsidiaries of a holding company, operate across state lines with different regulatory requirements, operate in different parts of the country, operate different percentages of natural gas or wholesale electric operations, or with different fuel compositions with varying levels of risk (i.e., coal-fired generation, nuclear power plants, renewable energy fuel sources, and so on). Pitts Direct 33. Dr. Pitts’ Proxy Group 1 consists of 12 electric utilities. Her Proxy Group 2 consists of 26 electric and gas utilities. Pitts Direct 34-36. The qualifications for Dr. Pitts’ Proxy Group 1 are: x At least 50% of revenues from regulated electric operations x Moody’s long-term rating of Baa3 or higher (investment grade) Recommended Decision Case No. 15-00261-UT 41 x Not involved in recent or ongoing merger/acquisition talks for electric or gas utilities x Electric utilities only The qualifications for Dr. Pitts’ Proxy Group 2 are: x x x x At least 60% of revenues from regulated electric operations Moody’s long-term rating of Baa3 or higher Not involved in recent or ongoing merger/acquisition talks for electric or gas utilities Listed in AUS Utility Reports, September 2014 edition, as either electric utility or combination electric and gas utility Pitts Direct 34-35. Selecting a proxy group is not an exact science and none of the proxy groups developed by the witnesses is clearly incorrect. Among the proferred proxy groups, Mr. Hevert’s Proxy Group, as revised in his Rebuttal Testimony, is the most appropriate. Mr. Hevert’s initial universe of utilities is electric only and taken only from Value Line. Dr. Pitts’ Group 2 and Dr. Woolridge’s Electric Proxy Group include electric and gas utilities and are taken from AUS Utility Reports. In its Opinion 531, FERC rejected reliance on AUS Utility Reports over Value Line. FERC explained, “Unlike Value Line, which is an investment-oriented publication, AUS Utility Reports is a service published primarily for regulators and is not typically relied on by investors.” Opinion 531, ¶ 100. Also, Dr. Pitts does not identify either of her proxy groups as being subject to a dividend screen. However, a dividend screen is used by FERC and is appropriate because the DCF model is based on investors’ expected return from dividend yield and growth. 2. GROWTH RATES The dividend growth rate in the DCF Method should reflect investors’ expectation of long-term dividend growth. As a substitute for actual investors, commissions rely on analysts’ projected growth rates. Hevert Direct 30. The PRC has consistently used an average of growth estimates from several rating agencies. 2007 PNM Electric Rate Case, Final Order 11, ¶ 29 (Zacks and BR growth rates should Recommended Decision Case No. 15-00261-UT 42 be considered); 2007 SPS Rate Case, Final Order 14 n.8 , ¶ 37 (using ((Value Line/Zacks average) + (Zacks/B*R average))/2); 2006 PNM Gas Rate Case, Final Order 7, ¶ 14 (using average of Value Line, Zacks, and B*R growth rates). Traditionally, the PRC has found Value Line to be a reliable source to use in conjunction with other rating agency (Zacks) projections and B*R growth because, among other reasons, Value Line is not in the business of selling securities, has its own independent research staff, and does not own stocks in the companies it analyzes. SPS 2007 Rate Case, Recommended Decision 71-72. In the 2012 SPS Rate Case, the PRC adopted SPS’s proposed 5.51% growth rate, which was an average of growth rates from Value Line, Zacks, and First Call. The Commission found that SPS had most closely followed Commission precedent. SPS 2012 Rate Case, Final Order 4, ¶ 8. In the 2006 PNM Gas Rate Case, the PRC rejected the argument that analysts’ growth rate projections extend over too short of a term to be relied on for the long-term growth estimate. This Commission quoted from a decision of the Illinois Commerce Commission stating, “‘[T]he Commission does not find merit in the Company’s assertion that a five-year period fails to adequately consider long-term growth expectations.” Recommended Decision 40 (quoting In Re Commonwealth Edison Co., 250 P.U.R.4th 161, 286 (Ill. C.C. 2006). It also cited to witness testimony that the longest security analyst forecasts go three to five years into the future. Id. In the context of rejecting PNM’s use of historic GDP growth, the Commission quoted from Case No. 2147, in which this Commission indicated its “‘agreement with the general approach of relying on projections for growth studies. Projections have been shown to be more accurate and more commonly relied on than historical rates.’” Id. at 41 (quoting Case No. 2147, Recommended Decision 43 (7-25-88)). For each proxy company, Mr. Hevert calculated mean, mean high, and mean low results. For the mean result, he combined the average of the Earnings Per Share (EPS) long-term growth rate estimates reported by Value Line, Zacks, and First Call, which he called the “average Recommended Decision Case No. 15-00261-UT 43 expected growth rate.” For the mean low result, he used the lowest of the three growth rates for each company in his proxy group, calculated the DCF result based on these lowests, and took the average of those low results across the proxy group. For the mean high result, he followed the same procedure, but used the highest of the three growth rates for each utility. Hevert Direct 33-34; Tr. (4-13-16) 579-80 (Hevert). The proxy group mean growth rates in his updated analysis are: Zacks First Call Value Line 4.98% 4.69% 5.52% Exh. RBH-1, p.8, to Hevert Rebuttal. To estimate the DCF growth rate, Dr. Woolridge reviewed both historical and projected growth rate measures and evaluated growth in dividends, book value, and EPS. He used the forecasted EPS growth rates of Wall Street analysts and the projected growth in EPS, DPS, and BVPS of Value Line. Dr. Woolridge calculated a range of growth rates for the Hevert Proxy Group from 4.75% to 5.0%. Based on this analysis, Dr. Woolridge used a DCF growth rate of 4.875% for the Hevert Proxy Group. Woolridge Direct 44-55. NMIEC witness Gorman calculated several growth rates, using various sources. Using the average of Zacks, Value Line, and a “br + sv” growth rate, he calculated an average 4.96% growth rate. Gorman Direct 35 & Exh. MPG-10 to Gorman Direct. Dr. Pitts separately calculated growth rates from several sources and using various growth periods: x Model 1: One-year growth rate for dividends (Thomson Reuters) x Model 1a: Five-year growth rate for dividends (Thomson Reuters) x Model 2: Historic 3-5 year growth rate for EPS (Zack's) x Model 2a: Historic 3-5 year growth rate for EPS, removing negative growth rates (Zack's) x Model 3: Historic 5-year growth rate for EPS (Thomson Reuters) Recommended Decision Case No. 15-00261-UT 44 x Model 3a: Historic 5-year growth rate for EPS, removing negative growth rates (Thomson Reuters) x Model 4: Projected 1-year growth rate for EPS (Zack's) x Model 4a: Projected 1-year growth rate for EPS, removing negative growth rates (Zack's) x Model 4b: Projected 1-year industry growth rate for EPS (Zack's) x Model 5: Projected 3-5 year growth rate for EPS (Thomson Reuters) Pitts Direct 46-47. Ms. Pitts’ calculated growth rates are in Exhibit 10 to her Direct Testimony. Dr. Pitts said that she “chose to estimate individual specifications using each selected growth rate and then create a zone of reasonableness from the median ROE estimates[.]” Pitts Direct 44. Both Mr. Hevert and Mr. Gorman followed Commission precedent in estimating growth rates. Mr. Hevert’s average growth rates should be used because they are based on more recent data than Mr. Gorman’s estimated growth rate. See PNM Gas Servs., 2000-NMSC-012, ¶ 87 (“[C]ommon sense requires that the latest available economic information should be utilized in order to insure that the projected figures bear a meaningful relation to future as well as past and present fiscal realities.”). In fact, Mr. Hevert’s average growth rates that Mr. Hevert used in his updated analysis dropped from the growth rates he used initially. Cf. Exh. RBH-3, p.8, to Hevert Direct with Exh. RBH-1, p.8, to Hevert Rebuttal. F. RECOMMENDED INPUTS TO CONSTANT GROWTH DCF ANALYSIS Adopting the above recommendations, the reasonable range of ROEs in this case should preliminarily be set using a Constant Growth DCF Analysis with the following inputs: x A 360-day average period to determine stock price x A full year’s dividend growth to determine dividend yield x The Hevert Proxy Group, as revised in Mr. Hevert’s Rebuttal Testimony to exclude five companies x Mr. Hevert’s updated average growth rate estimate Recommended Decision Case No. 15-00261-UT 45 G. RESULTS OF PARTIES’ AND STAFF’S CONSTANT GROWTH DCF ANALYSES The AG did not perform a Constant Growth DCF analysis using a 360-day average period to determine stock price. Woolridge Direct 42. He also did not perform a Constant Growth DCF analysis using one year of projected growth; he used only one-half year of projected growth. Woolridge Direct 44. NMIEC witness Gorman submitted a DCF analysis that uses all of the above inputs, except his proxy group includes two companies that both Mr. Hevert and Mr. Woolridge said should be excluded: Duke Energy and Empire District. Mr. Gorman’s analysis produced a 9.03% average ROE and an 8.88% median ROE. Exh. MPG-10 to Gorman Direct. Staff witness Pitts’ Constant Growth DCF analyses do not use the Hevert Proxy Group, as revised. For each of two proxy groups, Staff witness Pitts submitted the results of ten Constant Growth DCF analyses that all use a 360-day average period to determine the stock price used to calculate the dividend yield and all use a full year’s dividend growth to determine the dividend yield. The analyses differ only in how dividend growth was modeled. They incorporate different sources of historic and projected growth rates. Pitts Direct 43. The following table shows the median results of these analyses: Model DCF1: 1-year forecast dividend growth from ThomsonReuters/IBES gathered on 12-11-15 DCF1A: 5-year forecast dividend growth from ThomsonReuters/IBES gathered on 12-11-15 DCF2: 3-5 year historic EPS growth from Zacks gathered on 12-17-15 DCF2A: 3-5 year historic EPS growth from Zacks gathered on 12-17-15, excluding negative growth numbers DCF3: 5-year historic EPS growth from ThomsonReuters/IBES gathered in Dec., 2015 DCF3A: 5-year historic EPS growth from ThomsonReuters/IBES gathered in Dec., 2015, excluding negative growth numbers Recommended Decision Case No. 15-00261-UT 46 Proxy Group 1 Proxy Group 2 9.26% 9.26% 7.47% 7.91% 8.28% 8.25% 8.47% 8.47% 12.69% 11.25% 12.93% 11.96% DCF4: 1-year forecast EPS growth from Zacks gathered on 12-17-15 DCF4A: 1-year forecast EPS growth from Zacks gathered on 12-17-15, excluding negative growth numbers DCF4B: 1-year forecast industry EPS growth from Zack’s gathered on 12-17-15 DCF5: 3-5-year forecast EPS growth from ThomsonReuters/IBES gathered in Dec., 2015 -2.28% 5.68% 10.63% 9.55% 7.36% 7.28% 7.38% 7.95% Exh. HMP-3 to Pitts Direct. Dr. Pitts is the Staff witness who recommended a 9.225% ROE in the 2015 EPE Rate Case. Dr. Pitts explained that her recommended ROE in this case is lower because the data supporting her analyses in this case is from December 2015, whereas the data supporting her analyses in the 2015 EPE Rate Case was from summer 2015. The later data in this case produced lower ROEs. Tr. (4-29-16) 3473. Mr. Hevert’s Constant Growth DCF Analysis, as updated in his Rebuttal Testimony, is the only Constant Growth DCF Analysis in evidence that uses all of the recommended inputs. This analysis updated his results through January 29, 2016, and produced the following results: Mean Low Mean Mean High 8.49% 9.18% 9.97% Exh. RBH-1, p.8 to Hevert Rebuttal. Preliminarily, based on the results of Mr. Hevert’s analysis, a reasonable range of ROEs is from 8.49% to 9.97%. H. PROPOSED ADJUSTMENTS PNM witness Hevert testified that, in determining an authorized ROE, the PRC should also consider two risk factors specific to PNM: (1) the effect of PNM’s substantial capital expenditure plan, which rating agencies call “capital acquisitions risks”; and (2) PNM’s relative small size compared to its peer utilities. Mr. Hevert said that credit rating agencies as well as Recommended Decision Case No. 15-00261-UT 47 investors consider higher levels of capital investment as indicative of high risk requiring a greater return. Hevert Direct 52-53; Tr. (4-13-16) 530 (Hevert). Mr. Hevert did not propose or quantify an adjustment to reflect this risk but considered it in determining a fair ROE. Tr. (413-16) 524 (Hevert). While Mr. Hevert calculated a range for a risk premium associated with PNM’s smaller size, he did not propose or quantify any adjustment. Rather, Mr. Hevert considered PNM’s smaller size in determining whether in the range of reasonable ROEs it was appropriate to set PNM’s ROE. Hevert Direct 58. These factors should not be considered for the reasons that the PRC did not consider EPE’s claim in the 2015 EPE Rate Case that it faced a greater business risk than the average of the proxy companies. AG witness Woolridge proposed an adjustment to reflect PNM’s slightly riskier investment risk compared to the Hevert Proxy Group. At the time when PNM filed its initial testimony and when Intervenors filed their testimonies, Moody’s and S&P rated PNM’s senior unsecured debt at Baa2/BBB, respectively, which are both investment grade ratings. Woolridge Direct 26-27. Dr. Woolridge said in his prefiled testimony that, based on those ratings, PNM’s investment risk was “a little above” the investment risk of the Hevert Proxy Group. For the Hevert Proxy Group, the average Moody’s and S&P ratings are BBB+ and Baa1. He said that PNMR’s investment risk was also “a little riskier” than the Hevert Proxy Group. Woolridge Direct 26-28. After Mr. Woolridge filed his testimony, S&P upgraded its rating of PNM’s senior unsecured debt to BBB+. Tr. (4-19-16) 1663 (Eden). However, for the Moody’s rating, PNM’s rating remains below the average of the Hevert Proxy Group, and in May 2016, Moody’s published a report stating that the extension of the suspension period in this case was “credit negative.” Eden 5-25-16 Supp. 11; Exh. EAE-3 to Eden 5-25-16 Supp. Mr. Woolridge said that bond ratings “provide a good assessment of the investment risk of a company,” and he concluded that PNM’s ROE should be adjusted 0.30% because of its additional risk. Woolridge Recommended Decision Case No. 15-00261-UT 48 Direct 27, 69; Tr. (4-13-16) 607 (Hevert) (risk is a measure of uncertainty). Mr. Gorman agreed that bond ratings are relevant to assessing risk in determining ROEs. Gorman Direct 7-13. Dr. Pitts warned against putting too much reliance on bond ratings, but said that they are a consideration in setting a utility’s ROE. Pitts Direct 29. The evidence supports incorporating PNM’s slightly riskier investment risk compared to the Hevert Proxy Group in determining PNM’s authorized ROE. I. HEARING EXAMINER’S RECOMMENDED ROE As stated above, preliminarily, based on the results of Mr. Hevert’s analysis, a preliminary reasonable range of ROEs is from 8.49% — Mr. Hevert’s low mean DCF result — to 9.97% — Mr. Hevert’s high mean DCF result. Taking into account PNM’s slightly riskier investment risk compared to the Hevert Proxy Group, PNM’s authorized ROE should be on the higher end of this range. To place PNM’s authorized ROE in the higher end of this range, it is reasonable to set PNM’s authorized ROE as the average of Mr. Hevert’s mean DCF result — 9.18% — and Mr. Hevert’s high mean DCF result — 9.97%. This results in a 9.575% ROE. All of the ROE witnesses cite to Regulatory Focus, a quarterly publication by Regulatory Research Associates that reports on state commission authorized ROEs for electric and gas companies. Hevert Rebuttal 50; Woolridge Direct 70; Pitts Direct 48 & Exh. HMP-13 to Pitts Direct; Tr. (4-22-16) 2337 (Gorman). According to Regulatory Focus, the average authorized ROE was 9.68% for the first quarter of 2016 compared to 9.58% for the full year 2015. PNM Exh. 74. In determining the weight to be given ROEs authorized by other state commissions, this Commission said: [W]hile not ignoring the ROEs granted by the other states, the Commission’s decision should not be determined by the returns granted elsewhere, but must be determined relying on the Commission’s expert judgment and guided by the record evidence[.] Recommended Decision Case No. 15-00261-UT 49 2007 PNM Electric Rate Case, Final Order 8, ¶ 20 (quoting 2006 PNM Gas Rate Case, Final Order 7, ¶ 15). While not relying on other state commissions’ authorized ROEs, the reasonableness of the recommended 9.575% ROE is confirmed when compared with the data reported by Regulatory Focus. XII. PNM’S TEST PERIOD WEIGHTED AVERAGE COST OF CAPITAL PNM’s proposed Test Period weighted average cost of capital, based on its proposed 10.5% ROE, is 8.17%.19 Using PNM’s proposed cost of debt and cost of preferred stock and a 9.575% ROE, the resulting Test Period Weighted Average Cost of Capital is 7.71%, as follows: PNM Capital Structure and Weighted Average Cost of Capital Class of Capital % of Total % Cost Weighted Average Cost Long-Term Debt 50% 5.87% 2.94% Preferred Stock 0.39% 4.62% 0.02% Common Equity 49.61% 9.575% 4.75% Total 100% 7.71% XIII. RENEWABLE ENERGY RATE RIDER In the PNM Renewable Rider Case, the PRC approved PNM’s use of a rate rider — Rate Rider No. 36 or PNM’s Renewable Energy Rate Rider (Renewable Rider) — to recover the costs of renewable resources approved by the PRC for PNM to meet the Renewable Portfolio Standard (RPS). The Renewable Rider collects PNM’s costs of complying with the RPS on a per kWh basis. The PRC required an annual true-up, initiated by PNM’s filing of an advice notice annually no later than February 28. A hearing is not required prior to annual adjustments to the Rider so long as a hearing is held in PNM’s annual renewable energy portfolio procurement plan ͳͻ ‘” ‹”‡…– ƒ– ʹʹǦʹ͵Ǥ Recommended Decision Case No. 15-00261-UT 50 case. Recommended Decision 53-54, Decretal ¶ H. The PRC required PNM to apply for any desired continuation of the Renewable Rider in its next general rate case. Recommended Decision, Decretal ¶ J. In PNM’s Renewable Rider Case, the PRC approved PNM’s use of the Renewable Rider over objections that it was unreasonable to single out the costs of renewable energy on customers’ bills. The PRC said, “Instead of burying the recovery of costs in a rate case, and accruing carrying charges, the rider, utilizing the true-up process, makes it more likely that actual costs are recovered and recovered timely.” Final Order 6, ¶ 6. Absent a rider to collect costs of complying with the RPS, these costs are booked as a regulatory asset with carrying charges accumulating until recovery is authorized in a general rate case. Ortiz Direct 45. The PRC found that the Renewable Rider’s avoidance of carrying charges would allow for more headroom to purchase more renewable energy and that “[t]his savings alone justifies the rider.” PNM’s Renewable Rider Case, Final Order 7, ¶ 7. The PRC also rejected piecemeal ratemaking objections to the Renewable Rider. Id. at 5-6, ¶ 5. In this case, PNM argues that the advantages of using a renewable rider identified by the PRC in Case No. 12-00007-UT support continued use of the Renewable Rider. Another benefit of continuing to use the Renewable Rider is that ratepayers more quickly and timely realize the benefit of decreases in the revenue requirement for PNM-owned solar facilities. As these facilities age, their revenue requirement decreases to reflect their increased depreciation. Because the Renewable Rider revenue requirement has been adjusted annually, ratepayers receive the benefit of the declining revenue requirement for these facilities annually. Ortiz Direct 44-50. The current version of the Renewable Energy Rider is the 8th Revised Rider No. 36, and the current Renewable Energy Rider Rate is $0.0048409/kWh. This Rate took effect by operation of law. Advice Notice No. 520 (2-26-16). Recommended Decision Case No. 15-00261-UT 51 If recovery of the NMWEC procurement costs is moved from the FPPCAC to the Renewable Rider, PNM projects the Renewable Energy Rider Rate for the Test Period to be $0.0078966/kWh. Exh. JCA-19, p.3, to Aguirre 3-14-16 Supp. No party nor Staff opposes continuation of PNM’s Renewable Rider, and its continuation should be approved, as modified to collect PNM’s procurement costs for the New Mexico Wind Energy Center, as discussed in Section XIV(A). XIV. A. FUEL AND PURCHASED POWER COSTS ELIMINATION OF FUEL MISALLOCATION PNM requests approval of a $174,786,171 fuel revenue requirement20, which would be a ($41,075,092), or (19%) fuel revenue decrease from base fuel revenues of $215,861,26321. Two changes drive the decrease in the fuel revenue requirement: First, PNM entered into a new coal supply agreement for the SJGS in 2016 that resulted in substantial fuel savings. Second, in 2016, PNM ceased recovering from ratepayers a $47.8 million under-recovered balance in its FPPCAC balancing account that it had been collecting between July 2014 and December 2015. This recovery was authorized by the PRC in the 2013 PNM Fuel Clause Case. Ortiz Direct 25-28. PNM’s current FPPCAC was initially authorized in 2008. Case No. 08-00092-UT, Final Order (5-22-08). The PRC PNM most recently authorized PNM to continue using the FPPCAC in 2014. 2013 PNM Fuel Clause Case. PNM’s FPPCAC fuel clause year runs from July to June and the FPPCAC factor is reset quarterly. Ortiz 2-24-16 Supp. 2. Currently, PNM collects some of its fuel costs through its base rates. These are recovered through PNM’s “base fuel rate.” PNM’s current $0.02128/kWh base fuel rate was approved in the 2010 PNM Rate Case. This base fuel rate represents a part of PNM’s per kWh volumetric 20 21 Tr. (6-28-16) 3928-29 (Monroy). Monroy Direct 5, Table HEM-1, line 2, “PNM Retail” column. Recommended Decision Case No. 15-00261-UT 52 rates. If PNM’s actual fuel costs are greater than the revenues it collects from its base fuel rate, it recovers its undercollected fuel costs through its FPPCAC. Conversely, if PNM’s actual fuel costs are less than the revenues it collects from its base fuel rate, PNM returns its overcollected fuel revenues to customers through its FPPCAC. Carrara 3-8-16 Affidavit, ¶ 7. The FPPCAC rate is calculated by dividing the projected fuel cost for a period (which is adjusted by any over- or under-collection in the previous period) by the projected energy to be billed for the same period, in kWh. Carrara 3-8-16 Affidavit, ¶ 8. Currently, the projected energy to be billed (the Denominator of the equation) includes kWh generated by renewable energy. If kWh generated by renewable energy are removed from total kWh in the Denominator, there are less kWh over which to collect the projected fuel cost, and the per kWh FPPCAC rate therefore increases. Carrara 3-8-16 Affidavit, ¶ 9. In its Application, PNM proposed to not change its current $0.02128/kWh base fuel rate. Taylor Supp. (2-24-16) 10. In Orders issued in February and March 2016, the PRC ordered PNM to revise its energy rates so that no fuel costs are recovered through its energy rates, its base cost of fuel is zero, and it recovers all fuel costs through its FPPCAC. The PRC also ordered PNM to: x x Remove New Mexico Wind Energy Center (NMWEC) procurement costs from total costs in the FPPCAC calculation Remove sales of kWh generated by renewable energy from total kWh sales in the FPPCAC calculation Order Extending Suspension Period, Requiring the Filing of Additional Testimony, and Directing the Hearing Examiner to Set a New Procedural Schedule 5-6, ¶ B (3-2-16). The last two bulleted directives above came about from the 2015 PNM RPS Case, in which PNM had been required to identify a method of eliminating what had been referred to as the “disproportionate avoided fuel benefit” (DAFB) to PNM’s Exempt and Large Capped Customers. Exempt and Large Capped Customers are two categories of customers who are treated differently than “Other Customers” under the REA. The REA has a Renewable Portfolio Recommended Decision Case No. 15-00261-UT 53 Standard (RPS) requirement, which, for 2016, is that renewable energy shall comprise no less than 15% of a public utility’s total retail sales to New Mexico customers. NMSA 1978, § 62-164(A). Exempt Customers are exempt from paying all charges by a utility for renewable energy procurements used to comply with the REA in a year. Id., § 62-16-4(A)(3). Large Capped Customers are capped in the amount that they can be charged by a utility for renewable energy procurements used to comply with the REA. In 2016, the annual cap is $110,479, and in 2017, it is $113,241. 2015 PNM RPS Case, Recommended Decision 10. Evidence in Case No. 15-00166 showed that the so-called DAFB occurs because, while Exempt Customers are exempt from paying for renewable energy procured by PNM and Large Capped Customers are capped in their payment for renewable energy procured by PNM, they receive offsetting fuel savings based on the total amount they would pay for renewable energy if they were not capped. This is because the fuel savings resulting from PNM’s procurement of renewable energy flow through to customers through PNM’s FPPCAC, which has no mechanism to exempt the Exempt Customers from receiving the savings or capping the savings received by Large Capped Customers. 2015 PNM RPS Case, Recommended Decision 36-51. The Hearing Examiner recommended eliminating the DAFB by adjusting PNM’s Renewable Rider. In its Final Order, the PRC disagreed with the concept of a “DAFB,” explaining that rates are set to collect costs and not to allocate benefits from avoided costs. The PRC recharacterized the issue as one of fuel misallocation and found that it would be more appropriate to address the fuel misallocation issue by considering changes to PNM’s FPPCAC than by changing PNM’s Renewable Rider. 2015 PNM RPS Case, Final Order Superceding Vacated Final Order, ¶ 24. A source of the fuel misallocation identified by the Commission is that PNM includes sales of kWh generated by renewable energy in total sales in its FPPCAC calculation. If sales of kWh generated by renewable energy are excluded from total sales in the FPPCAC calculation, Recommended Decision Case No. 15-00261-UT 54 the FPPCAC Factor applies to only the portion of a customer’s energy use that is estimated to be generated by nonrenewable energy. Another problem identified by the Commission is that PNM recovers its New Mexico Wind Energy Center (NMWEC) procurement costs through its FPPCAC, rather than through its Renewable Rider, which is how it collects the costs of its other renewable procurements. The PRC approved PNM’s procurement of energy from the NMWEC before the Legislature required utilities to procure renewable energy through enactment of the Renewable Energy Act. Since enactment of the REA, PNM has applied RECs associated with NMWEC wind energy generation to its RPS requirement. PNM values these RECs at zero additional cost above the energy cost. Ortiz 3-14-16 Supp. 25. Because PNM recovers the NMWEC costs through its FPPCAC, those costs are being paid for by Exempt Customers and not being capped for Large Customers. However, because PNM applies Renewable Energy Certificates (RECs) associated with the NMWEC toward its Renewable Portfolio Standard (RPS), these costs, under the REA, cannot be recovered from Exempt Customers and are subject to the Large Customer Cap. Staff agrees that if PNM uses the NMWEC procurement to comply with its RPS, then the NMWEC procurement costs should be subject to the Exemption and Large Customer Cap. Carrara 3-8-16 Affidavit, ¶ 13. Moving recovery of the NMWEC procurement costs from the FPPCAC to the Renewable Energy Rider increases the revenue allocation to the Residential Class by $892,067. Aguirre 3-14-16 Supp. 5; Exh. JCA-2 (3-14-16 Supp.), p.1. Moving recovery of the NMWEC procurement costs from the FPPCAC to the Renewable Rider would increase by about $14.6 million the costs recovered through the Renewable Rider. Exh. JCA-2, p.1 to Aguirre 3-14-16 Supp. (column F, row 14 – row 28). In its February 10, 2016 Order, the Commission ordered PNM to address its current calculation of base fuel cost in its base rates as it relates to: Recommended Decision Case No. 15-00261-UT 55 1. PNM’s apparent inclusion of sales of kWh generated with renewable energy in its FPPCAC 2. PNM’s inclusion of NMWEC costs in its FPPCAC In supplemental response testimony, Mr. Ortiz testified that “it would be impossible,” using PNM’s current billing system, to not account for sales of kWh generated with renewable energy in the calculation of PNM’s FPPCAC. This is because, according to PNM, to do what the Commission requested would require charging two different energy rates to its customers: 0ne for kWh generated by renewable energy and one for kWh generated by non-renewable energy. “PNM’s billing system would not allow PNM to bill the base fuel charge and FPPCAC factor on only a portion of a customer’s energy usage.” Mr. Ortiz says that adding such capability to PNM’s billing system “would take considerable time and expense.” Ortiz 2-22-16 Supp. In its March 2, 2016 Order, the Commission said that PNM’s failure to correct its FPPCAC calculations because its billing system cannot bill more than one base variable charge and cannot bill the FPPCAC factor on only a portion of a customer’s energy use is simply unacceptable. The PRC said, “PNM’s current method of calculating its FPPCAC does not serve ‘the goal of providing reasonable and proper service at fair, just and reasonable rates to all customer classes,” which would be cause for the Commission to eliminate PNM’s FPPCAC.” Order Extending Suspension Period et al., ¶ 10. The PRC ordered PNM to (i) remove sales of kWh generated by renewable energy from total sales in the FPPCAC calculation by removing fuel costs from its base energy rates and collecting all of its fuel costs through its FPPCAC; and (ii) remove NMWEC procurement costs from total costs in the FPPCAC calculation. The Commission also ordered PNM Id., ¶ 11 & Decretal ¶ B. PNM witness Susan Taylor responded that, to carry out the PRC’s order, PNM would have to create three balancing accounts to apply to the following three different customer groups: i) Exempt Customers; (ii) Large Capped Customers; and (iii) Non-Capped/Non-Exempt Recommended Decision Case No. 15-00261-UT 56 Customers (Other Customers). Taylor 3-14-16 Supp. 8. She explained that separate balancing accounts are necessary because the percentage of a customer’s energy use that is generated by nonrenewable energy varies among the three groups. This is because (i) Exempt Customers are exempt from paying any costs incurred by PNM to comply with the Renewable Energy Act (REA); (ii) Large Capped Customers are capped in the amount they pay toward PNM’s costs of complying with the REA; and (iii) Other Customers pay a percentage of their energy use toward PNM’s RPS compliance costs that is equal to PNM’s net Renewable Portfolio Standard (RPS). Taylor 3-14-16 Supp. 8; Case No. 15-00166-UT, Recommended Decision 9-10 (10-20-15). To carry out all of the PRC’s directives, PNM proposed “Method A,” which makes the following changes to PNM’s current methods of billing fuel and purchased power costs and renewable costs: 1. Recovers all fuel and purchased power costs through PNM’s FPPCAC, and recovers no such costs through PNM’s base energy rates 2. Moves recovery of the NMWEC procurement costs from the FPPCAC to the Renewable Rider 3. Excludes kWh generated by renewable energy from the calculation of total kWh sales in calculating PNM’s FPPCAC, thereby applying the FPPCAC Factor to only the portion of a customer’s energy use that is estimated to be generated by nonrenewable energy 4. Breaks the FPPCAC charge on a customer bill into two parts: a. One FPPCAC Factor applies to the estimated percentage of a customer’s fuel use generated by non-renewable energy b. The other FPPCAC Factor applies to the estimated percentage of a customer’s fuel use generated by renewable energy — this FPPCAC Factor is zero because no fuel use is associated with use of renewable energy. Therefore, the charge on this line of a customer’s bill will always be zero. Recommended Decision Case No. 15-00261-UT 57 5. Uses a different estimated percentage of fuel use generated by renewable energy for each of the three customer classes, which is then multiplied by the FPPCAC factor. The estimated percentage of energy use generated by renewable energy for Large Capped Customers is 5.176% and 13.258% for Other Customers. Aguirre 3-14-16 Supp. 9; Exh. SAT-2 to Taylor 314-16 Supp. For Exempt Customers, the percentage, of course, is zero. A sample bill that results from these changes is Exhibit A to this Recommended Decision (attach Exh. GTO-1, p.2 to Ortiz 3-14-16 Supp.). Under Method A, Exempt Customers and Large Capped Customers pay more fuel costs. Taylor 3-14-16 Supp. 9. Exempt Customers would pay an estimated $593,641 more in fuel costs; Large Capped Customers would pay an estimated $1,470,511 more in fuel costs; and Other Customers will pay an estimated $2,064,152 less in fuel costs. Exh. SAT-2 to Taylor 3-14-16 Supp. This is because Large Capped Customers would be billed a higher percentage of their energy use at the resulting higher non-renewable FPPCAC factor and Exempt Customers would be billed for 100% of their energy use at the resulting higher non-renewable FPPCAC factor. Ortiz 3-14-16 Supp. 32. PNM agrees in this case, as it has in previous cases, that a fuel misallocation exists, but PNM, as in previous cases, does not take a position on whether it should be eliminated. Ortiz 314-16 Supp. 23. PNM has made the following statements among others: x The PRC’s suggested changes to the FPPCAC calculation would result in “only a de minimus impact” for the vast majority of PNM customers. Taylor 3-14-16 Supp. 10. x “While WRA characterizes the issue as a matter of misallocated fuel cost, I see this as a distinction without a difference.” Ortiz 3-30-16 Supp. 6. x “Including the renewable energy production in the calculation of [PNM’s] base fuel and FPPCAC adjustment factor rates does not affect the amount of fuel and purchased power costs recovered from customers.” Ortiz 3-14-16 Supp. 14. x There would be no benefit to removing kWh generated by renewable energy from total kWh in the FPPCAC calculation because “[d]oing so would not affect the total amount billed.” Ortiz 2-24-16 Supp. 5. Recommended Decision Case No. 15-00261-UT 58 PNM’s repeated minimizing of the impact of the cost misallocation is exasperating and shows concern only with its bottom line and not fairness to its customers. NMIEC witness James Dauphinais testified in this case that “NMIEC does not believe that under New Mexico law there is a disproportionate fuel benefit for exempt and capped customers.” Dauphinais 3-23-16 Supp. 4. And, Mr. Dauphinais testified that the cost and burden of Method A are not justified by the amount of fuel costs that would be reallocated. If the PRC adopts Method A, he recommends that it be done gradually to minimize the rate impact on Exempt and LCC. Dauphinais 3-23-16 Supp. 6. WRA witness Howe testified that PNM’s Billing Method A would provide a more accurate presentation and fair outcome for all PNM customers. He says that “the new billing method will replace the confusing current situation with a more consistent system that identifies all REA-procured renewable energy costs in a single place.” Howe 2-23-16 Supp. 2. He explains that Billing Method A would clarify the costs and benefits of renewable energy: Customers would see explicitly the fuel costs of the conventional resources that serve them, and the zero fuel cost of the renewable energy that serves them. Mr. Howe aptly concluded, [T]he Commission’s suggested methodology, and PNM’s proposed implementation of that method through its “Method A,” makes sense of what until now has been a very confusing cost recovery and billing methodology that resulted in uncapped non-exempt customers subsidizing the fuel costs for generation used to serve capped and exempt customers. Howe 3-23-16 Supp. 3. CCAE supports WRA’s position. CCAE’s Initial Posthearing Brief 56. The PRC has made clear that it believes that currently there is a fuel misallocation that it wants corrected. The reason for correcting the fuel misallocation is the well-established policy of cost causation, which is the basis in large part of PNM’s proposed rates. Because a greater proportion of the energy used by Large Capped Customers is served by conventional resources, they should bear the cost of those resources. Howe 3-23-16 Supp. 4. Recommended Decision Case No. 15-00261-UT 59 Staff refers to the PRC’s suggested method of correcting the fuel misallocation as “the Clarification Method.” Staff recommends that if the Clarification Method is used, it should be modified. It recommends what it calls an “Energy Balancing Method,” which makes three changes to PNM’s Method A. First, Staff says that PNM should recalculate its FPPCAC rate monthly rather than quarterly because the actual percentage amounts of renewable energy used will vary too much from the assumed and applied percentage amounts renewable energy. Carrara 3-23-16 Affidavit, ¶ 13. Second, Staff calculates the percentages of renewable energy used to serve LCC and NonLarge/Non-Capped Customers differently than does PNM. PNM percentages reflect total renewable generation as a percentage of total load. Staff’s percentages reflect total renewable generation as a percentage of total generation. Taylor 3-30-16 Supp. 2, 4-5. Mr. Carrara testified that total generation can exceed total load by 10-20% because of factors such as line losses, unbilled amounts, unaccounted for energy consumption and off-system sales. Carrara 323-16 Affidavit, ¶ 14. Because PNM’s percentages reflect total load, PNM’s method assigns all of these energy reductions to non-renewable resources. PNM says that it is appropriate to assign line losses to non-renewable resources because non-renewable resources frequently are located on the distribution system and do not have the same level of losses associated with them as do non-renewable resources that located on the transmission system, farther from load centers. Taylor 3-30-16 Supp. 4. Staff’s method assumes that off-system sales are served in part by renewable resources. Staff estimates the amount of renewable energy supplied to Large Capped Customers based on the relative contribution of revenues from Large Capped Customers toward the Renewable Rider. Under Staff’s method, the estimated percentage of energy use generated by renewable Recommended Decision Case No. 15-00261-UT 60 energy for Large Capped Customers is 1.675% and 98.324% for Other Customers. Aguirre 3-3016 Reply 4; Exh. BEC-AFF2-4 (1.675% = 0.196% + 1.479%). PNM disagrees with Staff’s assumption that off-system sales are served in part by renewable resources because renewable resources are “must take” resources that are dispatched first. PNM also says that because generation associated with off-system sales is not used to serve retail load, off-system sales MWh should be removed from total generation if Staff’s method is used. Taylor 3-30-16 Supp. 4-5. Third, Staff calculates the percentage of renewable energy supplied to LCCs differently. PNM’s estimate of 5.176% is roughly equal to the LCCs’ energy contribution to PNM’s 2016 RPS. Aguirre 3-30-16 Reply 4. In estimating the amount of renewable energy supplied to LCCs, PNM, in general terms, divided each LCC’s dollar cap by $0.0167/kWh, which is the net per kWh cost of renewable energy approved by the PRC in Case No. 15-00166-UT for purposes of calculating PNM’s RPS. PNM then converted the estimated kWh of renewable energy supplied to each LCC to a percentage of the LCC’s total estimated use. The average percentage of renewable energy supplied to the LCCs is 5.176%, which is the percentage of renewable energy that PNM estimates is provided to LCCs. Aguirre 3-14-16 Supp. 8 & Exh. JCA-3, p.2, to Aguirre 3-14-16 Supp. While Staff’s examination of the merits of Method A is appreciated, Method A is preferable and should be adopted for the time being. However, as Dr. Howe pointed out, Method A does not completely remove the cost misallocation with respect to Large Capped Customers. Howe 3-23-16 Response 2-3; see also Tr. (4-11-16) 90-91 (Ortiz). Dr. Howe estimated that Method A only corrects one-third of the fuel misallocation, and that the total fuel misallocation amount is more. Tr. (4-25-16) 2591, 2679. The remaining fuel misallocation occurs because PNM determines the amount of renewable energy to procure for Large Capped Customers on a net cost basis (gross cost less fuel savings), but PNM pays gross cost when it procures the renewable energy. Dr. Howe calls the dollar difference “the missing money.” Tr. Recommended Decision Case No. 15-00261-UT 61 (4-25-16) 2678-80. WRA recommends eliminating the remaining fuel misallocation in PNM’s next renewable energy plan case, to be filed in 2017. Dr. Howe suggests that the remaining fuel misallocation could be eliminated by recovering an additional amount from Large Capped Customers under the Renewable Rider equal to the per MWh dollar difference between the gross and net cost of procuring renewable energy multiplied by the percentage of renewable energy allocated to a Large Capped Customer (5.176% under Method A). The total additional amount charged to Large Capped Customers would be subtracted from the total amount charged to Other Customers under the Renewable Rider. Howe 3-23-16 Response 2-3; Tr. (4-25-16) 268788 (Howe). Another way to correct the remaining fuel misallocation might be to calculate the amount of each Large Capped Customer’s dollar cap on a net cost basis rather than on a gross cost basis. PNM currently calculates the dollar cap on a gross cost basis. Tr. (4-11-16) 87-90 (Ortiz); Tr. (4-25-16) 2679-80 (Howe). Calculating the dollar cap on a gross cost basis may be inconsistent with the PRC’s holding in PNM 2015 Renewable Plan Case, Recommended Decision 18-19, that the Large Customer Adjustment be calculated on a net cost basis. It may be inconsistent to calculate the Large Customer Adjustment on a net cost basis but calculate each Large Capped Customer’s dollar cap on a gross basis.22 Method A should be adopted in this case. Method A incorporates the PRC’s recent directive to EPE in the 2015 EPE Rate Case: The Commission further finds that EPE should be ordered to remove all fuel and purchased power costs from base rates and instead, bill such costs entirely through the FPPCAC. Such treatment of these costs should provide greater transparency to ratepayers with regard to the nature of the charges on their bills. In particular, separation of the highly variable fuel costs from other, more fixed costs, should assist ratepayers in understanding the factors influencing the magnitude and variability of their bill charges. In fact, page 2 of Exh. SG-6 to Shane Gutierrez’s August 28, 2015 Supplemental Testimony in the PNM 2015 Renewable Plan Case showed that when net cost was used to calculate the Large Customer Adjustment and gross cost was used to calculate a Large Capped Customer’s dollar cap, one Large Capped Customer (Customer W) was no longer a Large Capped Customer. 22 Recommended Decision Case No. 15-00261-UT 62 2015 EPE Rate Case, Final Order 106, ¶ 242. In its renewable energy plan case to be filed in 2017, PNM should file testimony that: 1. Addresses and corrects the remaining cost misallocation identified by Dr. Howe 2. Discusses the merits of correcting the remaining fuel cost allocation through the method suggested by Dr. Howe 3. Discusses the merits of correcting the remaining fuel cost allocation by calculating each Large Capped Customer’s dollar cap on a net cost basis 4. Discusses the merits of any other method of correcting the remaining fuel allocation that might be suggested by PNM 5. Discusses whether each Large Capped Customer’s dollar cap should be calculated on a net cost basis to be consistent with calculating the Large Customer Adjustment on a net cost basis. Removing kWh generated by renewable energy from the FPPCAC calculation requires reprogramming PNM’s Banner billing system to create complementary fixed non-renewable and renewable energy percentages that always total 100% of a customer’s monthly energy use. Reprogramming would be necessary because PNM’s billing system currently is configured to apply the FPPCAC Factor to 100% of a customer’s energy use. Ortiz 3-14-16 Supp. 7, 10. PNM witness Houck estimated that it would take about 100 days to reprogram the Banner System. However, Mr. Houck testified in April 2016 that PNM had begun the reprogramming and will be ready to go forward with the changes when new rates take effect if the Commission so orders. Tr. (4-12-16) 412-13. PNM estimates the cost of reprogramming its Banner System at $271,767, broken down as follows: Recommended Decision Case No. 15-00261-UT 63 Type of Cost Outside Services (Applications Development) Outside Services (Testing) Internal Labor Total Amount of Cost $121,081 $67,436 $83,250 $271,767 Houck 3-14-16 Supp. 10. PNM says that because it is not otherwise necessary to redesign its billing system in order to bill customers in accordance with existing rate structures and FPPCAC calculations that have been approved by the PRC, the costs of reprogramming the billing system should be deferred for future cost recovery through a regulatory asset. Ortiz 3-14-16 Supp. 36. WRA opposes PNM’s request to create a regulatory asset for the cost of changing the Banner billing system because it is not clear that the cost is an extraordinary one that requires separate treatment apart from traditional regulatory recovery. Howe 3-23-16 Response 7. As explained in Section XIX(A), regulatory asset treatment should be the exception, not the norm. PNM’s request for regulatory asset treatment should be denied. PNM’s argument that it should be allowed to defer the costs for later recovery because reprogramming its Banner System isn’t necessary reflects PNM’s repeated showing of its concern only with its bottom line and not fairness to its customers. B. FUEL HANDLING EXPENSES AND PURCHASES AND SALES OF SPINNING RESERVES PNM requests that certain fuel costs that are currently recovered in non-fuel base rates instead be recovered through PNM’s FPPCAC. PNM seeks to include its coal and nuclear fuel handling expenses and purchases and sales of spinning reserves in its FPPCAC calculations. If PNM’s request is approved, the estimated amount on a total company basis that would be added to recovery under PNM’s FPPCAC is $14.4 million. Taylor Direct 12-14. PNM records coal handling expenses under FERC Account 501, Taylor Direct 12, which defines this Account as including: [T]he cost of fuel used in the production of steam for the generation of electricity, including expenses in unloading fuel from the shipping media and handling Recommended Decision Case No. 15-00261-UT 64 thereof up to the point where the fuel enters the first boiler bunker, hopper, bucket, tank or holder of the boiler-house structure. 18 C.F.R. Pt. 101, Operation & Maintenance Expense Chart of Accounts, Account No. 501, Fuel. More specifically, coal handling expenses include (i) supervising purchasing and handling of fuel; (ii) all routine fuel analyses; (iii) unloading from shipping facility and putting in storage; (iv) moving of fuel in storage and transferring fuel from one station to another; (v) handling from storage or shipping facility to first bunker, hopper, bucket, tank or holder of boiler-house structure; (vi) operation of mechanical equipment. Id. PNM records nuclear fuel handling expenses under FERC Account 518, Taylor Direct 12, which defines this Account as including the net cost of nuclear fuel assemblies used in the production of energy. 18 C.F.R. Pt. 101, Operation & Maintenance Expense Chart of Accounts, Account No. 518, Nuclear Fuel Expense (Major Only). Spinning reserves represent electrical capacity immediately available to respond to an increase in load relative to the generation available to serve load, such as occurs when a power plant experiences a sudden forced outage. Without such reserves, the sudden loss of a large power plant on an interconnected electrical system can cause widespread blackouts. Golden Spread Elec. Coop., Inc. v. Denver City Energy Assocs., 269 S.W.3d 183, 186 (Tex. Ct. App. 2008). Spinning Reserve is the on-line reserve capacity that is synchronized to the grid and ready to meet electric demand within ten minutes of a dispatch instruction. Non-Spinning Reserve is off-line generation capacity that can be ramped to capacity and synchronized to the grid within ten minutes of a dispatch instruction and that is capable of maintaining that output for at least two hours. PNM records the cost of purchasing spinning reserves in FERC Account 555 (Purchased Power), Taylor Direct 12, which defines this Account as including the cost of spinning reserve Recommended Decision Case No. 15-00261-UT 65 capacity. 18 C.F.R. Pt. 101, Operation & Maintenance Expense Chart of Accounts, Account No. 555, Purchased Power. PNM records sales of spinning reserve in FERC Account 447 (Sales for Resale), Taylor Direct 12, which defines this Account to include the net billing for electricity supplied to other utilities for resale purposes. 18 C.F.R. Pt. 101, Income Chart of Accounts, Account No. 447, Sales for Resale. PNM excludes sales of spinning reserves from off-system sales, which are passed through its FPPCAC, to match its recovery of the cost of purchasing spinning reserves, which are not passed through its FPPCAC. Taylor Direct 12. Commission Rule 17.9.550 (Rule 550), titled “Fuel and Purchased Power Cost Adjustment Clauses for Electric Utilities,” contains a sample FPPCAC Report Form for guidance to utilities that file monthly FPPCAC reports. It identifies components of fuel and purchased power expense with lines for amounts to be filled in. The components listed on the sample form include: x x x x “Account 501 – Fuel Expense;” “Account 518 – Nuclear Fuel Expense;” “Account 555 – Purchases;” and “Account 447 – Sales for Resale” 17.9.550.20 NMAC. Therefore, under Rule 550, fuel handling expenses and the costs of purchasing and revenues from selling spinning reserves are, at a minimum, eligible to be passed through under a FPPCAC. In fact, in past years, Southwestern Public Service Company recovered its fuel handling costs through its FPPCAC, as has El Paso Electric Company. Tr. (4-29-16) 3535-3540 (Gunter); Ortiz Rebuttal 36. To actually receive authority to pass these costs and revenues through its FPPCAC, PNM must show that these costs and revenues “periodically fluctuate and cannot be precisely Recommended Decision Case No. 15-00261-UT 66 determined in a rate case.” 17.9.550.17 NMAC; 2008 PNM Fuel Clause Case, Final Order 10, ¶ 26. PNM witness Taylor testified that fuel handling costs “vary directly with the quantity of fuel used in the production of energy.” She produced a chart showing that these costs periodically fluctuate, although nuclear fuel handling costs fluctuate more widely than coal fuel handling costs. Taylor Rebuttal 7. She also produced a chart showing that the cost of purchasing spinning reserves periodically fluctuate. Taylor Rebuttal 8. PNM seeks to include its coal and nuclear fuel handling expenses and purchases and sales of spinning reserves in its FPPCAC calculations to ensure that customers pay actual amounts for these expenses and receive actual revenues for sales of spinning reserves. Taylor Direct 11-12. Staff opposes PNM’s request to include coal and nuclear fuel handling expenses and the purchases and sales of spinning reserves in its FPPCAC calculations, for four reasons. First, Staff says that in two cases, the PRC has not allowed PNM to pass through these costs and revenues through its FPPCAC. Second, Staff believes that all fuel and purchased power costs should be recovered through base rates, not through a FPPCAC. Third, Staff says that “a development in New Mexico in recent years has been to allow fewer, less volatile costs to be recovered in base fuel expense,” citing to the 2007 SPS Rate Case in which the PRC approved SPS’s request to move recovery of its purchased power capacity costs from its FPPCAC to base rates. Fourth, Staff argues that PNM has not provided a compelling reason to include coal and nuclear fuel handling expenses and purchases and sales of spinning reserves in its FPPCAC calculations. Gunter Direct 10-12. Staff’s arguments are not persuasive. First, it is inaccurate to say that in two cases the PRC did not authorize PNM to include coal and nuclear fuel handling expenses and the purchase and sale of spinning reserves in its FPPCAC. In those cases, the PRC approved stipulations in Recommended Decision Case No. 15-00261-UT 67 which the parties agreed that PNM would flow through its FPPCAC only those costs and revenues approved for flow-through by the PRC when PNM’s FPPCAC was most recently approved in Case No. 08-00273-UT. Second, the PRC does not share Staff’s belief that all fuel and purchased power costs should be recovered through base rates, not through a FPPCAC. In the 2015 EPE Rate Case, the PRC, on its own motion, ordered EPE to recover all of its fuel and purchased power costs through its FPPCAC, stating: The Commission further finds that EPE should be ordered to remove all fuel and purchased power costs from base rates and instead, bill such costs entirely through the FPPCAC. Such treatment of these costs should provide greater transparency to ratepayers with regard to the nature of the charges on their bills. In particular, separation of the highly variable fuel costs from other, more fixed costs, should assist ratepayers in understanding the factors influencing the magnitude and variability of their bill changes. Final Order 106, ¶ 242. Third, the 2007 SPS Rate Case is not precedent for “a development in New Mexico in recent years has been to allow fewer, less volatile costs to be recovered in base fuel expense.” In that case, the PRC approved SPS’s request to move recovery of its purchased power capacity costs from its FPPCAC to base rates, and, in doing so, explained that Staff supported SPS’s request in part because Staff believed that capacity costs are relatively stable. 2007 SPS Rate Case, Corrected Recommended Decision 166-67. Additionally, in a later case in which the PRC approved reinstatement of PNM’s FPPCAC, the PRC said, “Contrary to the assertions of some of the parties in this case, Rule 550 does not require a showing that purchase power and fuel costs are “volatile,” or fluctuate more than some other cost of providing service.” It was sufficient that PNM showed that its costs periodically fluctuate. Case No. 08-00092-UT, Final Order 15, ¶ 37. PNM has shown that its coal and nuclear fuel handling expenses and the cost of its purchases and sales of spinning reserves fluctuate periodically and therefore qualify under Rule 550 to be passed through its FPPCAC. PNM’s request is reasonable and should be granted. However, this recommendation should not be viewed as a conclusion that PNM is entitled to Recommended Decision Case No. 15-00261-UT 68 pass these costs and revenues through its fuel clause. Use of a fuel clause is a privilege which can be eliminated if it does not serve the goal of providing reasonable and proper service at fair, just, and reasonable rates to all customer classes. C. REVENUES FROM CHEMICAL PRETREATMENT OF COAL FOR SJGS The federal government provides tax incentives to entities that reduce NOx (nitrogen oxide) emissions by chemically treating coal before combustion. Several entities have developed proprietary processes for coal pretreatment that meet Internal Revenue Service requirements. On behalf of the SJGS owners, PNM entered into a License and Access Agreement with San Juan Fuels, LLC (SJF) under which SJF installed a pretreatment facility at the SJGS. SJF pays a licensing and access fee based on the tonnage of coal treated, of which PNM’s retail share would be about $5.6 million annually. PNM Ortiz Direct 62-63. In April 2014, when the PRC last approved PNM’s continued use of its FPPCAC, the PRC approved a stipulation allowing PNM to keep 100% of the revenues from the SJR Contract through the effective date of new rates approved in PNM’s next general rate case (this case). The PRC required PNM to include in its next general rate case (this case) a proposal for the ratemaking treatment of the revenues going forward. 2013 PNM Fuel Clause Case, Certification of Stipulation 18-19, 23, 28. In the same case, PNM agreed to forego recovery of $10.5 million of undercollected fuel and purchased power costs in its FPPCAC balancing account, in part to compensate ratepayers for PNM’s retention of 100% of the SJR Contract revenues. Id. 13, 19, 23. At that time, PNM had not yet received any revenues under the SJR Contract because SJR had not located an investor for the project. PNM was expected to begin receiving payments under the SJR Contract in January 2014 or earlier. Id. 19, 23. It was also expected that new PNM general rates would take effect in January 2016. Id. 23. The Certification of Stipulation in Case No. 13-00187-UT, citing Staff’s testimony, says: “PNM’s agreement to the $10.5 million credit is intended to compensate ratepayers for PNM’s Recommended Decision Case No. 15-00261-UT 69 retention of coal pre-treatment revenues and is not dependent on when the coal pre-treatment payments start, if ever.” Id. 23. In this case, PNM proposes that it continue to retain 100% of the SJR Contract revenues through December 31, 2016, and beginning January 1, 2017, credit 50% of the SJR Contract revenues against fuel handling costs through its FPPCAC. Ortiz Direct 63. In support of its proposal to continue to retain 100% of the SJR Contract revenues through December 31, 2016, PNM emphasizes that it did not begin to receive revenues under the Contract until much later than expected. Ortiz Direct 63. Staff recommends that 100% of the SJR Contract revenues be credited to customers. Staff argues that PNM’s purchase and use of coal at San Juan “are the fundamental activities that allow coal pre-treatment revenues to be realized in the first place, and are activities for which PNM’s New Mexico customers reimburse PNM in full through the FPPCAC.” Staff’s understanding is that PNM shareholders did not invest any capital to obtain the revenues. In response to PNM’s point that it did not receiving the revenues as soon as expected, Staff says that it is inappropriate for ratepayers “to protect PNM from risks the Company undertook, and was aware of in advance, when it entered into a settlement in a previous case.” Gunter Direct 15-16. Staff’s argument that ratepayers should receive 100% of the SJR Contract revenues because they pay for the coal that produces the revenues, is persuasive, and should be adopted. D. WRA’S RISK SHARING PROPOSAL WRA proposes that PNM’s FPPCAC be modified to include a 70%-30% risk sharing mechanism in which PNM would be allowed to pass through only 70% of any under or over recovery of its fuel and purchased power costs. PNM’s customers would be at risk for 70%, and PNM’s shareholders would be at risk for 30% of the variation in PNM’s fuel and purchased power costs. Howe Direct 30-40. Recommended Decision Case No. 15-00261-UT 70 CCAE supports WRA’s proposal. CCAE’s Initial Posthearing Brief 55-56. PNM opposes WRA’s proposal. WRA made a similar risk sharing proposal in a rulemaking in which the PRC most recently amended Rule 17.9.550 NMAC. WRA proposed that a utility be allowed to pass through only 90% of any under recovery of its fuel and purchased power costs and be allowed to keep only 25% of any over recovery. In its Final Order in the rulemaking, the PRC said: WRA . . . suggested the addition of a new term, ‘performance adjusted,’ which is essentially a revenue-sharing incentive-disincentive mechanism for fuel and purchased power cost fluctuations. That suggestion was opposed by the other commenters. The Commission agrees that WRA’s suggested mechanism would go beyond the intent of the rule. Case 07-00389-UT, Final Order 2, ¶ 6 (emphasis added). In this case, WRA has not said why the PRC should reverse its decision in Case No. 0700389-UT. Therefore, its argument should be rejected. See Mountain States Tel. & Tel. Co. v. New Mexico State Corp. Comm’n, 1986-NMSC-019, ¶ 26, 104 N.M. 36 (change in policy must be preceded by proper notice and supported by reasonable justification). XV. PALO VERDE CAPACITY A. BACKGROUND/HISTORY PNM entered into sale/leaseback arrangements of its interests in PV Units 1 and 2 in 1985 and 1986. The initial lease terms resulting from the sale/leasebacks expired in 2015 and 2016. Before the leases expired, PNM had to decide what its future relationship would be, if any, to the lease assets. PNM had to decide whether to: (1) allow its interests to terminate; (2) extend some of the PV Unit 2 leases for two years; (3) extend the PV Unit 1 leases and one PV Unit 2 lease for eight years; or (4) purchase the interests from each Lessor at FMV. Recommended Decision Case No. 15-00261-UT 71 PNM decided to extend five leases in Units 1 and 2 covering 114.6 MW for eight years and to purchase three leasehold interests in Unit 2 covering 64.1 MW. The prudence and ratemaking treatment of PNM’s decisions are at issue here. In 1977, the PRC granted PNM a certificate of public convenience and necessity (CCN) to participate in the Arizona Nuclear Power Project, known as the Palo Verde Nuclear Generation Station. PNM was granted authority to participate in PV as a tenant in common and allowed to own, operate, and maintain an undivided 10.2% interest in each of three PV units (PV Units 1, 2, and 3) together with common facilities incident to the Units. Case No. 1216, Findings of Fact and Order, Decretal ¶ C (2-8-77).23 In 1985, the PRC authorized PNM to sell and lease back substantially all of its 10.2% undivided ownership interest in PV Unit 1 to third party investors, who simultaneously leased the assets back to PNM. Case No. 1995, Order (11-27-85). In 1986, in Phase I of Case No. 2019, the PRC authorized PNM to sell its 10.2% undivided ownership interest in PV Unit 2 and the remainder of its PV Unit 1 interests to third party investors, who simultaneously leased these assets back to PNM. Case No. 2019, Phase I, Order (7-8-86). The investors purchased interests through owner trusts. Altogether, for Units 1 and 2, the PRC approved PNM sales to 11 investors. Each investor held a beneficial interest in a trust, which, in turn, held legal title to the leased assets. Case No. 2444, Recommended Decision 4 (717-92), adopted by Final Order (8-24-92). 23 At issue in this proceeding are PNM’s PV Units 1 and 2 interests. PV Unit 3 is not at issue but was an issue in the San Juan Case. PV Unit 3 provided service to PNM ratepayers from the 1988 date of commercial operation until it was excluded from PNM’s jurisdictional rates and decertified in 1990. San Juan Case, Certification of Stipulation 130-31 (4-8-15). In December 2015, the PRC granted PNM a CCN for its 134 MW Interest in PV Unit 3 to serve retail customers effective January 1, 2018. San Juan Case, Final Order. Recommended Decision Case No. 15-00261-UT 72 The Owner Trusts financed the purchases with debt and equity. About 80% of the purchase price of the PV interests was provided through debt, while the remainder was provided by the equity investors. First PV Funding Corporation initially used bank borrowings to provide the debt for the lease transactions and then refinanced the borrowings through two public offerings of its lease obligation bonds (LOBs). The Series 1986A LOBs, due 1991-2014, were issued with interest rates ranging from 8.3% to 10.3%, depending on maturity. The Series 1986B LOBs, due 1992-2016, were issued at interest rates ranging from 8.05% to 10.15%, depending on maturity. Eden 5-25-16 Supp. 5. The debt was payable from lease payments due from PNM under the leases. Each owner trustee, as Lessor, leased the interest in the Unit back to PNM under a separate lease agreement, having an initial lease term for the Unit 1 leases expiring on January 15, 2015, and for the Unit 2 leases expiring on January 15, 2016. PNM made lease payments to the Indenture and Collateral Trust, which provided an equity return to the Owner Trust and made debt service payments to bondholders, which ultimately flowed through to the Owner Trusts’ equity investors and supported contributions to a sinking fund that provided a “return of” capital to the lenders. In return for the lease payments, PNM received the right to PV power. Case No. 2444, Recommended Decision 5; Eden 5-25-16 Supp. 4. The lease payments were a function of the interest rates payable on the debt, among other things. Case No. 2019, Phase I, 4, ¶ 6. The equity investors’ returns on equity and implied capital structures were not disclosed to PNM. Eden 5-25-16 Supp. 5. PNM received about $900 million in total for its PV Units 1 and 2 interests, which provided $644 million for PNM’s construction costs plus $256 million in profit. Eden 5-25-16 Supp. 5. PNM recognized an after-tax gain of about $5.5 million on its Unit 1 interests and $37 million on its Unit 2 interests. Case No. 2262, Final Order 43. The PRC ordered those gains to be amortized back to ratepayers over a 15-year period. Case No. 2019, Phase I, 11, ¶ 14. Recommended Decision Case No. 15-00261-UT 73 Ratepayers were credited with those gains via reductions to O&M expense and rate base. Tr. (627-16) 3909-10 (Monroy); Tr. (6-28-16) 3951 (Harland). The PRC rejected PNM’s later request to increase the amortization period to the life of the leases. Case No. 2262, Final Order 43-44. The purpose of sale/leaseback transactions was to levelize the rate impact of PV. Under an ownership scenario, the revenue requirement associated with the capital costs of a major plant addition such as PV is very high initially and then declines as the plant depreciates. In contrast, under a lease scenario, the lease payments, at least initially, are lower than the revenue requirement of ownership. “Thus, by leasing PV, rate shock was mitigated when the new capital intensive PV plant entered service.” Case No. 2444, Recommended Decision 5. The thenpresent value of the ratepayer revenue requirement of the facilities was projected to be significantly lower under the lease transactions than under PNM ownership over 30 years. Case No. 2146, Part II 12; Case No. 2262, Final Order 13. The owners of the PV leases were able to capitalize the investment with more debt financing than PNM and more fully use tax benefits transferred to them under the sale/leaseback transactions. Ortiz 5-25-16 Supp. 10. “For ratemaking purposes, the lease payments [were] equated with capital costs.” The lease payments were based on the sales price of the units. Case No. 2146, Part II, Final Order 12; Case No. 2262, Final Order 13-14. Since Case No. 2262, or roughly the past 30 years, the costs of the PV leases have been recovered in base rates. PNM used the proceeds from each of the sale/leaseback transactions (1) to repay utility obligations; (2) for stock repurchases; (3) for tax payments; and (4) for maintenance of utility service. Eden 5-25-16 Supp. 5; Exh. EAE-2 to Eden 5-25-16 Supp. A trio of cases between 1987 and 1990 addressed the initial ratemaking treatment of the PV units. Rather than filing a rate case, PNM filed Case No. 2146 on August 14, 1987, in an attempt to reorganize itself into separate generation and distribution subsidiaries, which would then Recommended Decision Case No. 15-00261-UT 74 subject its rate regulation to FERC instead of the PRC. PNM eventually withdrew its request, but the PRC established a second phase of the proceeding to consider the PV units in the larger context of PNM’s excess capacity problems. In Case No. 2146, Part II, the PRC considered PNM’s excess capacity problems. The PRC ruled that ratepayers and investors must share the economic consequences caused by too much capacity. Final Order 58 (4-5-89). The PRC also framed the issues to be considered in PNM’s pending rate case, Case No. 2262. Among other things, the PRC: x Found that PV Units 1 and 2 were not used and useful, but nevertheless conditionally included the Units in PNM’s base rates, subject to the outcome of Case Nos. 2087 and 2262. Final Order 109, 116-117, Decretal ¶ I x Excluded PV3 from rate base and ordered PNM to propose appropriate treatment of PV Unit 3 in Case No. 2262, including any proposal for decertification and abandonment. Final Order 109, 116-117, Decretal ¶¶ F, I x Excluded 130 MW of SJ Unit 4 from rate base and stated that PNM may apply to include this excluded portion of SJ Unit 4 in rate base when it was no longer part of PNM’s excess capacity. Final Order, Decretal ¶ G x Disallowed 105 MW of a purchased power agreement that PNM had with M-S-R. Public Power Agency24 In Case No. 2087, the PRC considered the prudence of PNM’s investment in PV. The PRC approved a stipulation which resulted in a $90 million disallowance from PNM’s rate base, comprised of $80 million of inventory costs and $10 million associated with PV Units 1 and 2. Case No. 2087, Final Order (3-6-90). “Effectively, under the Final Stipulation, all of the energy from PNM’s share of PVNGS Units 1 and 2 would serve PNM customers, but at a reduced cost due to this disallowance.” Final Order 40. In Case No. 2285, the PRC issued an order granting abandonment and decertification of PNM’s interest in PV Unit 3. In Case No. 2296, the PRC issued an order denying PNM’s application for abandonment and decertification of SJ Unit 4, which was affirmed by the New Mexico Supreme Court. Public Serv. Co. of New Mexico v. New Mexico Pub. Serv. Comm’n, 1991-NMSC-083, 112 N.M. 379. Thus, while SJ Unit 4 was not included in PNM’s rate base, it continued to be available to serve PNM retail customers. 24 Recommended Decision Case No. 15-00261-UT 75 Case No. 2262 was a rate case filed in January 1989, after the conclusion of Cases Nos. 2146 Part II and 2087, in which the PRC reduced PNM’s then-existing rates by $2,887,382, adopted the result of Case No. 2087, and included PV Units 1 and 2 lease payments in base rates. Case No. 2262, Final Order (4-12-90). In 1992, in Case No. 2444, the PRC authorized PNM to purchase, for $16.75 million, the beneficial interests in the Owner Trusts associated with leases of 58 MW (29 MW in each of Units 1 and 2) as part of the bankruptcy reorganization of Drexel Burnham Lambert Group, Inc. (Drexel) (the Drexel Assets). Final Order (8-24-92), adopting Recommended Decision (7-1792). The leases were originally held by Burnham Leasing Corporation (BLC), a subsidiary of Drexel. PNM acquired the equity interests at substantially less than the original price paid by BLC due to the bankruptcy reorganization. Upon collapse of the trust holding the Drexel Assets and termination of the lease in 1998, the ownership interest in the assets transferred to PNM. Ortiz 5-25-16 Supp. 10; Eden 5-25-16 Supp. 9. In Case No. 2444, PNM represented that, as a result of the purchase, it could lower customer rates by about $1.8 million annually beginning in 1999 and continuing through 2015, and $0.9 million in 2016. PNM proposed to lower rates once the revenue requirement of owning the Drexel Assets became lower than the revenue requirement would have been under continued leasing. Case No. 2444, Recommended Decision 8-9. The Recommended Decision adopted by the PRC says: At the end of the respective leases, the 58 MW shall be brought into rate base at zero value. PNM will not seek to include in rate base the unamortized portion of the acquisition adjustment. PNM can, however, include in rate base any just, reasonable and prudent costs for betterments and improvements related to the 58 MW as well as any appropriate expenses. Recommended Decision 10-11 & 18, ¶ G. In Case No. 2567, the PRC approved a stipulated $30 million rate reduction based on PNM’s $180 million write down, on a pre-tax basis, in 1993, of assets. The stipulation approved Recommended Decision Case No. 15-00261-UT 76 in that case included provisions for restructuring or refinancing PV Units 1 and 2 leases upon demonstrated net benefits, with any cost of service reductions allocated 60% to shareholders and 40% to customers. Case No. 2567, Final Order (11-28-94). In Case No. 2567, the PRC approved a provision in the stipulation stating that PNM’s interests in PV Units 1 and 2 were used and useful and appropriately included in rates. Id. at 73, ¶ DD. However, the PRC made clear that this used and useful declaration was not “indefinite or open-ended.” Id. at 50. The PRC explained: [T]he conferral of “used and useful” status on PV Units 1 and 2 at this juncture will not give these units any more insulation from the normal vicissitudes of economics, regulation, or the other circumstances to which they are subject, than they now have. For example, PNM, Staff, the AG and the other signatories agree that the provision under consideration would not serve to preclude future reconsideration or redetermination by the Commission of the costs of these PV units in PNM’s rates. . . . .... In sum, we can give no more assurances on the future ratemaking treatment of PV Units 1 and 2 than we can for any other utility asset in rate base. . .. .... The foregoing “used and useful” finding should only remain in force until the Stipulation is superseded or otherwise terminated. If PNM desires at that time to have the Commission declare that PV Units 1 and 2 are then currently used and useful, PNM will have the burden of proof. Id. at 50-52; see also id. at 73, ¶ DD. In Case No. 2567, the PRC also approved a transition mechanism for PNM to repurchase the PV assets from the Lessors to facilitate a sale to another owner, even at a loss, to reduce costs and avoid decommissioning expense. In describing the transition mechanism, the PRC relied on the testimony of Staff witness Gee in which Mr. Gee described a hypothetical sale of PNM’s repurchased Drexel Burnham share of PV Units 1 and 2, which had been written down to $805/KW. If PNM sold this capacity at $700/KW, the $105/KW difference would remain in PNM’s rates as a regulatory asset and would be paid for by PNM’s customers over the remaining life of that asset. Id. at 61. Again relying on Mr. Gee’s testimony, the PRC said that the purpose Recommended Decision Case No. 15-00261-UT 77 of the transition mechanism was “to give PNM some assurance that additional write-offs associated with PVNGS capacity would not occur in the event that it became beneficial to stockholders and ratepayers to sell such capacity at a loss.” Id. at 62. The PRC explained that, according to Mr. Gee, the signatories agreed that getting rid of some PV, regardless of whether included or excluded, would be beneficial because of the reduction or elimination of potential risks associated with nuclear plant decommissioning and operation, “not to mention costs.” Id. at 63. The PRC specifically approved the transition mechanism because it could improve PNM’s standing with the financial community and result in possible ratepayer benefits. Id. B. PNM’S CURRENT INTERESTS IN PV UNITS 1 AND 2 Under each of the PV sale/leaseback agreements, PNM had three options upon expiration of the initial lease period: 1. Allow the lease to expire at the end of the initial lease term and relinquish control of the Leased Assets 2. Renew the lease with a 50% reduction in lease payments (some leases could be extended for two years and some for eight years) or 3. Purchase the lease assets for FMV at end of initial lease term Tr. (6-27-16) 3726-27 (Ortiz); Eden 5-25-16 Supp. 6. ; Eden 5-25-16 Supp. 7. Under the sale/leaseback agreements, PNM was to provide each Lessor with two notices before the initial lease terms expired. The First Notice was due three years bef0re expiration and the Second Notice was due two years before expiration. Because PNM had already repurchased the Drexel Burnham and the First Chicago Interests, the First and Second Notice provisions for these Leases did not apply. In the First Notice, PNM was required to irrevocably elect to terminate or continue its use of the PV plant. The Second Notice only applied if PNM committed in the First Notice to continue to use the PV plant. In the Second Notice, PNM was required to state how it wanted to Recommended Decision Case No. 15-00261-UT 78 continue to use the PV plant: either (1) extend the Lease; or (2) purchase the Lessor’s interest and own the PV plant. Both the First and Second Notices are irrevocable to PNM. If PNM elected to purchase the plant, the Lessor and PNM had a period of time in which to negotiate a fair market sales value, or, if they could not agree, an appraisal process would determine the FMV. Eden 5-25-16 Supp 6. The initial terms of PV Unit 1 Leases expired in January 2015. Therefore, PNM had to provide its First Notice in January 2012. On January 6, 2012, PNM gave irrevocable notice to four PV Unit 1 Lessors that it would continue to use PV Unit 1 lease assets. This meant that PNM relinquished its opportunity to terminate its relationship to these PV Unit 1 interests at that time. On January 9, 2013, PNM gave irrevocable notice to the Unit 1 Lessors that it would renew the leases for an additional eight years at 50% of then-current lease payments, until January 2023. The initial term of the PV Unit 2 Leases expired in January 2016. Therefore, PNM had to provide its First Notice in January 2013. On January 9, 2013, PNM gave irrevocable notice to all of the PV Unit 2 Lessors that it would continue to use the PV Unit 2 lease assets. This meant that PNM relinquished its opportunity to terminate its relationship to these PV Unit 2 interests at that time. On December 30, 2013, PNM notified JPMorgan Chase that it would extend the PV Unit 2 lease representing 10.4 MW at 50% of then-current lease payment for eight years, until January 2024, reducing its lease payments by $1.6 million annually. Eden Direct 25; Exh. DVW-14 to Van Winkle Direct. On January 13, 2014, PNM notified the Lessors of the other three Unit 2 leases — totaling 64.1 MW — that it would exercise the FMV purchase options specified in the leases rather than extend the leases for two years. PNM negotiated agreements with each lessor regarding the purchase price for each lease. Eden Direct 26. Recommended Decision Case No. 15-00261-UT 79 On February 24, 2015, PNM entered into a letter agreement with CGI Capital, Inc., specifying a FMV for 31.25 MW of generating capacity at Unit 2 of $78.2 million or $2,500/kW as of the end of the original lease term, January 15, 2016. On May 1, 2015, PNM entered into a letter agreement with Cypress Verde LLC and Cypress Second PV Partnership, specifying a FMV for 32.76 MW of generating capacity at Unit 2 of $85.2 million or $2,600/kW as of the end of the original lease terms, January 15, 2016. Eden Direct 26-27. In total, PNM paid about $163.5 million for the 64.1 MW. Tr. (6-28-16) 3990 (Eden). PNM purchased the 64.1 MW of PV Unit 2 interests on January 15, 2016. Peters Direct 13-15. The leases have terminated, and PNM now owns these PV Unit 2 interests. Eden 5-25-16 Supp. 8. PNM’s current PV interests are comprised of a combination of direct ownership and leasing arrangements. PNM has ownership interests of 2.3% in Unit 1, 9.4% in Unit 2, and 10.2% in Unit 3, and has leasehold interests of 7.9% in Unit 1 and 0.76% in Unit 2. Eden Direct 23. At issue in this case are the following leasehold and ownership interests: Lessor MW Chrysler Financial Corp. MFS Leasing Corp. Chase Manhattan Realty Leasing Corp. Chase Manhattan Realty Leasing Corp. 49.1 PV UNIT 1 -- LEASEHOLD INTERESTS Expiration Expiration Annual of Initial of Renewed Lease Lease Term Lease Term Payment Before Extension 2015 2023 $15,693,862 Annual Lease Payment After Extension $7,846,931 17.9 2015 2023 $5,580,122 $2,790,061 14.9 2015 2023 $4,757,770 $2,378,885 22.3 2015 2023 $6,974,314 $3,487,157 Eden Direct 25; Monroy 5-25-16 Supp. 8-9. Recommended Decision Case No. 15-00261-UT 80 PV UNIT 2 -- 10.4 MW OF LEASEHOLD INTERESTS AND 64.1 MW OF OWNERSHIP INTERESTS Lessor MW Expiration of Initial Lease Term Expiration of Renewed Lease Term Annual Lease Payment After Lease Extension 2018 Annual Lease Payment Before Lease Extension $9,246,447 Citigroup (CGI Capital) Cypress Cypress Chase Manhattan Realty Leasing Corp. 31.3 2016 14.9 17.9 10.4 2016 2016 2016 2018 2018 2024 $4,403,887 $5,331,503 $3,038,572 N/A N/A $1,636,280 N/A Eden Direct 25; Monroy 5-25-16 Supp. 9. C. PNM’S RESPONSIBILITIES FOR O&M, LEASEHOLD IMPROVEMENTS, AND DECOMMISSIONING EXPENSE 1. GOVERNING DOCUMENTS Several documents govern PNM’s participation in PV. These include, among others: a. The Arizona Nuclear Power Project Participation Agreement, as amended, entered into among the PV Participants. The original Participants were PNM, Arizona Public Service Company, Salt River Project Agricultural Improvement and Power District, Tucson Gas and Electric Company, and El Paso Electric Company. The Arizona Nuclear Project Participation Agreement establishes the rights and duties of the various owners of interests in the PV nuclear plant. PNM is a party to the Agreement through its ownership interests in PV Units 1, 2, and 3. b. The Sale/Leaseback agreements applicable to each of the Leased Assets, as amended. Under the Sale/Leaseback Agreements, PNM as Lessee is solely responsible for all costs associated with the underlying assets, including lease payments, capital investments, O&M expenses, and decommissioning liabilities. When a lease expires, PNM continues to be Recommended Decision Case No. 15-00261-UT 81 responsible for decommissioning expenses and any capital project costs for projects pending at the date of the lease expiration. Eden 5-25-16 Supp. 23. 2. LEASE PAYMENTS During the periods of time when PNM has been a Lessee under the sale/leaseback agreements, PNM has not included the value of the Lease Assets (excluding improvements) in rate base but has treated the amount of its lease payments as an O&M expense. Tr. (6-27-16) 3894 (Monroy). As explained above, for ratemaking purposes, the lease payments have been equated with capital costs. 3. O&M PNM, as Lessee, is and was responsible for all O&M expenses of the Leased Assets in proportion to its Generation Entitlement Share, which is 10.2%. Case No. 2019, Phase I, Final Order 3, ¶ 6; Tr. (6-28-16) 4131 (Eden); Arizona Nuclear Power Participation Agreement, ¶ 11.2. 4. LEASEHOLD, COMMON PLANT, CAPITAL IMPROVEMENTS & DEPRECIATION PNM, as Lessee, was and is responsible for the cost of leasehold improvements to the Leased Assets and associated Common Plant in proportion to its Generation Entitlement Share. Tr. (6-28-16) 4131 (Eden); Arizona Nuclear Power Participation Agreement, ¶ 18.3. For all of the leasehold improvements paid for in part by PNM, title or ownership to the improvements has belonged to the Lessors when such improvements were installed. Tr. (6-2816) 4077, 4122-23 (Eden). PNM said that if it did not extend a lease or purchase Leased Assets at the end of the initial lease term, title to and possession of improvements would remain with the Lessor. NEE Exh. 30. PNM has always included improvements to its Leased Assets in rate base (including improvements to its share of common plant), has earned a return of and on these assets, and has depreciated these assets according to PRC-approved depreciation rates for each respective time period. Tr. (6-29-16) 4291-93 (Peters). PNM has used the length of the NRC-operating licenses Recommended Decision Case No. 15-00261-UT 82 as the service life of the PV improvements for purposes of calculating depreciation rates. The depreciation rates in effect in 1985 and 1986 were based on a 40-year remaining life. In the 2008 PNM Rate Case, the PRC approved new PV depreciation rates incorporating a life extension to 2046 to coincide with the NRC’s extension of the operating licenses. Peters 5-25-16 Supp. 22-23, 31; Tr. (6-29-16) 4220 (Peters). PNM tracks PV improvements by Unit, but not by lease. Therefore, to calculate the value of improvements associated with the Lease Assets of a single lease, PNM divides the value of the total improvements for a Unit by the megawatts of capacity associated with the Lease Assets associated with a lease. Tr. (6-27-16) 3879 (Monroy). 5. DECOMMISSIONING Under the sale/leaseback agreements, PNM is responsible for paying decommissioning costs of PV in proportion to its Generation Entitlement Share even if PNM relinquishes its interests in the Leased Assets. Tr. (6-28-16) 4131 (Eden). Therefore, PNM is already obligated for decommissioning PV units 1 and 2 whether it continues its participation with the PV units. Exh. 12 to Van Winkle Direct. D. PRESERVATION OF PRC’S RATEMAKING AUTHORITY INCLUDING RECOVERY OF PLANT IMPROVEMENTS AND RECOVERY OF DECOMMISSIONING COSTS PNM argues that PRC orders relating to PV Units 1 and 2 indicate that PNM would be entitled to recover the full purchase price for PV Unit 2 interests if PNM were to purchase the interests at their FMV. PNM Supplemental Brief 7-8. These orders granted PNM authority to purchase the PV Unit 2 interests at FMV, but the orders did not dictate or approve the ratemaking treatment of any such purchases. The orders also did not rule on the question of whether such a purchase would be reasonable and prudent in light of the facts present at the time of a future purchase decision. In its Final Order in Case No. 1995, the PRC said with respect to its approval of the sale/leasebacks for PV Unit 1: Recommended Decision Case No. 15-00261-UT 83 PNM is hereby granted authority to exercise its options to renew the Leases or any of the Leases and to repurchase all or any portion of the facilities in accordance with the terms of the Leases at the FMV of the Facilities at the time of such renewal or repurchase. Final Order 21, ¶ C. The PRC, however, also preserved its full ratemaking authority: All other issues of ratemaking treatment for the Lease Transactions and whether the Commission’s Final Order in Case No. 1804, which established Inventorying, needs to be amended to reflect the approval of the Lease Transactions, should be determined in a separate case which PNM shall file at least 120 days prior to the date that it needs to obtain a final order from the Commission. The Commission retains full authority over the Facilities, and over all issues of ratemaking treatment for the lease payments, the costs of and any gains or losses from the sale and leaseback concerning said Facilities, including the authority to disallow any or all of the lease expenses and transaction costs on a used-and-useful basis, on the basis of imprudency in the cost of the Facilities, or on any other lawful basis, and the approval of the Lease Transactions granted by this Order is contingent on the Commission’s retention of such full authority[.] Id. at 7, ¶ 24. Similarly, in its Final Order in Case No. 2019, Phase I, approving the sale/leasebacks for PV Unit 2, the PRC said: The Company [PNM] is hereby granted authority to renew the Leases or any of the Leases and to repurchase all or any portion of the Facilities as may be permitted or required by the terms of the Leases. Final Order 8, ¶ 3. Again the Commission preserved its full ratemaking authority: Except as expressly determined herein, nothing contained herein shall be considered as a determination by the Commission of the value of any of the Company’s properties, the justness or reasonableness of any costs or expense incurred by the Company, the propriety of including any item in the Company’s cost of service or any ratemaking determination. The Commission retains full authority over the Facilities, and over all issues of ratemaking treatment for the lease payments, the costs of and any gains or losses from the sale and leaseback concerning said Facilities, including the authority to disallow any or all of the lease expenses and transaction costs on a used-and-useful basis, on the basis of imprudency in the cost of the Facilities, or on any other lawful basis, and the approval and authorization of the Lease Transactions granted by this Order is contingent on the Commission’s retention of such full authority. Id. at 12, ¶ 16. Recommended Decision Case No. 15-00261-UT 84 E. SUMMARY OF THE PARTIES’ POSITIONS 1. PNM PNM requests approval of an acquisition adjustment to include the 64.1 MW in rate base in the amount of the $163.5 million purchase price, as depreciated through September 30, 2016, at a value of $2,550/kW. It requests authority to separately include in rate base the leasehold improvements for plant additions made during the lease period. It also requests recovery of lease expenses for the five extended PV leases. PNM’s Supplemental Brief 59. 2. WRA WRA does not support a $2550 valuation for PV. WRA does, however, believe that PV is an important and valuable resource. WRA’s Supplemental Brief. 3. NEE NEE recommends that the PRC disallow PNM’s request to recover from ratepayers the cost of the 114 MW of lease extensions and the acquisition of the 64.1 MW. NEE’s Supplemental Brief 29. 4. CITY/COUNTY The City/County assert that the rate base value of the 64.1 MW of PV Unit 2 should be no more than $1,306 per kW, which they say is justified by capital expenditures and lease payments that have been made and paid for by ratepayers over the years. They argue that extending the leases for two years instead of purchasing the Leased Assets would have saved consumers about $20 million over the two-year period. The City/County take no position on whether the purchase was an arm’s length transaction or whether the 64.1 MW is used and useful. City/County’s Supplemental Brief 7. 5. ABCWUA ABCWUA recommends: (1) denying PNM’s requested acquisition adjustment; (2) including the 64.1 MW of PV Unit 2, including associated capital improvements and common Recommended Decision Case No. 15-00261-UT 85 plant, in rate base at a zero value; and (3) excluding from PNM’s cost of service the lease payments for the extended leases and associated capital improvements, common plant, depreciation, and property taxes. ABCWUA’s Supplemental Brief 13. ABCWUA argues that PNM’s decision to extend the PV Unit 1 and Unit 2 leases was unreasonable. It recommends that, because of questions about PNM’s actual (not book) gains from its PV Unit 1 and 2 sales, the PRC should require PNM to detail and support: (1) its calculations of actual after-tax sale proceeds; (2) taxes it paid on the sales; (3) the amount of after-tax sale proceeds that have been returned to ratepayers; (4) when the returns were made; (5) the amount of after-tax sale proceeds that remain to be returned; and (5) when it intends to return the remaining sales proceeds to ratepayers. Id. at 10. ABCWUA also recommends that before allowing PNM to recover through rates costs related to the 64.1 MW of PV Unit 2 or the extended leases, PNM should be required to (i) consider whether there are better alternatives; (ii) grant future case participants reasonable access to the inputs and assumptions that PNM uses when it runs the Strategist® model; and (iii) conduct a reasonable number of Strategist® runs on behalf of such participants. Id. at 13. 6. AG The AG recommends denying PNM’s request for an acquisition adjustment and, instead, including the 64.1 MW in rate base at a NBV of $39,086,062 or $610 per kW. Alternatively, the AG says that if the PRC determines that a 30-year depreciation period should be used to calculate the NBV of the 64.1 MW, then it would be appropriate to include the 64.1 MW of capacity in rate base at a zero value. Regardless of valuation, the AG says that common plant and capital improvements “would continue to be reflected in rate base as they have been in prior cases.” AG’s Supplemental Brief 16. 7. NMIEC Recommended Decision Case No. 15-00261-UT 86 NMIEC recommends that the PRC adopt a reasonable ratemaking value for the 64.1 MW. NMIEC suggests that a reasonable approach would be to deny PNM’s request for an acquisition adjustment and include the 64.1 MW in rate base at PNM’s revised NBV calculation $1,306/kw. NMIEC recommends excluding the unamortized portion of the leasehold improvements associated with the 64.1 MW (i.e., $44.6 million) from PNM’s revenue requirement. NMIEC believes that it is important for the PRC to make two separate findings on these issues to put PNM on notice that double recovery of leasehold improvements will not be allowed in the event of any future PV sale/leaseback repurchases. NMIEC’s Supplemental Brief. 8. CFRE CFRE opposes PNM’s request for recovery of the cost of purchasing the 64.1 MW. CFRE’s Posthearing Response Brief 21. 9. STAFF Staff’s position is unknown. Staff filed supplemental testimony, but at the supplemental hearing, Staff’s testimony was stricken and withdrawn. Staff did not file a supplemental brief. F. PNM’S DECISIONS TO EXTEND THE LEASES AND TO REPURCHASE THE 64.1 MW WERE NOT PRUDENT 1. PRUDENCE To be included in rates, expenditures on utility plant must (1) have been prudently incurred; and (2) be used and useful. Case No. 2146, Part II, Final Order 53; Accounting for Pub. Utils., § 4.03. The prudent investment theory provides that ratepayers are not to be charged for negligent, wasteful or improvident expenditures, or for the cost of management decisions which are not made in good faith. “In other words, ratepayers are not expected to pay for management’s lack of honesty or sound business judgment.” Case No. 2146, Part II, Final Order 50 (4-5-89). Recommended Decision Case No. 15-00261-UT 87 A utility only receives a profit on “prudent investments at their actual cost when made . . . [and is] limited to a standard rate of return. . . .” Duquesne Light Co. v. Barasch, 488 U.S. 299, 309 (1989). “Prudence is that standard of care which a reasonable person would be expected to exercise under the same circumstances encountered by utility management at the time decisions had to be made. In determining whether a judgment was prudently made, only those facts available at the time judgment was exercised can be considered. Hindsight review is impermissible. Imprudence cannot be sustained by substituting one’s judgment for that of another. The prudence standard recognizes that reasonable persons can have honest differences of opinion without one or the other necessarily being ‘imprudent.’” Case No. 2087, Order on Burden of Proof and Specific Issues to be Addressed (10-4-98). PNM’s burden of proving that its proposed changes in rates are just and reasonable requires PNM to prove the amount of its prudent investment in PV to earn a return on that investment. PNM witness Ortiz acknowledged that the PRC, in Case No. 2019, Phase I, retained authority to disallow costs related to the PV Assets on the basis of imprudence. Tr. (6-27-16) 3728. PNM admits that it has the burden of proof to show that its decision to repurchase the 64.1 MW was prudent. PNM’s Initial Posthearing Brief 58. PNM’s decisions to extend the five PV leases and purchase the 64.1 MW PV were imprudent because it failed to show by a preponderance of the evidence that it (i) reasonably examined alternative courses of action and that its decisions to extend the leases and purchase the 64.1 MW were its most cost effective resource choices; and (ii) adequately and timely notified the PRC of its decisions regarding PV Units 1 and 2. 2. FAILURE TO CONSIDER ALTERNATIVES The preponderance of the evidence shows that when PNM decided to extend the leases and purchase the 64.1 MW, it only considered whether to extend the leases or purchase the Lease Assets. There is no evidence that PNM adequately considered ending its use of PV to Recommended Decision Case No. 15-00261-UT 88 provide service to retail ratepayers following the initial expiration of the Leases or that PNM considered alternatives to extending the leases or purchasing the Lease Assets. A December 2013 Memo from PNM management to the Finance Committee/Board of Directors that recommended purchasing the leases says: ALTERNATIVES CONSIDERED: PNM considered two alternatives: x Renewing all the leases (10 MW Unit 2 lease to 2024 and the other three Unit 2 leases to 2018), and x Buying all four Unit 2 leases at January 2016. Exh. DVW-14 to Van Winkle Direct; Tr. (6-27-16) 3671-72 (Ortiz). This is consistent with PNM’s repeated statements that retention of all of its PV assets “has been an integral part of PNM’s long-term planning since the inception of the sale-leaseback arrangement.” It has “repeatedly and consistently been analyzed as an existing resource whose retention and continued use is a necessary part of PNM’s generation portfolio to meet its system load.” Ortiz Rebuttal 23. “It was always contemplated that PNM would reacquire the fee ownership of its certificated interests in Palo Verde in accordance with the terms of the leases.” Eden Rebuttal 7. PNM did not consider the possibility that the PRC would not include the price for the 64.1 MW in its rate base. Tr. (6-27-16) 3725-26 (Ortiz). Ms. Eden admitted that PNM’s strategy was to retain PV capacity: She explained that PNM developed a strategy to purchase leases with short renewal periods through purchases, and maintain its option to purchase the leases with longer renewal periods by renewing the leases. Consistent with this strategy, PNM extended the terms of five PV Unit 1 and 2 leases representing 114 MW for eight years and purchased the three remaining Unit 2 leases representing 64 MW. Eden Direct 24. PNM did not provide any quantitative analysis in this case demonstrating the benefits of extending the five PV leases over alternatives. Tr. (6-27-16) 3645, 3818 (Ortiz). Recommended Decision Case No. 15-00261-UT 89 PNM performed no Strategist runs, economic modeling, or financial analysis to determine whether purchasing the PV Unit 2 interests was its most cost-effective resource option with respect to the PV Unit 2 interests. Exh.DVW-15 to Van Winkle Direct; NEE Exh. 12; Dauphinais Direct 11-12; Exh, JRD-1. Until its decisions to extend the leases and purchase the 64.1 MW were challenged, the only evidence that PNM submitted to support the prudency of its decisions was testimony that PNM believed that extending the five leases was a better alternative than purchasing those Lease Assets and that purchasing the 64.1 MW was a better alternative than extending those leases. Tr. (4-11-16) 103 (Ortiz); Ortiz Rebuttal 19, 25; Eden Direct 27-28. PNM’s analysis of which of two available alternatives to pursue to retain an existing resource, without consideration of alternatives to retaining the existing resource, is not only insufficient but lacking even with respect to analysis of whether to continue leasing or purchase the PV assets. PNM performed no analysis of the revenue requirement impact of buying versus continuing to lease the 64.1 MW. PNM explained: This analysis is entirely dependent on the assumed FMV purchase price for the end of the two year lease renewal term on January 15, 2018. This speculation on future FMVs severely limits the usefulness of the analysis and, therefore, PNM has not performed this analysis. Exh. 12 to Van Winkle Direct. PNM also said that it did not calculate the revenue requirement of continued leasing of the 64.1 MW over the life of PV. NEE Exh. 17. At the June 2016 hearing, however, PNM witness Mr. Monroy agreed that, analyzed over a two-year horizon, extending the three leases comprising the 64.1 MW would have saved ratepayers roughly $10.5 million in the first extended year and roughly $10.5 million — perhaps less — in the second extended year, compared to ownership. Tr. (6-28-16) 3923. While PNM denies that it purchased the 64.1 MW to increase its rate base and earnings, PNM told its Board of Directors that it was a factor, demonstrated by the statement in the Recommended Decision Case No. 15-00261-UT 90 December 2013 Memo from PNM Management that “[p]urchasing the other three Unit 2 leases will increase rate base, allowing shareholders to earn a return on the assets.” Exh. DVW-14 to Van Winkle Direct. In fact, PNM had an incentive to retain its interests in PV Units 1 and 2. PNM will continue to be responsible for decommissioning costs of PV Units 1 and 2 even if PNM had relinquished its rights to the units and the lessors/investors sold the units to a third party. PNM would also be responsible for the capital project costs on projects pending at the date of the lease expiration. Eden 5-25-16 Supp. 21. PNM did not analyze whether, and to what extent, it would be responsible for the cost of any pending capital projects if it had relinquished its interests in the Lease Assets that it purchased or the Lease Assets subject to the extended leases. Tr. (6-2916) 4155-56 (Eden). PNM argues that whether it considered alternative generation resources to the 64.1 MW of PV Unit 2 is irrelevant because “[t]his is not a resource planning or a resource acquisition case.” It says that Intervenors’ arguments that it should have considered other generation resources “are erroneously premised on a new resource acquisition portfolio under current conditions.” Ortiz Rebuttal 28. Nevertheless, in response to arguments that PNM did not consider alternatives other than extending the PV Unit 2 leases or purchasing the Lease Assets, PNM witness Ortiz identified the following benefits of selecting PV generation over other sources of generation: 1. 2. 3. PV is a zero emissions plant. PV has a strong performance record. PV is PNM’s lowest cost resource from an economic dispatch perspective. Ortiz Rebuttal 19; Tr. (6-27-16) 3637-38. Additionally, PNM argues that the results of the analysis in its 2011 Integrated Resource Plan (IRP) show that its decisions to extend the five PV leases and purchase the 64.1 MW were prudent. PNM witness Ortiz said that PNM’s 2011 IRP “was the relevant IRP when we were Recommended Decision Case No. 15-00261-UT 91 making the decision to maintain control of PV 1 and 2 in its entirety[.]” Tr. (6-27-16) 3639-40; see also Tr. (6-28-16) 4062 (Ms. Eden saying that PNM relied on the results of its 2011 IRP in deciding to extend the leases). NMIEC argues, for the following reasons, that the 2011 IRP analysis does not show that PNM’s January 2013 and January 2014 decisions to keep control of, and then purchase, the 64.1 MW were PNM’s lowest reasonable cost resource option regarding the 64.1 MW: 1. The 2011 IRP analysis was completed over two years before PNM’s January 2014 decision to purchase the leases at FMV and therefore was stale. 2. The 2011 IRP did not consider, or model, acquiring 178 MW of PV lease acquisitions in 2018 as part of PNM’s most cost-effective resource portfolio. Rather, the 2011 IRP assumed that PNM would renew the leases at then current lease prices through 2030. PNM did not perform a run that allowed Strategist to determine whether abandoning the leases, renewing the leases or purchasing the leases was the most cost effective option. Dauphinais Direct 13-15. PNM’s 2011 IRP does not meet PNM’s burden of proof to show prudence for several reasons. First, PNM’s inclusion in its IRPs of all of its PV interests as part of its most cost effective resource portfolio does not establish that, if PNM needs the capacity, the five extended leases and the 64.1 MW of PV2 capacity are the best resource to meet that need. ABCWUA’s Initial Brief 8. The PRC does not approve IRPs. The most affirmative action that the PRC takes in response to an IRP is accept the IRP as compliant with the PRC’s IRP Rule. 17.7.3.12 NMAC. The PRC did not even do this much in response to PNM’s 2011 IRP. Tr. (6-27-16) 3644 (Ortiz). Protests were filed in August 2011 to PNM’s 2011 IRP, and the proceeding established to review the protests was closed without prejudice effective October 18, 2013, when no party filed a response to the PRC’s Notice of Proposed Dismissal. Case No. 11-00317-UT, Notice of Proposed Recommended Decision Case No. 15-00261-UT 92 Dismissal, Case No. 11-00317-UT. The proceeding was closed without the PRC even accepting the IRP as compliant with the Commission's IRP rule. Second, PNM’s IRPs are not evidence in this case. Tr. (6-27-16) 3640-41 (Ortiz). PNM claims that it conducted sensitivity analyses in the preparation of its 2011 IRP that considered alternatives to PNM's continued use of PV Units 1 and 2 and showed the value of continuing its use of the PV units. Id. at 3639-3640, 3653, 3664. However, neither the analyses nor the 2011 IRP are in evidence. Third, the 2011 IRP was stale at the time that PNM made its irrevocable decisions to retain control of the Leased Assets. For example, the price of natural gas changed significantly between 2011 and 2014. ABCWUA Exh. 10. In September 2013 — before PNM made its decision in December 2013 to purchase the 64.1 MW and extend the 10.4 MW lease for Unit 2 — PNM filed a Notice of Material Event that had the effect, according to PNM, of changing the results of PNM’s 2011 IRP. The Material Event was the New Mexico Environmental Improvement Board's approval of a revised State Implementation Plan (SIP) that provided for the retirement of SJ Units 2 and 3. As a result of the Material Event, PNM stated that it had accelerated aspects of the development of its 2014 IRP and, in December 2013, would request PRC approval of a regulatory plan to comply with the Revised SIP, including any needed revisions to the four-year action plan in the 2011 IRP. Fourth, even if the 2011 IRP was not stale, it did not test extension of the leases and purchase of the 64.1 MW against a wide range of futures/scenarios and input assumptions. PNM uses Strategist, a computer software tool, to rank portfolios. Strategist can consider alternative resource portfolios with the goal of identifying through an optimization algorithm the most cost effective combination of resources as measured by NPV. It varies its assumptions, i.e, cost of fuel, carbon costs, through sensitivity analyses. Strategist determines Recommended Decision Case No. 15-00261-UT 93 whether PNM needs resources and recommends the most cost-effective portfolio to serve load over 20 years. Dauphinais Direct 20-21; Tr. (4-11-16) 224-25 (Ortiz). The 2011 IRP analysis ran Strategist with and without all 178 MW of PNM’s leased PV capacity, assuming under both scenarios that all of these leases expired in 2020. Tr. (4-11-16) 103-04, 320-22 (Ortiz). The scenario without the 178 MW of PV assumed replacement in 2020 of the 178 MW with a new 252 MW combined cycle gas turbine plant, which yielded a NPV that was $51 million higher than the continued leasing scenario ($212 million vs. $161 million). Dauphinais Direct 15. PNM’s sensitivity analyses did not model the option of purchasing the leases at FMV or continued leasing at 50% for two years, but it assumed continued leasing at the then current lease cost through 2020. Mr. Ortiz suggested that the absence of the purchase option being modeled is not material because the revenue requirement of ownership would, over time, go lower than the revenue requirement of continued leasing. Tr. (4-11-16) 272 (Ortiz). While PNM says in this case that whether it considered alternative generation resources to the 64.1 MW of PV2 is irrelevant, PNM submitted evidence of its consideration of alternative resources in Case No. 08-00305-UT, in which PNM sought approval to acquire a beneficial ownership interest in the PV2 owner trust, representing the 29.79 MW. In that case, in support of its request, PNM submitted evidence of the cost of acquiring the 29.79 MW compared to other generation resources (including biomass, coal, nuclear, combined cycle gas-fired) that would be available in the 2016 time frame when that Unit 2 lease interest expired and compared to purchasing the 29.79 MW in 2018 at the end of a renewed lease term. This evidence showed that PNM’s purchase of the 29.79 MW was the lowest cost option on a levelized basis over 30 years. PNM also submitted evidence that purchasing the 29.79 MW would save about $5 million over extending the lease for two years, and lease payments are reduced to one-half for that two-year period. Case No. 08-00305-UT, Certification of Stipulation 43-44. Recommended Decision Case No. 15-00261-UT 94 The AG does not challenge the prudency of PNM’s purchase of the 64.1 MW but argues that the purchase price was not reasonable or supported by PNM. Nevertheless, Ms. Crane aptly explained why PNM’s purchase of the 64.1 MW or its extensions of the leases might not have been “prudent”: [If PNM did need the 64.1 MW of capacity,] I think they should have said, “Do we have other ways to get this capacity? Can we do this more cheaply? Can we do this with less long-term risk to ratepayers? Can we do this in a way that maybe matches this increase in the peak demand that they claim is coming, instead of doing it through additional base load? You know, they didn’t do any of that. And instead, they relied upon, you know, a 2011 IRP which was outdated by the time this decision had to be made. A 2014 IRP, which basically had said, “We’re buying it. We’ve — we’re going to buy it, period.” So there was really no analysis done there. .... But regardless of what recommendations we [the AG] made, one would expect the company to evaluate, at that point in time — when it made the decision to extend those leases, one would expect the company to have evaluated its options to determine, you know, what was the most cost-effective option for ratepayers, what option perhaps minimized the risk of the company. Tr. (6-27-16) 3797-3802. Under PRC precedent, a reasonable utility must consider alternatives before going forward with a project, and a new resource will not be approved if a better alternative is available. For example, in Case No. 2382, the PRC rejected PNM's request for a CCN for the Ojo Line Extension because "PNM's alternatives analysis [was] not sufficiently reliable to determine whether OLE is in fact the best alternative among those presented by PNM. Recommended Decision 98 (7-5-95), adopted by Final Order (11-20-95). The PRC said, “Thus even assuming a need on the transmission system for the sake of argument, the Commission remains unconvinced that the public convenience and necessity require or will require the OLE Project as the proper response to such a need." Id. at 102. The Commission recognized that it has authority to examine alternatives to needs identified by a utility, that there may be various solutions for such needs, and that it would not be in the public interest for the PRC to grant a Recommended Decision Case No. 15-00261-UT 95 CCN for a proposed project that might meet needs but is the worst among a range of alternatives. Id. at 49. In the San Juan Case, the PRC also discussed the requirement that need a CCN be issued for the most cost effective proposal. The PRC found that PNM had sufficiently analyzed the alternatives and that the Stipulation's proposal was the most cost effective alternative. The PRC rejected NEE’s argument that a RFP was necessary to demonstrate that the CCNs for PV Unit 3 and for the additional capacity in SJ Unit 4 were the best replacement resources when SJ Units 2 and 3 were abandoned because PNM already had relevant market information through other recent RFPs and had performed extensive resource planning modeling to show that PV Unit 3 and SJ Unit 4 were its most economical options. PNM had performed a myriad of modeling of different resources, both intermittent like solar and wind, and for baseload like natural gas: PNM’s Strategist analyses in the January and October hearings assessed the costs to operate and maintain a large number of potential resource portfolios to replace San Juan Units 2 and 3. Mr. O’Connell stated that the Strategist modeling considered solar, wind, natural gas, coal and nuclear generation alternatives and assumed the continued growth of PNM’s energy efficiency and distributed generation programs. He said the Strategist modeling evaluated thousands of potential combinations of these resources. San Juan Case, Final Order 7, ¶ 16 (emphasis in original). Other state commissions have ruled similarly. See, e.g., BP Pipelines (Alaska) Inc., Docket No. IS09-395-004 et al., 2014 WL 897389, ¶ 130 (AL Reg. Comm’n 2-27-14) (“[W]hile the Carriers need not consider every possible alternative, they should consider the more reasonable options and provide some rationale for selecting one option over the other.”), corrected on other grds. by Errata (3-7-14). In 2012, the Oregon Public Utility Commission faced similar circumstances when it considered a request by PacifiCorp, d/b/a Pacific Power, for recovery of $661 million for capital investments in emissions control equipment at seven coal-fueled generation units. The investments included projects to reduce emissions of sulfur dioxide, nitrous oxides, and Recommended Decision Case No. 15-00261-UT 96 particulate matter. The PUC found that Pacific Power failed to act prudently in two areas. First, Pacific Power’s claims that there were not legitimate alternative courses of action that could have allowed Pacific Power to meet its air quality requirements at a lower cost and risk to ratepayers were not convincing: “Pacific Power did not alter its course of action or consider alternatives of any kind.” Second, the PUC found that Pacific Power failed to perform appropriate analyses to determine the cost-effectiveness of the investments: “Pacific Power’s contemporaneous cost-effectiveness analyses were demonstrably deficient, and did not demonstrate the rigorous review that a prudent utility should have performed prior to making these significant investments.” In re Pacificorp, 2012 WL 6644237, UE 246, Order No. 12 493, § IV(C)(3)(b) (Or. P.U.C. 12-20-12). Before making its ruling, the PUC addressed the question whether “our prudence standard focuses solely on the decision made by the utility, without regard to the decisionmaking process used to reach that decision.” The PUC said that a utility does not automatically fail its burden of proof if it is unable to present contemporaneous evidence of its own actions; it is possible that the utility may be able to present sufficient information of what it should have known to establish prudence regardless of what it knew. Nevertheless, the PUC explained: [T]he utility’s decision-making process may be highly relevant as to whether a capital investment was prudently incurred. It is often central to the inquiry of whether the utility exercised the standard of care which a reasonable person would be expected to exercise under the same circumstances encountered by utility management at the time the decision had to be made. Id., 2012 WL 6644237, § IV(C)(3)(a). The PUC found that Pacific Power’s cost-effective analyses were flawed in a number of ways, including a lack of meaningful sensitivity and scenario analyses and a failure to update analyses. Id., 2012 WL 6644237, § IV(C)(3)(b). Because of Pacific Power’s failure to reasonably examine alternative courses of action and perform adequate analysis to support its investments, the PUC concluded that a partial Recommended Decision Case No. 15-00261-UT 97 disallowance was warranted. It said, “Because the purpose of a prudence review is to hold ratepayers harmless from any amount imprudently invested, a disallowance should equal the amount of the unreasonable investment.” However, the PUC was not able to easily calculate the precise amount of a proper disallowance because quantifying the impact of the utility’s imprudence “[was] hindered by the very actions that underlie our finding of imprudence — the utility’s inadequate analysis and decision-making.” Had Pacific Power considered alternatives, the PUC would have had the information to calculate the harm to ratepayers for the utility’s decision to proceed with its investments rather than pursuing other, least-costly, options. Id., 2012 WL 6644237, § IV(C)(3)(c). The PUC decided to adopt a percentage disallowance applied to all of the unit upgrades. In doing so, the PUC rejected total disallowance, saying that “even [the Citizens’ Utility Board of Oregon] acknowledges the difficulty of excluding from rate base investments that enable the affected plants to continue to operate and provide service to customers.” Accepting the fact “that it is impossible, on this record, to precisely quantify the impact of Pacific Power’s imprudence,” the PUC found sufficient evidence to support a 10%, or $17 million, disallowance. It further found that the effect of this disallowance, combined with other decisions in the rate case, resulted in rates that were just and reasonable. Id., 2012 WL 6644237, § IV(C)(3)(c). Because PNM has not shown that it considered alternatives to retaining control of PV capacity when the initial leases expired, its decisions to extend the five leases and purchase the 64.1 MW were not prudent. 3. INSUFFICIENT NOTICES TO PRC PNM’s decisions to extend the five leases and purchase the 64.1 MW were imprudent also because PNM did not show that it timely and adequately notified the PRC in advance of its decisions. Recommended Decision Case No. 15-00261-UT 98 Ms. Eden testified that PNM’s decision to renew the leases and purchase the 64.1 MW of PV2 capacity did not require PRC pre-approval because the PRC had already approved PNM’s exercise of either the lease renewal or FMV purchase option when it approved the original leases in Case No. 1995 and Case No. 2019, Phase I. Eden Direct 28. The Hearing Examiner agreed in her Order Denying Motion to Dismiss, issued on February 9, 2016, and the PRC denied NEE’s Motion for Interlocutory Appeal of that Order. Order Denying NEE’s Motion for Interlocutory Appeal (4-13-16). However, the absence of a legal requirement for PNM to receive prior PRC approval to purchase the 64.1 MW and extend the five leases does not mean that PNM acted reasonably in not adequately notifying the PRC before it made these decisions. Ms. Eden said: PNM made the Commission aware of its actions with respect to the proposed PVNGS lease renewals and the repurchase of the ownership interests in PVNGS Unit 2 at an open meeting on October 30, 2013, as well as in letters dated January 13, 2014, February 28, 2014, and May 2, 2014. Eden 5-25-16 Supp. 18. The PRC’s October 30, 2013 Open Meeting was after PNM had given the PV Unit 1 Lessors b0th the First and Second Irrevocable Notices and after PNM had given the PV2 Lessors the First Notice of PNM’s irrevocable decision to retain control of the lease assets, either by extending the leases or purchasing the lease assets. Eden 5-25-16 Supp. 18; Tr. (6-2816) 4107-08 (Eden). Yet Ms. Eden recognized the importance of properly notifying the PRC in advance of its decisions. She said, “If the Commission wanted us to take a different course, we certainly would expect them to have voiced some of those concerns . . . We certainly would have wanted to be aware if the Commission wanted us to take a different course.” Tr. (6-18-16) 4004-05 (Eden). Seeking advance input from the PRC is exactly what a prudent utility would do. Filing appropriate information and giving the PRC and interested persons a reasonable amount of time Recommended Decision Case No. 15-00261-UT 99 to review this information, would have enabled the PRC to assure that PNM’s actions were in the public interest and would have given PNM the assurance that Ms. Eden wanted. How could the PRC have voiced its concerns with PNM’s decisions — as Ms. Eden said PNM would have expected — when PNM did not notify the PRC in advance of, with respect to the PV Unit 1 lease extensions, giving both the First and Second Notices, and, with respect to the PV Unit 2 decisions, giving the First Notice? PNM’s own actions, therefore, prevented the Commission from expressing its opinion on PNM’s plan to continue to participate in the expiring leasehold interests. In conclusion, the following admonition from the Oregon Public Utility Commission to Pacific Power is well said and adopted by this Commission: Regardless of whether a utility intends to use the IRP process for a resource decision, we expect to be kept informed [in advance] about anticipated major utility investment. As this case demonstrates, investments made by a utility to serve its customers can significantly impact the rates paid by those customers. The communications between [PNM] and this Commission with regard to the utility’s investments related to its [PV investments] were not sufficient. Pacificorp, 2012 WL 6644237, § IV(C)(3)(e). This applies even when, as in this case, a utility receives Commission pre-approval to extend a lease or purchase an existing resource. See Gulf States Util. Co. v. Louisiana Pub. Serv. Comm’n, 578 So. 2d 71, 85 (La. 1991) (prudence inquiry encompasses a utility’s continuation of investment as well as its decision to enter into that investment); Washington Utils. & Transp. Comm’n v. Cascade Nat’l Gas Corp., 1995 WL 735608, Docket No. UG-941408 (Wash. Utils. & Transp. Comm’n 10-30-95) (under terms of existing contract, utility must consider annually whether to renegotiate and document thoroughly the information it acquires and considers, and its reasons for making its decisions, to demonstrate that its actions are prudent). G. PNM DID NOT PROVE THAT IT REPURCHASED THE 64.1 MW AT FMV Because PNM did not show that its decision to purchase the 64.1 MW was prudent, it is unnecessary to address PNM’s request for an acquisition adjustment to include the 64.1 MW in Recommended Decision Case No. 15-00261-UT 100 rate base at the alleged FMV of the purchase prices. Nevertheless, this Section addresses PNM’s request to dispel PNM’s repeated claims that the purchases were at FMV. PNM performed no appraisal of the assets comprising the 64.1 MW. PNM's primary support for the $2,500 per kW value lies in two prior sales of PV interests in 2008 and 2012. The first arises from a 2007 auction in which PNM Resources (PNMR), parent of PNM, purchased from First Chicago, a beneficial ownership interest in a PV Unit 2 owner trust, representing 29.79 MW for about $2,850/kW. PNM relies on the PRC’s approval of a stipulation which, among other things, approved (i) PNM’s acquisition from PNMR of that beneficial ownership interest, representing the 29.79 MW; and (ii) inclusion of the 29.79 MW in PNM’s rate base at $2,579/kW. Eden Direct 28-29. The stipulation that the PRC approved in Case No. 08-00305-UT, and upon which PNM relies, is dated September 10, 2008, over six years before PNM entered into the sales agreements in 2015 to purchase the 64.1 MW. Even PNM witness Eden admitted that the electric energy market has changed since 2009. Tr. (4-19-16) 1802-03. Evidence of “contemporaneous sales of comparable properties is generally the preferred method” to resolve disputes over the FMV of property. El Paso Natural Gas Co. v. Federal Energy Regulatory Comm’n, 96 F.3d 1460, 1464 (1996). The validity of this method, however, “depends on both the timing of the sales and the ‘comparability’ of the properties involved.” Where the properties are not shown to be sufficiency similar or the sales sufficiently contemporaneous, the trier of fact must resort to other means of determining FMV. Id. The PRC’s approval of including PV capacity in rate base at a value of $2,579/kW over six years ago and in the context of approving a stipulation, renders that example incomparable. Second, PNM relies on its attempted purchase of another PV Unit 2 lease in 2011. In August 2011, MFS Leasing Corporation (Mellon) informed PNM of an auction it was initiating to sell its 14.89 MW PV Unit 2 lease interest. PNM submitted an offer of about $37.3 million or Recommended Decision Case No. 15-00261-UT 101 $2,505/kW. Mellon told PNM that there were two higher bids and provided PNM the opportunity to increase its bid. PNM raised its bid to $2,578/kW. PNM relies on its higher bid not being accepted and the sale of the PV Unit 2 lease to another bidder — Cypress Second PV Partnership (Cypress) — presumably at a higher price. Id.; Eden Rebuttal 10; Eden 5-25-16 Supp. 19. The sale to Cypress occurred on February 24, 2012, and, like the previous example relied on by PNM, is not sufficiently contemporaneous to be relevant. Moreover, the sale to Cypress is not sufficiently comparable to be relevant. Cypress only bought Mellon’s beneficial interest in the property. Eden 5-25-16 Supp. 20-21. Cypress did n0t assume risks of typical ownership, such as O&M costs, decommissioning costs, and costs of capital improvements. The interest acquired by Cypress is substantially different than PNM’s ownership interest in the 64.1 MW. Cypress’ interest is more like the interest of a bondholder and therefore is likely more valuable and highly valued. PNM also relies on an appraisal for its 134 MW of PV Unit 3 that showed a value of $2500 per kW. This appraisal is not in evidence. Furthermore, the appraisal for PNM's interest in PV Unit 3 was criticized by the PRC in the San Juan Case for its reliance on questionably high projections of future natural gas prices. The Hearing Examiner recommended that the appraisal not be accepted as justification of the value for PV Unit 3 proposed by PNM in that case. San Juan Case, Certification of Stipulation 132-134 (4-18-15). PNM did not provide evidence of the FMV of any sales of true ownership interests in nuclear plants. The PRC, in the San Juan Case, highlighted the need to consider the comparability of the ownership interests at issue in the “comparable sales” used to value a property. In doing so, the PRC relied on PNM witness Reed’s identification of a transaction’s allocation of responsibility for decommissioning costs as a major factor in determining a FMV: Further, Mr. Reed’s comments on the valuation of nuclear power plant sales indicate that Mr. Horn’s analysis is too simplistic. Mr. Reed stated the value of a Recommended Decision Case No. 15-00261-UT 102 transaction is dependent on all of a transaction’s terms, not just price. He said “[i]t’s a little bit more complicated because the value of the transaction is not just dependent on the price. It’s also dependent on the terms associated with decommissioning, the terms associated with transfer of contracts, transfer of people. There are dozens of commercial terms that affect the value of a transaction. So in some cases I’ve, in fact, advised clients to reject an offer that was a higher price but had worse non-price terms. Id. at 134 (emphasis added). In this case, NEE witness Van Winkle attached excerpts from the Direct Testimony of David C. Rode filed on behalf of PRC Staff in the San Juan Case that emphasized significant risks of nuclear ownership. Exh. DVW 21A to Van Winkle Direct. FMV is the price that a seller is willing to accept and a buyer is willing to pay on the open market and in an arm’s-length transaction, neither being under any compulsion to buy or to sell and both having reasonable knowledge of relevant facts. Black’s Law Dictionary 743 (2nd pocket ed. 2001); United States v. Cartwright, 411 U.S. 546, 551 (1973). The elements of an arm’s length transaction are absent from PNM’s purchase of the 64.1 MW of PV Unit 2. As explained by AG witness Crane, once PNM notified the Lessors that it would exercise its option under the relevant leases to buy the capacity interests, it could not walk away from the obligation to buy. In other words, once PNM selected its options to buy, it was compelled to buy the 64.1 MW of capacity. In addition, the December 2013 Memo in which PNM Management highlighted that purchasing the 64.1 MW would increase rate base, indicates that the Lessors and PNM might have shared an interest in obtaining a high price from the sale. If so, there was no true bargaining over price. See Tr. (6-27-16) 3760, 3771 (Crane). Ms. Crane aptly and concisely summed up the evidence when she said that “in no way, shape or form did [PNM] justify the market value.” Id. at 3808 (Crane). PNM had a substantial financial incentive to buy the 64.1 MW. Mr. Ortiz conceded that if PNM did not buy the beneficial interest in this capacity, there was some risk that PNM, not Recommended Decision Case No. 15-00261-UT 103 ratepayers, would bear the cost of non-depreciated capital improvements and decommissioning expenses associated with the capacity after expiration of the leases. Tr. (6-27-16) 3835-3836, 3845-3846. Although Mr. Ortiz downplayed the risk that PNM, rather than ratepayers, would pay for non-depreciated capital improvements and decommissioning costs, this issue has not been decided. In Case No. 1995, the Hearing Examiner found that: It is the policy of the commission that ratepayers should not be responsible for decommissioning costs associated with Palo Verde Nuclear Generating Station Unit 1 associated with that portion of the life of such unit during which it is not owned or leased. Recommended Decision 19, ¶ 19. PNM filed exceptions to this finding, and the PRC in its Final Order reserved for a future case the issue of “what responsibility, if any, ratepayers should bear for decommissioning costs associated with PV Unit 1 (and related common facilities) associated with that portion of the life of such unit during which it is not owned or leased by PNM.” Case No. 1995, Final Order 8. More specifically, the PRC said: The case described in paragraph 24 of the Findings and Conclusions should include, but not necessarily be limited to determining: .... b. what responsibility, if any, ratepayers should bear for decommissioning costs associated with PVNGS Unit 1 (and related common facilities) associated with that portion of the life of such Unit during which it is not owned or leased by PNM[.] Id. at 8, ¶ 25. PNM has not shown that it ever pursued the issue. PNM never obtained approval to recover from New Mexico retail ratepayers, decommissioning costs of PV plant for any period that PV plant is not used to serve New Mexico customers. In Cases Nos. 1995 and 2019, the PRC clearly preserved its authority to rule on this issue. The Stipulation approved in the San Juan Case is consistent with the policy expressed in Case No. 1995. The PRC adopted the parties’ agreement that ratepayers only bear responsibility for decommissioning costs for PV Unit 3 in proportion to the amount of time the plant is used Recommended Decision Case No. 15-00261-UT 104 for retail purposes. The PRC ruled that if the Unit operates to the 2047 expiration of its renewed license, PNM's retail customers will be responsible for about one-half of PNM's 10.2% share of the Unit's decommissioning costs. Certification of Stipulation 25 (11-16-15). H. ALLEGED DOUBLE-COUNTING OF IMPROVEMENTS PNM seeks to recover the cost of the 64.1 MW purchase from ratepayers by including the 64.1 MW in rate base at its FMV and recovering a return of and on this amount. PNM also seeks to recover from ratepayers, separately, a return of and on the improvements to that 64.1 MW and related common plant. PNM acknowledges that the alleged FMV that it paid for the 64.1 MW includes the value of the leasehold improvements to the 64.1 MW. Tr. (6-28-16) 4127 (Eden). The purchase price “is a fair market value for the entire asset.” Tr. (6-29-16) 4152 (Eden). When asked: “PNM did acquire the capital improvements subsumed at $163.5 million and yet it wants to recover the capital improvements on top of that?” Ms. Eden responded: Because it’s really two transactions. There’s two transactions here. We leased the asset for 30 years or 29 ½ years. That’s one transaction that has certain responsibilities. Then we purchased the asset for fair market value, and now we have ownership in the asset until the end of life. Tr. (4-28-16) 4129-30 (Eden). The total amount that PNM seeks to include in Plant in Service in rate base for the Test Period for the 64.1 MW is $207,437,165 million25, consisting of: x $158,300,85226, representing the depreciated value of the $165.3 million paid by PNM for the 64.1 MW; and x $49,136,31427, representing the rate base value of the improvements to the 64.1 MW (including improvements to common plant). Mr. Peters said that PNM’s purchase of the 64.1 MW was to purchase ownership of that 64.1 MW as of January 15, 2016, and going forward. Therefore, he said that including the PNM Exh. 88, p.1, column H, row 27. PNM Exh. 88, p.2, column O, row 27; Tr. (6-29-16) 4249 (Peters). 27 PNM Exh. 88, p.2, column M, row 27. 25 26 Recommended Decision Case No. 15-00261-UT 105 alleged FMV in rate base is to allow PNM to recover a return of and on that asset going forward. On the other hand, according to Mr. Peters, the $49,136,314 represents the net value of the improvements, which have not been fully depreciated. Mr. Peters viewed the capital improvements as a separate asset. He said that from a ratemaking perspective, “we have to treat them as separate assets, because they were separate transactions, separate purchases that PNM had to make.” PNM argues that its ratepayers have not fully paid for PNM’s cost of these improvements because they have not been fully depreciated through the end of their service lives, which the PRC extended to 2046 in Case No. 08-00273-UT. Tr. (6-29-16) 4229-31, 423745, 4311-12 (Peters). Ms. Eden said that even if PNM had returned the Leased Assets to the Lessors at the end of the original lease terms, and even though PNM would not have had title to leasehold improvements to those Leased Assets, PNM would have still sought to recover the cost of the undepreciated improvements from ratepayers. Tr. (6-28-16) 4127-29 (Eden). PNM’s position is precarious, at the least, considering that when it repurchased the assets from the Lessor, it obtained title to the capital improvements. I. SALE/LEASEBACK PAYMENTS PNM recovered its original cost of construction of PV Units 1 and 2 of approximately $644.3 million plus a before-tax-gain of approximately $236 million when it sold the Units to the investors in 1985 and 1986. Harland 5-25-16 Supp. 3. The investors recovered their cost plus a return through the lease payments made to them. PNM recovered its lease payments from ratepayers. Indeed, in Case No. 2262, the Commission explained that for ratemaking purposes the lease payments are the equivalent to paying capital costs. Over the lease terms, PNM, on a total company basis, paid $2.083 billion in lease payments. Exh. JAP-3, p.5 to Peters 5-25-16 Supp. PNM made the payments to the Indenture and Collateral Trust, which made debt service payments to bondholders and provided an equity Recommended Decision Case No. 15-00261-UT 106 return to the Owner Trust, which ultimately flowed through to the Owner Trusts’ equity investors and supported contributions to a sinking fund for the “return of” capital to the lenders. Eden 5-25-16 Supp. 3-5. The “return of” capital is similar to recovering depreciation expense. Thus, irrespective of what length of time PNM was authorized to depreciate PV Units 1 and 2, ratepayers fulfilled their “return of” capital obligations by funding a sinking fund over 30 years. Ratepayers’ contributions to the sinking fund or the amount that made up the “return of” capital was based upon $900 million of capital and not “the original cost” of approximately $644.3 million. Thus, ratepayers not only fully paid for the depreciation of the original cost of PV Units 1 & 2, they also returned an additional $256 million of capital less the then calculated gain net of tax, or $42.5 million. The debt associated with the sinking funds has been fully paid and there are no sinking fund contributions included in PNM’s requested revenue requirement. Eden 525-16 Supp. 22. The 30-year term of the lease payments also accelerated recovery of capital expenditures. If the return of capital obligations had been recovered through depreciation expense, recovery would have occurred over the 40-year useful life authorized by the Nuclear Regulatory Commission. Because recovery occurred over 30, not 40, years, Ms. Crane said that it would be reasonable to calculate the NBV of the 64.1 MW using a 30-year service life. She explained: It would be reasonable to utilize a 30-year period for this plant because that is the term of the original lease. Accordingly, it is reasonable to assume that the investors used a 30-year time horizon to evaluate the proposed transaction. While the Lessors also had the option to sell the plant at the end of the lease period, the valuation at that time was uncertain. Therefore, my expectation is that the original lessors evaluated the transaction based on a 30-year timeframe. Therefore, it would be reasonable for the Commission to utilize a 30-year period to determine the accumulated depreciation associated with the original transaction. Crane 6-14-16 Supp. 14. J. HEARING EXAMINER’S RECOMMENDATION The result of finding that PNM’s decisions to extend the five PV leases and purchase the 64.1 MW of PV2 capacity were imprudent is to exclude from PNM’s non-fuel revenue Recommended Decision Case No. 15-00261-UT 107 requirement the depreciated value of the purchase price of the 64.1 MW and the annual lease payments for the five extended leases. Since PV Units 1 and 2 will continue to serve ratepayers, ratepayers should continue to pay some costs associated with these assets, such as fuel and O&M expenses and unrecovered costs of previous capital improvements and common plant that have been previously approved as reasonable. Mr. Peters believes that a Commission’s decision to exclude the 64.1 MW from rate base at PNM’s requested value or to include it at a zero value would result in a write-off. For accounting purposes, a write-off involves removing the asset from the balance sheet and recording a one-time regulatory disallowance on the income statement that reduces equity through retained earnings. A write-off of $158,300,852 would result in a $158,300,852 pre-tax reduction to retained earnings. Tr. (6-29-16) 4304 (Peters). This would equal a 28.2% decrease in PNM’s 2015 retained earnings reflected in Staff Exh. 1, p. B-14. PNM argues that excluding all or most of its investments in PV from rate base will cause serious harm to PNM. The only evidence of an adverse impact on PNM associated with such disallowances is PNM’s claim that its credit rating might be negatively impacted. PNM has not submitted evidence of the amount of the minimum revenue increase it believes it needs to maintain its creditworthiness. The Hearing Examiner is recommending that rates be approved for PNM to result in over a 6.0% increase in current revenues. This recommendation results in just and reasonable rates that provide PNM an opportunity to earn a fair rate of return on its investments, based on a balancing of investor and ratepayer interests. Maintenance of a particular bond rating is not the essential criterion, or indicator, of PNM’s financial health. Case No. 2146, Part II, Final Order 69. PNM survived the PRC’s exclusion of PV Unit 3 from rate base, which, according to PNM at the time, equated to a disallowance of $306.5 million from its rate base, which was 19% of PNM’s net electric plant. Case No. 2146, Part II 101. Recommended Decision Case No. 15-00261-UT 108 This Commission has said: A review of the previous Supreme Court cases shows that even in the extreme situation of a utility bankruptcy, utility commissions are still allowed wide latitude in determining just and reasonable rates. In balancing the sometimes irreconcilable differences between ratepayers and investors, utility commissions may constitutionally deny a utility a return on its investments without violating constitutional due process or taking of property principles, even if the investments were prudently made and even if it causes substantial financial harm to the utility. Case No. 2146, Part II, 45. Excluding an asset from rate base but continuing to regulate the asset for its use in serving ratepayers does not necessarily violate a utility’s due process rights, as the New Mexico Supreme Court held in Public Service Company v. New Mexico Public Service Commission. There, the Court squarely rejected PNM’s argument that the PRC’s refusal to decertify SJ4, while at the same time excluding it from PNM’s rate base, violated PNM’s due process rights. The Court explicitly “disagreed” with PNM’s argument that “exclusion of an asset from rate base is logically inconsistent with continued regulation of the same asset.” 1991-NMSC-083, ¶ 16. It found that PNM’s assertion that exclusion of SJ4 from rate base represented a loss of capital costs alone of $19.8 million per year fell “well short of the required showing.” 1991-NMSC-083, ¶ 18. Furthermore, whether the Commission should consider the financial effects of a prudence disallowance is questionable. A used and useful disallowance may be appropriate even if a utility is prudent. And under the circumstances of a used and useful test, the Commission should balance the interests of shareholders and ratepayers and determine just and reasonable rates that are in the public interest. In addressing the interests, the Commission may appropriately consider financial effects on the utility. A disallowance due to imprudence is, however, quite different; and to consider financial harm in determining a disallowance founded on the utility being imprudent would, in essence, be rewarding a utility for its imprudent acts. Recommended Decision Case No. 15-00261-UT 109 However, the disallowances of costs from PNM’s revenue requirement in this case, as a result of the findings of imprudence, are not necessarily permanent disallowances. PNM in its next base rate case filing can attempt to show that the PV repurchase and lease extensions are the most cost effective resources among available alternatives to meet customers’ needs at that time. PNM did not attempt to make that showing in this case. In PNM’s next base rate case, the PRC will consider any evidence and arguments submitted as to what type of resources are needed and represent the most cost effective alternatives at that time. At a minimum, any such evidence presented by PNM shall include the average cost per kWh of each option considered. Because PNM’s decisions to extend the five PV leases and purchase the 64.1 MW of PV2 were not prudent, it is not necessary to address whether these PV capacities are used and useful, whether PNM’s request for an acquisition adjustment should be approved, or the NBV of the 64.1 MW. XVI. BALANCED DRAFT PNM seeks approval to include in rate base $78,210,784, representing the cost to PNM of installing Selective Non-Catalytic Reduction Equipment (SNCR) on San Juan Generating Station (SJGS or SJ) Units 1 and 4 and converting these Units to a balanced draft configuration. PNM combined installing the SNCRs and converting to a balanced draft configuration into a single capital project to reduce costs and the need for multiple unit outages. Olson Rebuttal 5-6. However, SNCR and balanced draft can be installed independently from one another. Tr. (4-1816) 1490 (Olson). The table below shows the cost breakdown between installing the SNCRs and converting to a balanced draft system: SJ Unit 1 SJ Unit 4 Total Balanced Draft $21,784,082 $30,492,959 $52,277,041 Recommended Decision Case No. 15-00261-UT SCNR $14,140,867 $11,792,876 $25,933,743 110 Total $35,924,949 $42,285,835 $78,210,784 Olson Direct 28. No parties or Staff oppose PNM’s recovery of the $25,933,743 cost of SCNR in rate base. SCNR is a post-combustion control technology for control of emissions of nitrogen oxides (NOX). A balanced draft system is a gas handling system that pushes and pulls gas through the entirety of a boiler and environmental controls of a unit. Olson Direct 26. Several Intervenors — WRA, the AG, the City/County, CCAE, CFRE and NEE — oppose PNM’s request to include in rate base the $52,277,041 cost of converting SJ Units 1 and 4 to balanced draft. WRA witness Howe submitted the most extensive testimony against including balanced draft in rate base. Staff takes a different approach with respect to PNM’s proposed recovery of the costs of the balanced draft conversion. Staff Witness Carrara acknowledged that balanced draft provides operational and environmental benefits and is used and useful. He also said that balanced draft will substantially eliminate the need for O&M related to duct leaks. However, Staff has reservations about whether balanced draft was “required” versus “accepted” by the New Mexico Environment Department (NMED) or the Environmental Protection Agency (EPA). Staff also believes PNM has not demonstrated that balanced draft is a more cost-effective solution than possible (although unidentified) alternative solutions. For these reasons, Staff proposes that PNM’s revenue request for the balanced draft conversion be limited to $2.1 million instead of $5.2 million. Carrara Direct 6-8. A. BACKGROUND ON INSTALLATION OF BALANCED DRAFT The EPA finalized its Regional Haze Rule in 2005. In 2006, NMED told PNM that SJGS would be subject to this Rule. In 2011, the EPA issued its Federal Implementation Plan (FIP) for New Mexico and the SJGS. The FIP required installation of SCR and balanced draft on all four SJ Units. Tr. (4-18-16) 1512, 1592-93 (Olson). Recommended Decision Case No. 15-00261-UT 111 In October 2010, Black & Veatch (B&V) prepared a cost study of installing SCR on SJ Units 1, 2, 3, and 4. The study’s estimated cost included $18.7 million at each of SJ Units 3 and 4 for balanced draft conversions. In a letter discussing the cost study, John Bunyak of the U.S. Department of Interior told the EPA that, while increased draft was needed to support a SCR (which would be delivered by induced draft fans), a balanced draft conversion was not required for a SCR. Howe Direct 4. On August 22, 2011, the EPA, in discussing whether balanced draft was required as part of the FIP, said that B&V’s SJGS costs were unusually high for four reasons. One reason was “including costs for equipment that is not necessary for a SCR (e.g., balanced draft conversion, sorbent injection, SCR bypass)[.]” Id. Nevertheless, a white paper dated September 2011, prepared by PNM consultant Class One Technical Services, references a meeting on August 10, 2011, between PNM and NMED to discuss a forthcoming permit application by PNM to address the FIP requirements. The white paper says that the Units will convert to balanced draft because of the “extremely stringent NOx limit of 0.05 lbs/mmBTU” and also states that an added benefit of balanced draft will be elimination of fugitive emissions. Id. at 5. In September 2011, PNM and NMED appealed the EPA’s FIP decision, arguing that the New Mexico State Implementation Plan (SIP), requiring a much less expensive control — SCNR — would meet the requirements of the Regional Haze Rule. Id. On April 3, 2012, Class One Technical Services submitted an application for revision of a NMED air permit on behalf of PNM. The application was for the addition of SCR on all four SJ Units, but noted that the FIP was on appeal and, if vacated, PNM would instead implement the SIP which called for the installation of SNCR on all four Units. These were called Scenarios A and B, respectively. The application called for balanced draft to be installed in both scenarios: “Both scenarios include modifications to the fan system to achieve “balanced” draft configuration Recommended Decision Case No. 15-00261-UT 112 allowing for the elimination of emission units E501, E502, E503 and E504.” Tr. (4-18-16) 1593 (Olson). The air permit, approving both Scenarios A and B, was issued by the NMED on August 31, 2012. This permit required the installation of balanced draft under either Scenario A or B. Tr. (4-18-16) 1513, 1593 (Olson). A Sargent & Lundy Report dated March 29, 2013, estimated the costs of SNCR and SCR at SJGS. At this time, PNM and NMED were still contemplating either a 4-unit SCR project or a 4-unit SNCR project. This document identifies balanced draft as part of both the SCR and SNCR conversion scenarios. With respect to SCR conversion, the document states that PNM may be required to address fugitive emissions at the plant, and it goes on to clarify that these fugitive emissions are fine ash particles pushed out through small openings in the ductwork: As part of the SCR design review process, PNM verified that the SJGS may be required to reduce existing fugitive emissions at the plant. Most of the fugitive emissions result from the boilers' pressurized design. Internal boiler and ductwork pressures push fine ash particles out through any small openings in the existing ductwork. Converting the plant to a balanced draft operation, which means internal pressures in the boiler and ductwork will be close to atmospheric pressure, will minimize or eliminate these fugitive emissions. Howe Direct 6. The Report continues to explain that SCR would create pressure drops that would need to be addressed by new fans “regardless of whether the units are converted to balanced draft.” However, with respect to SNCR conversion, the report concludes that none of the balanced draft costs are part of, or necessary for, the SNCR project: As discussed above, PNM may be required to reduce existing fugitive emissions at the plant. Fugitive emissions from the existing boilers could be minimized or eliminated by converting the boilers to balanced draft operation. Balanced draft conversion would require the installation of new fans, boiler and ductwork stiffening, and upgrades to the existing electrical systems. However, unlike the SCR project, SNCR could be installed and operated on the units without these upgrades; therefore, balance draft costs were not included in the SNCR cost estimate. Id. at 7. Recommended Decision Case No. 15-00261-UT 113 Subsequently, an agreement called the Revised State Implementation Plan (RSIP) was reached. It required the closure of SJ Units 2 and 3 and the installation of SNCR on Units 1 and 4. Id. The New Mexico Environmental Improvement Board approved the RSIP in September 2013. Tr. (4-18-16) 1594 (Olson). After the agreement was approved, PNM’s NMED permit was revised in May 2015 to change the requirement for SCR technology to SNCR technology, but PNM did not seek revision of the permit to remove the requirement for installation of balanced draft. Tr. (4-18-16) 1367, 1555, 1594 (Olson). PNM admits that it did not seek such a revision and “has no basis to believe that the NMED would reverse itself with respect to the requirement for conversion of San Juan to balanced draft.” PNM’s Posthearing Response Brief 54. After the agreement was approved, PNM was issued two permits: a New Source Review permit and a Title V permit. Both require installation of balanced draft. Olson Rebuttal 7; Tr. (4-18-16) 1554 (Olson). PNM presented the balanced draft project to the SJ Participant Coordination Committee. There are two owners of SJ Unit 1 — Tucson Electric and PNM — and they both approved installation of balanced draft on Unit 1. They shared the cost of installing balanced draft on SJ Unit 1. There are four owners of SJ Unit 4. PNM is the only owner who voted to install balanced draft on Unit 4. However, PNM, as the operator of SJ Unit 4 was able to invoke operator prerogatives under an operating agreement to install balanced draft on SJ Unit 4. While the other owners did not vote to approve installation of balanced draft on SJ Unit 4, they are sharing in the cost of its installation. Tr. (4-18-16) 1372-74, 1498-1502, 1570-72 (Olson). Conversion of SJ Unit 1 to balanced draft was complete on April 28, 2015, and conversion of SJ Unit 4 was complete on November 25, 2015. Olson Rebuttal 5. In December 2013, PNM filed its Application in the San Juan Case, in which in sought: x Approval to abandon SJ Units 2 and 3 x Issuance of a CCN for PV Unit 3 Recommended Decision Case No. 15-00261-UT 114 x Issuance of a CCN for 78 MW in SJ 4 x Approval to recover the costs of SNCR and balanced draft to be installed on SJ Units 1 and 4 San Juan Case, April 8, 2015 Certification of Stipulation 1. Paragraph 28 of the initial Stipulation filed in the San Juan Case said that the Stipulating Parties agreed that PNM’s reasonable and prudent costs of the SNCR project, including balanced draft, shall be included in rate base. Id. at 23. At the hearing, the Stipulating Parties clarified that, under Paragraph 28, they were not requesting approval of the SNCR project or that the PRC determine whether the costs of the project were reasonable or prudent, and that the prudence or reasonableness of the costs would be determined in a rate case or other case. According to PNM, the intent of Paragraph 28 was that the presumption that the costs of the SNCR project were reasonable is limited to $90.6 million. Opponents of the Stipulation challenged the reasonableness of the balanced draft portion of the SNCR project. Id. 23, 58-60, 118. In his Certification of the Initial Stipulation, the Hearing Examiner found, “The evidence raises serious questions regarding the prudence of the balanced draft portion of the project[.]” Id. at 118, 139. He recommended that Paragraph 28 be modified to reflect that PNM must file direct testimony on the prudence of the SNCR costs when it seeks recovery of those costs. He said, “PNM should provide affirmative proof of the reasonableness of the balanced draft expenditures in the proceeding in which PNM seeks to recover those costs.” Id. at 140. In the Supplemental Stipulation filed after issuance of the Certification of the Initial Stipulation, the Stipulating Parties agreed that installation of SNCR on SJ Units 1 and 4 was prudent, and that PNM should be authorized to recover the reasonable costs of SNCR in rates, but they also agreed that the prudence and reasonableness of the costs of balanced draft “will be determined in a PNM general rate case. PNM shall make an affirmative demonstration that incurrence of the costs of balanced draft was prudent and reasonable.” The Hearing Examiner Recommended Decision Case No. 15-00261-UT 115 recommended approval of this provision, and it was approved by the PRC in its Final Order issued in December 2015. San Juan Case, Nov. 16, 2015 Certification of Stipulation 55. B. ANALYSIS OF PNM’S ARGUMENTS IN SUPPORT OF ITS CLAIM THAT INSTALLATION OF BALANCED DRAFT WAS PRUDENT PNM has not shown that its decision to install balanced draft was prudent. PNM says that it “is not justifying the investment in balanced draft based on O&M cost or other savings; rather the primary driver for the conversion is the permit requirements and continued compliance with applicable emissions limitations.” It further says that there are other benefits associated with the balanced draft conversion such as improved workplace conditions, including the mitigation of ammonia exposure, and enhanced operations. PNM’s Initial Posthearing Brief 85. One of PNM’s repeated arguments in support of including balanced draft in rate base is that balanced draft is an express condition in its SJ air permits. E.g., PNM’s Initial Posthearing Brief 85. A reasonable conclusion to draw from PNM’s reliance on balanced draft being an express condition in PNM’s SJ air permits is that NMED required installation of balanced draft as a condition of issuing the permits. However, an e-mail produced by WRA shows that NMED did not in fact require installation of balanced draft as a condition of issuing the permits. In an email to Bruno Carrara, the PRC’s Utility Division Director, dated December 9, 2014, Richard Goodyear, Air Quality Bureau Chief of the NMED wrote: The balanced draft project is not a requirement of the proposed BART determination (State Alternative) and is not required, in general, by the Regional Haze regulation. The balanced draft project is also not a requirement for the installation of SNCR technology. According to the technical analyses submitted by PNM to the Department, the balanced draft project is a requirement for installation of either an SCR system or a hybrid SCR/SNCR system, but would not be required for the installation of SNCR. PNM included the costs of a balanced draft project in their cost analyses of both the SCR and SCR/SNCR systems in its BART submittal, but did not include the costs of the balanced draft conversion project in its cost analysis of the SNCR system. Recommended Decision Case No. 15-00261-UT 116 Please note that PNM's assertion that the state of New Mexico required the balanced draft conversion is incorrect. PNM's request to implement the balanced draft project was entirely voluntary and only appears in the air quality permit because PNM requested the inclusion of the project in their air quality application. Exh. DJG-2 to Howe Direct (emphasis added). Mr. Olson admitted that it was PNM that proposed balanced draft as an environmental compliance measure. Tr. (4-18-16) 1363, 1550. PNM says, “The NMED’s decision to include this condition in both of these permits is a final determination on the need for balanced draft for San Juan Units 1 and 4, and this determination is conclusive and binding in the context of this proceeding pursuant to the Air Quality Act.” PNM argues, “The Commission is not at liberty to overrule or second-guess the requirement imposed by the NMED that the two San Juan units be converted to balanced draft.” PNM’s Initial Posthearing Brief 86-87. PNM’s emphasis that its air quality permits require balanced draft suggests that if the PRC denies PNM’s request to include balanced draft in rate base, PNM will be in violation of the permit. However, when Mr. Olson was asked: Q. So a denial by the Commission of recovery by PNM of the cost of balanced draft in rates, that wouldn’t result in PNM’s noncompliance with the permit, right? Mr. Olson responded, “Right.” Tr. (4-18-16) 1581. PNM argues, at least by implication, that because balanced draft is a requirement of its operating permits, the Commission lacks authority to deny cost recovery. That is not the case. The Commission is vested with considerable discretion in determining the justness and reasonableness of utility rates. Attorney General v. New Mexico Pub. Serv. Comm’n, 1984NMSC-081, ¶ 12, 101 N.M. 549. A PRC decision to exclude balanced draft from rate base would not result in the balanced draft requirement being removed from the permits, and Mr. Olson said that he could not see any benefit to asking NMED to remove the balanced draft requirement from the permits if the PRC excludes balanced draft from rate base. Tr. (4-18-16) 1580-81. Recommended Decision Case No. 15-00261-UT 117 PNM says compliance with applicable air quality requirements was the primary driver for the balanced draft requirement. PNM’s Initial Posthearing Brief 85. However, PNM admits that SJGS had not been found to be non-compliant with the National Ambient Air Quality Standards (NAAQS) or any other air quality requirements for fugitive emissions from duct leaks, and was currently in compliance with those requirements. Tr. (4-18-16) 1372, 1376, 1569 (Olson). Mr. Olson said that balanced draft was not needed to comply with regional haze requirements. Tr. (4-18-16) 1374. This is confirmed by the following sources: 1. Mr. Goodyear’s e-mail, in which he said: As PNM was in compliance with all applicable ambient air quality standards in effect prior to the proposed installation of the balanced draft project, it should be noted that the project is not required to comply with any applicable ambient air standard. Howe Direct 11. 2. PNM’s own expert witness on environmental matters in the San Juan Case, Mr. Cichanowicz, testified that the SJGC complied with all NAAQS requirements without balanced draft and that, after 2017 when units 2 and 3 are retired and emissions are cut in half, SJGS would continue to be in compliance with NAAQS without balanced draft. Mr. Cichanowicz agreed that the impact of PNM’s balanced draft installation on reducing CO2 would only be about 1%. Id. at 11-12. Nevertheless, PNM says that installing balanced draft was prudent because NAAQS modeling showed that SJGS was close to the compliance limit for the NAAQS PM 2.5 standard28 “with the very real prospect that the standard would be lowered such that San Juan would be at the maximum compliance level.” Olson Rebuttal 15; Tr. (4-18-16) 1378 (Olson). PNM argues that because PNM is now at, but not exceeding, the particulate compliance level (PM2.5) of 12 Particulate matter is inhalable particles. PM2 means fine particles, with diameters generally 2.5 micrometers and smaller. The PM2 standard specifies the maximum amount of PM2 to be present in outdoor air. See Tr. (4-18-16) 1374-75 (Olson). 28 Recommended Decision Case No. 15-00261-UT 118 ug/m3 (micrograms/cubic meter), and because the closure of Units 2 and 3 will not occur until the end of 2017, PNM had to install balanced draft to provide a margin of cushion for compliance. Olson Rebuttal 40. PNM says, and no party denies, that balanced draft eliminates emissions of pollutants, including PM2.5 and NOx. PNM Exh. 68; Tr. (4-19-16) 1595-97 (Olson). However, PNM did not quantify the amount of emission reductions from balanced draft. Tr. (4-18-16) 1578-79 (Olson). In support of the prudence of installing balanced draft, PNM also says that, in the absence of the balanced draft conversion, employees would potentially be exposed to ammonia fumes, and that balanced draft mitigates the need for San Juan employees to use respiratory protection due to exposure to fugitive dust emissions. Olson Rebuttal 21-22. PNM says, and no party denies, that installing balanced draft eliminates or reduces fugitive emissions associated with duct leaks. WRA does not dispute that balanced draft will reduce workplace exposure to emissions from duct leaks. Tr. (4-25-16) 2623-24, 2652 (Howe). About 330 employees work at SJGS. Tr. (4-18-16) 1385 (Olson). During the last seven years, PNM recorded five written complaints due to fugitive emissions from San Juan Units 1 and 4 and 15 employees filed worker’s compensation claims due to fugitive dust emissions. Tr. (4-19-16) 1599-1560, 1605-06(Olson). However, Mr. Olson did not know how the complaints were resolved. PNM did not identify any OSHA violations that would occur in the absence of balanced draft. Tr. (4-18-16) 1386 (Olson). And, Mr. Olson said that employees in some areas of SJGS continued to wear respiratory protection after installation of balanced draft. Tr. (4-1816) 1384. Evidence that installation of balanced draft was not necessary to comply with air quality requirements supports the imprudency of the PNM’s decision. However, the fact that installation was not necessary to comply with regulations does not necessarily mean that PNM’s decision to install balanced draft was imprudent. Evidence that installation of balanced draft Recommended Decision Case No. 15-00261-UT 119 produced significant environmental or health benefits might have met PNM’s burden to show prudence. However, PNM submitted no evidence that reducing emissions resulted in environmental benefits. Tr. (4-25-16) 2625 (Howe). PNM submitted no studies or analyses showing qualitative or quantitative health benefits from installation of balanced draft. WRA’s Posthearing Response Brief 8. PNM also submitted no evidence that installation of balanced draft improved workplace safety. Evidence of complaints and worker’s compensation claims are non-specific and not compelling. PNM’s decision to install balanced draft was imprudent and the $52,277,041 cost of balanced draft should be excluded from PNM’s rate base. XVII. PREPAID PENSION ASSET PNM seeks to include in rate base the amount of contributions made by PNM’s shareholders to the employee pension plan trust in excess of the amounts that were expensed and recovered from customers in accordance with ASC 715-30. Peters Direct 21-22. The contributions to PNM’s pension plan are determined by the PNMR Investment Committee based on annual actuarial studies by Towers Watson based on what is required under the Pension Protection Act of 2006. Eden Direct 34. The resulting amounts in excess of what is expensed and recovered from customers are referred to as the Prepaid Pension Asset. The total amount of the Prepaid Pension Asset that PNM seeks to include in its retail jurisdictional rate base is $137.8 million. Peters Rebuttal 9. PNM’s Prepaid Pension Asset proposed to be included in rate base takes into account the total pension expense through September 30, 2016, and contributions that have been or will be funded to the pension plan through that date. Peters Direct 22. PNM’s Prepaid Pension Asset reflects mandatory shareholder contributions as determined by Towers Watson. Eden Direct 33-34. Recommended Decision Case No. 15-00261-UT 120 In the 2007 PNM Electric Rate Case, the Commission allowed PNM to include the amount of its Prepaid Pension Asset in its rate base after a showing that PNM’s shareholder had contributed more to the pension plan than had been expensed and paid by ratepayers and that there was a benefit to ratepayers in the form of lower pension costs. NMPRC No. 07-00077-UT at ¶¶ 5-52. PNM has also been allowed to include its Prepaid Pension Asset in it rate base in PNM’s 2008 rate case (08-00273-UT) and in its 2010 rate case (10-00086-UT). PNM-47 at 22:16-18. PNM prepared a cost-benefit analysis as required by the Final Order in the 2007 PNM Electric Rate Case.29 This analysis demonstrates that revenue requirements, including a full return on the Prepaid Pension Asset included in rate base, are slightly higher than the expense that would have been included in PNM’s revenue requirement calculation absent the additional shareholder funding. Therefore, PNM proposes to only include the amount of Prepaid Pension Asset in rate base up to the breakeven point in the revenue requirement for the expense without the contributions compared to the revenue requirement associated with the Prepaid Pension Asset in rate base. This reduces rate base by $22 million that would otherwise be requested for the Prepaid Pension Asset in this proceeding. Peters Direct 23. PNM’s request is opposed in whole or in part by the AG, NMIEC, and ABCWUA. The New Mexico Supreme Court very recently issued an opinion announcing the standard for including prepaid pension assets in rate base, which in large part disposes of the Intervenors’ objections. In the Supreme Court case, Southwestern Public Service Company (SPS) sought approval to include a prepaid pension in rate base. A prepaid pension asset is the amount by which investor contributions to a pension trust and earnings on these contributions exceed pension expenses. AG v. PRC, 2015-NMSC-032, ¶ 3. 29 Final Order 19, ¶ 51. Recommended Decision Case No. 15-00261-UT 121 Pension contributions and expenses differ because the federal Employee Retirement Income Security Act of 1974, 29 U.S.C. §§ 1001-1461 (2011) and the Internal Revenue Code, 26 U.S.C. §§ 1-59 (2012), dictate how much a utility must contribute to its employee pension program, whereas the Financial Accounting Standards Board (FASB) promulgates accounting standards that govern how pension expenses are determined. Because of these differing federal and industry standards, pension contribution and expense calculations use different assumptions, attribution methods, and periods of time over which the costs are required to be recognized. These similarities often result in differing annual contribution and expense amounts. When mandated contributions and income earned on the contributions exceed expenses, a prepaid pension asset accrues. Id., ¶ 4. A utility cannot legally withdraw any funds from pension trusts except to pay pension benefits and expenses. However, ratepayers benefit from a prepaid pension asset because the earnings on this asset are deemed to be income for a utility, which reduces the amount of revenue it must collect from customers. Id., ¶ 5. The New Mexico Supreme Court provided the following hypothetical to illustrate the indirect benefit that customers receive: [S]uppose that in a given year the utility had a revenue requirement of $300, and that it expected to earn a 6% return on the pension fund. The $3.00 return on [a hypothetical] $50 prepaid pension asset (0.06 x $50) . . . would be credited against the revenue requirement, so that the utility could only collect $297 from its customers through [the] rates. Thus, the revenue requirement is reduced by $3.00 as a result of the prepaid pension asset. Id. The New Mexico Supreme Court characterized a prepaid pension asset as working capital, i.e., “‘an allowance for the sum which the company needs to supply from its own funds for the purpose of enabling it to meet its current obligations as they arise and to operate economically and efficiently.’” Id., ¶ 15 (quoting Gov’t of Guam v. Fed. Mar. Comm’n, 329 F.2d 251, 256 (D.C. Cir. 1964)). The Court explained that, as a result, only utility contributions, not Recommended Decision Case No. 15-00261-UT 122 ratepayer contributions, can be properly included in rate base as working capital. It said, “For example, if a utility were to prepay for natural gas with investor funds, the utility should expect to receive a reasonable return on its investment.” Id. Therefore, the Court said, “A utility can include prepayments for pension expenses in its rate base ‘because the utility is out-of-pocket for such costs until they are recovered from ratepayers and is therefore entitled to recover its cost of financing such prepaid expenses.’” Id., ¶ 16 (quoting S. Co. Servs., Inc., 122 FERC ¶ 61,218, at *62235, 2008 WL 630079, slip copy at 5). The Court further explained that, in the context of prepaid pension assets, income earned on the pension fund is reported under GAAP as a reduction in pension expense, which reduces the amount collected from ratepayers. Under these circumstances, the utility must finance the reduction because it cannot use the income from the pension trust to pay other current obligations; as a result, the utility is allowed to recover the costs of financing the reduction by including the pension income in rate base. Id. The Court said that the amount of the prepaid pension asset included in rate base is not limited to the amount of the reduction in the utility’s pension expense because “a utility may not only be out-of-pocket for reductions in its revenue requirement resulting from pension fund earnings. A utility may also be out-ofpocket for investor-funded contributions that are in excess of pension expenses.” Id. 16-17. The value of SPS’s prepaid pension asset was $36.9 million and produced $1.7 million in annual earnings that reduced SPS’s pension expense by $1.7 million, which reduced SPS’s revenue requirement by $1.7 million. SPS sought to include the $22 million net amount of the prepaid pension asset in rate base (the $36.9 million asset minus a $14.9 million tax deferred asset). Id., ¶ 6. The PRC granted SPS’s request to include the $22 million net amount in rate base because doing so “recognizes that ratepayers benefit from the prepaid pension asset and that the Recommended Decision Case No. 15-00261-UT 123 utility should earn a return on the prepaid pension asset in order for the utility to recover its full cost of service.” Id., ¶ 8. The AG appealed the PRC decision, making an evidentiary argument that including “the net prepaid pension asset in rate base will result in ratepayers paying more to SPS than the benefit ratepayers have enjoyed from the pension fund earnings.” The Court, interpreting the AG’s argument to be that SPS did not prove how much of the net prepaid pension asset resulted in consumers paying $1.7 million less to SPS, “disagree[d],” saying, “We hold that some or all of a prepaid pension asset should be included in rate base to the extent that the evidence evinces that the asset was investor-funded, as opposed to ratepayer-funded.” Id., ¶ 19. The Court said that while the AG was correct to make an evidentiary argument, “the premise of the argument is incorrect.” The Court reiterated that “[u]tilities are able to recover the costs of financing their business operations through the inclusion of investor-supplied working capital in the rate base.” Id., ¶ 21. Because the evidence was uncontested that SPS investors made contributions to the pension fund that were required by law and that “exceeded expenses and generated earnings that effectively reduced SPS’s — and consequently the ratepayers’ — pension expense . . . . the net prepaid pension asset was properly included in the rate base.” Id. The Court made clear that its basis for approving including the prepaid pension asset in rate base was to allow “utilities to recover the costs of financing their business operations through the inclusion of investor-supplied working capital in the rate base.” Id. In this case, the AG, as he did in the Supreme Court case cited above, contests including PNM’s Prepaid Pension Asset in rate base. AG witness Crane does not dispute that shareholders have contributed amounts in excess of ratepayer expense, but she contends that ratepayers have not benefited from the Prepaid Pension Asset. Specifically, AG witness Crane contends that PNM’s pension investments have not fared well during the period from the end of 2007 to 2014. Ms. Crane also contends that the pension plan was frozen at the end of 2007 and that the plan Recommended Decision Case No. 15-00261-UT 124 was fully funded at that time. Ms. Crane recommends an adjustment to the amount of prepaid asset to be included in rate base to remove losses incurred between December 31, 2007 and December 31, 2014. Crane Direct 53-56. The premise of the AG’s argument is the same as the premise of the argument that the New Mexico Supreme Court said was “incorrect.” Ms. Crane’s statement that her understanding is that “in the SPS case, the Court found that a prepaid pension asset had provided a ratepayer benefit” is correct. Crane Direct 61. What Ms. Crane does not correctly understand is that the Supreme Court found that the benefit existed because the utility made mandatory contributions to the pension fund that exceeded expenses and generated earnings that reduced the utility’s pension expense justified including the prepaid pension asset in rate base; the Court rejected the need for a more granularized analysis of whether ratepayers benefitted from the prepaid pension asset. AG v. PRC, 2015-NMSC-032, ¶ 19. NMIEC also opposes including PNM’s Prepaid Pension Asset in rate base. NMIEC witness Gorman contends that PNM failed to demonstrate that it has not fully recovered its actual pension contributions from customers and, therefore, PNM’s proposal to include the Prepaid Pension Asset should be denied. Gorman Direct 75. PNM rebutted NMIEC’s evidentiary claim in Mr. Peters’ Rebuttal Testimony, in which he attached Exhibit JAP-1, showing that the cumulative amount of excess shareholder cash contributions over the life of the prepaid pension asset is $188.2 million, which is higher than the $155.2 million included in PNM’s rate base. In determining the dollar amount of ratepayer contributions, Mr. Peters added the annual pension revenue requirements included in PNM’s cost of service in past PNM rate cases. Exh. JAP-1 to Peters Rebuttal. NMIEC argues that Exhibit JAP-1 does not meet PNM’s burden of proof because some of the past PNM rate cases were settled, and there is only an Illustrative Cost of Service available to support the total revenue requirement in those cases. NMIEC argues that an Illustrative Cost of Service “is not Recommended Decision Case No. 15-00261-UT 125 reliable data to determine the actual level of any specific expense item, such as an annual pension contribution, in the cost of service.” NMIEC’s Brief in Chief 12-13. In the 2010 PNM Rate Case, the approved revenue requirement was based on a stipulation, as modified by the PRC. The Signatories to the Stipulation did not agree on any single cost of service study to derive the proposed rates, nor, for the most part, did they agree on the appropriate level of items comprising the cost of service. PNM submitted an Illustrative Cost of Service for the purpose of demonstrating to the Commission that the proposed rates were just and reasonable and within the zone of reasonableness. PNM emphasized, however, that the Signatories did not agree on any single cost of service study. Certification of Stipulation 33. While the cost of service was illustrative, PNM, after issuance of the Final Order, adjusted its books to reflect the Illustrative Cost of Service as modified by the PRC. Tr. (4-15-16) 1293 (Monroy). NMIEC does not suggest how PNM could have alternatively calculated the amount of ratepayer contributions. Revenues received by a utility through rate recovery are not placed into separate FERC accounts, so it is not evident how PNM could trace the actual dollar amount of ratepayer contributions. In a case in which the PRC approves a rate increase based on an Illustrative Cost of Service, and does not explicitly approve a dollar amount for a particular expense, it is reasonable to rely on the dollar amount of the expense included in the Illustrative Cost of Service. ABCWUA recommends reducing the rate base amount for the Prepaid Pension Asset by about $28 million. ABCWUA’s Initial Posthearing Brief 34. ABCWUA’s recommendation is based on its argument that PNM’s cost-benefit analysis is faulty and therefore PNM’s determination of the “breakeven” point, where the amount of PNM’s pension expense decreases because of excess shareholder contributions, is incorrect. ABCWUA Witness Dittmer disputes PNM’s treatment of Accumulated Deferred Income Taxes (ADIT) in its cost-benefit calculation, Recommended Decision Case No. 15-00261-UT 126 and argues that because of PNM’s current and recent Net Operating Loss (NOL) carryforward position, customers are not receiving the cash benefit of the ADIT liability attributable to PNM’s pension contributions. Dittmer Direct 37.30 The premise of ABCWUA’s argument is the same as the premise of the argument that the New Mexico Supreme Court said was “incorrect” in AG v. PRC. ABCWUA’s argument hinges on its position that PNM did not correctly calculate the benefit to ratepayers in its cost/benefit analysis. While the PRC required a cost/benefit analysis in the 2007 PNM Electric Rate Case, the Supreme Court found that this type of cost/benefit analysis is not relevant to whether a prepaid pension asset can be included in rate base. As stated above, the Court made clear that its basis for approving including the prepaid pension asset in rate base is to allow “utilities to recover the costs of financing their business operations through the inclusion of investorsupplied working capital in the rate base.” Because Exhibit 1 to Mr. Peters’ Rebuttal Testimony shows by a preponderance of the evidence that PNM investors, like SPS investors in AG v. PRC, made contributions to the pension fund that were required by law and that “exceeded expenses and generated earnings that effectively reduced [PNM]’s — and consequently the ratepayers’ — pension expense . . . . the net prepaid pension asset is properly included in the rate base.” balanced draft should be excluded from PNM’s rate base. 30 In the early years of an asset, accelerated depreciation creates larger tax deductions than book expenses. For book purposes, the tax savings are deferred, not saved. The tax savings are credited to an account called Accumulated Deferred Income Tax (ADIT), which is a liability account that reduces rate base. The cash savings is viewed as a cost-free source of capital to a utility. Hahne & Aliff, § 17.01(1]. Use of accelerated depreciation can create not only an ADIT liability, but also an ADIT asset if a utility has negative taxable income as a result of taking large tax deductions created by accelerated depreciation. In this situation, the utility experiences a NOL, which may be carried forward to reduce taxable income in future periods. A utility records the NOL as an ADIT asset, which increases rate base. 2010 PNM Rate Case, Certification of Stipulation 67. Recommended Decision Case No. 15-00261-UT 127 XVIII. LAS VEGAS DECOMMISSIONING PNM seeks approval to begin recovering previously approved regulatory assets and liabilities relating to the decommissioning of the Las Vegas Generating Station, over two years. This request is unopposed and should be granted. PNM’s Initial Brief 119. XIX. A. NEW REGULATORY ASSETS AND LIABILITIES GENERAL PNM seeks approval of new regulatory assets and liabilities and for rate treatment of other existing regulatory assets and liabilities. A regulatory asset represents the amount of a current expense incurred by a utility that will be recovered in future rates. A regulatory liability represents an amount collected by a utility from ratepayers for an un-incurred future cost, which the utility owes to ratepayers. The purposes and consequences of creating regulatory assets and liabilities vary significantly. A commission may approve a regulatory asset only for the purpose of preserving a utility’s ability to request recovery for a cost in a future rate case; the commission defers a decision on whether the expense represented by the regulatory asset can later be recovered in rates. The regulatory asset is not included in rate base and the utility does not earn a return on the regulatory asset, i.e., accrue carrying charges. E.g., Application of Pacificorp for an Accounting Order to Establish a Regulatory Asset, No. PAC-E-08-02, 2008 WL 2588047 (Idaho P.U.C. 6-26-08). This treatment preserves a utility’s ability to recover an expense from ratepayers that would otherwise be financed by shareholders or bondholders. Tr. (4-27-16) 3027-31 (Crane). Alternatively, a commission’s approval of a regulatory asset can result in the asset being included in rate base over an amortization period and the utility earning a return on the Recommended Decision Case No. 15-00261-UT 128 regulatory asset over a period of time. The rationale for allowing the utility to earn a return on the regulatory asset is to compensate shareholders or bondholders for the time value of money for a cost that they have incurred before it is recovered from ratepayers. Tr. (4-27-16) 3031 (Crane). Another alternative is that a commission’s approval of a regulatory asset can result in the utility amortizing the regulatory asset as an expense in the period in which the utility receives revenues through rates to recover the regulatory asset but receiving no return of or on the asset. E.g., Application of Sierra Pac. Power Co., Docket Nos. 13-06002 & 13-06003, 2014 WL 576310, § VI(A) (Nev. P.U.C. 2-3-14). This is the treatment that the AG is recommending in this case for PNM’s rate case expenses. Tr. (4-27-16) 3026 (Crane). Yet another alternative is that a commission’s approval of a regulatory asset can result in the utility earning both a return on and return of a regulatory asset, resulting in a utility both earning a return on the regulatory asset and recovering amortization expense for the regulatory asset. If a regulatory asset is amortized, the total cost reflected in the regulatory asset is recovered over a period of years, usually on a straight-line basis. Pub. Util. Deprec. Practices 44. If the total cost reflected in a regulatory asset is not recovered through amortization expense by a utility’s next rate case, the unamortized amount generally is amortized in that next rate case. Once the total cost of the regulatory asset has been amortized, the regulatory asset is removed from rate base. If a utility is not allowed to create a regulatory asset, the costs incurred would have to be expensed in the year incurred. If incurred outside the Test Period in the utility’s next rate case, the utility loses the ability to recover the costs from ratepayers. 2010 PNM Abandonment Case, Recommended Decision 17. Excepting PNM’s requested regulatory assets for its proposed Credit Card Program and its impairment of state net operating loss, PNM asks to earn a return on its requested regulatory Recommended Decision Case No. 15-00261-UT 129 assets equal to PNM’s average weighted cost of capital as determined in this case. Tr. (4-11-16) 249 (Ortiz); Tr. (4-15-16) 1149-50 (Monroy). PNM also asks to earn a return of the regulatory assets over a five-year amortization period for the credit card regulatory asset and a two-year amortization period for its other requested regulatory assets and liabilities. Tr. (4-15-16) Monroy 1149. It appears that utility requests for creation of regulatory assets and liabilities are becoming more frequent. It also appears that this Commission has not discussed the circumstances in which creation of a regulatory asset or liability is appropriate. The significant evidence on this issue in this case creates a good opportunity for the Commission to provide guidance on this issue: the Commission shares the concerns expressed by AG witness Crane about creating regulatory assets. Ms. Crane said that the AG opposes creation of regulatory assets in principle because they insulate shareholders from risk and shift risk to ratepayers. Tr. (4-27-16) 2989 (Crane). Ms. Crane believes that a utility’s incentive to manage its business gets lost when regulatory assets are permitted. Tr. (4-27-16) 2995. Ms. Crane aptly observed: But I do find it ironic that you would go out and execute a buy-back of Palo Verde leases for $150 million or something without batting an eye, and yet you want to come in and have a regulatory asset for a couple of hundred thousand dollars in credit card fees. I mean, you are at risk, you should be at risk, and guess what? That’s why you get a rate of return that’s more than a risk-free rate. If you didn’t have any risk, then we should be talking about a treasury rate of return for the company. We’re not. Even Dr. Woolridge is not recommending that. And so, you know, there’s risk. You’re going to be compensated for that risk, and you need to take that risk instead of pushing everything off on ratepayers, you know, and trying to get — trying to get the Commission to guarantee recovery of costs that you haven’t even incurred yet. Tr. (4-27-16) 3003. The basis for Ms. Crane’s opposition to regulatory assets is well-founded. While this Commission does not oppose creation of regulatory assets and liabilities, they should be the exception, not the norm. Recommended Decision Case No. 15-00261-UT 130 B. ALVARADO SQUARE LEASE REGULATORY ASSET To reduce costs and increase employee efficiency, PNM in 2012 moved all of its employees out of the Alvarado Square building that it had been leasing and consolidated its downtown Albuquerque employees into the PNM Headquarters building located across the street. In vacating Alvarado Square, PNM incurred costs to demolish a skywalk that connected the Alvarado Square and Headquarters buildings, to separate the heating and cooling systems for each building to have stand-alone systems, and to construct improvements to the Headquarters building to accommodate the increased use and capacity of the facility, including remodeling each floor and installing cubicles. At the time, PNM’s parent company, PNM Resources Inc., owned leasehold improvements at the Alvarado Square building having a net book cost of $4,557,557, a portion of which is allocable to PNM for ratemaking purposes. Monroy Direct 43-46. For accounting purposes, PNM has already booked a regulatory asset for costs incurred in exiting Alvarado Square. Tr. (4-15-16) (1252) (Monroy). Here, PNM asks approval to create a regulatory asset for ratemaking purposes. PNM proposes to include in rate base $11.3 million of the improvements made to the Headquarters building as general plant, before corporate allocation, as well as a regulatory asset in the amount of $3.8 million reflecting the PNM Retail share of the Alvarado Square exit costs. Monroy Direct 44; Exh. HEM-4 WP RA-4 to Monroy Direct. The total costs deferred include the remaining balance of the Alvarado Square leasehold improvements as of December 31, 2012, as well as costs associated with removing the skyway between the Headquarters and Alvarado Square buildings, as well as the heating and cooling system modifications. PNM proposes to amortize and recover the Alvarado Square Lease Regulatory Asset over a five-year period to reflect the long-term nature of these relocation costs. Monroy Direct 44. The resulting annual amortization expense included in the Test Period revenue requirement is $766,997. PNM also Recommended Decision Case No. 15-00261-UT 131 includes the unamortized balance as of the end of the Test Period in rate base. Exh. HEM-4 WP OA-2. To justify the reasonableness of recovering the Alvarado Square Lease Regulatory Asset, PNM filed a cost-benefit analysis that PNM says demonstrates an overall net benefit of $1.4 million from the relocation taking into account the incremental costs to improve the Headquarters building and to vacate the Alvarado Square building. Monroy Direct 43; Exh. HEM-4 RA-5 to Monroy Direct. Staff and the AG oppose PNM’s proposed recovery of the costs reflected in the Alvarado Square Regulatory Asset, arguing that it is untimely and that PNM should have requested preapproval of the regulatory asset. The AG further argues that granting PNM’s request would be retroactive ratemaking. De Cesare Direct 8-9; Crane Direct 36. The AG removed from PNM’s revenue requirement the regulatory asset associated with Alvarado Square relocation costs, associated ADIT and amortization expense. Crane Direct 37. PNM argues that it is unnecessary to obtain pre-approval of a regulatory asset as a precondition to seeking cost recovery in a rate case. PNM’s Initial Posthearing Brief 125. However, in 2010, PNM, in connection with its request to abandon the Las Vegas Generating Station, did seek approval to create a regulatory asset to record future decommissioning costs and recovery the costs in its next rate case. 2010 PNM Abandonment Case, Recommended Decision 12. The AG’s retroactive ratemaking argument is dispositive and requires rejection of PNM’s request. In a Southwestern Public Service Company (SPS) rate case, the PRC rejected SPS’s request for approval of a regulatory asset based on the same argument made by the AG in that case. SPS proposed to include in rate base a regulatory asset of $3.5 million in historic demand side management expenses. In rejecting SPS’s request, the PRC said: From the record it appears SPS did not request the Commission’s permission to defer recovery of the claimed DSM expenditures. The 2003 correspondence with Recommended Decision Case No. 15-00261-UT 132 Staff does not rise to the level of, and is no substitute for Commission approval to defer costs or create a regulatory asset subject to recovery in a subsequent rate case. Consequently, inclusion of these historic DSM expenditures in rate base would constitute retroactive ratemaking. Therefore, SPS’s proposed adjustment to rate base to include pre-Test Period DSM expenditures is disallowed. 2007 SPS Rate Case, Recommended Decision 104. PNM’s request for approval of an Alvarado Square Regulatory Asset should be rejected and PNM’s revenue requirement should be adjusted to remove the regulatory asset associated with Alvarado Square relocation costs, associated ADIT and amortization expense. C. TIME OF USE REGULATORY ASSET PNM withdrew its original request to establish a regulatory asset to defer and begin recovering expenses to be incurred to re-program PNM’s Time of Use (TOU) meters to accommodate proposed changes in PNM’s TOU rates. Instead, PNM has included the cost in its Test Period revenue requirement. Tr. (4-15-16) 1307-08 (Monroy). D. RATE CASE EXPENSE REGULATORY ASSET PNM proposes approval of a regulatory asset in the amount of $3,790,023, representing its Test Period rate case expenses, and to amortize the cost of this asset over two years. This request should be approved, as explained below in Section XX(C). E. DEPARTMENT OF ENERGY REFUNDS REGULATORY LIABILITY PNM requests approval of to a regulatory liability to accumulate and pass on to customers refunds from the Department of Energy (DOE) received under a settlement agreement relating to spent nuclear fuel costs at Palo Verde. Under the settlement agreement, PNM received a settlement payment in October 2014 in the amount of $3,784,423, as reimbursement for costs incurred through June 2011. In addition, PNM received a second settlement payment in 2015 for costs incurred from July 2011 through June 30, 2014, in the amount of $2,117,844. PNM proposes that the jurisdictional portion of these refunds be credited back to PNM retail customers through base fuel over a two-year amortization period in Recommended Decision Case No. 15-00261-UT 133 the annual amount of $2,815,812. Monroy Direct 50-51; Exh. SAT-2 WP Fuel-3: Test COS Column R, to Taylor Direct. This request is unopposed. In light of the recommendation that PNM collect no fuel costs through its base rates, the total retail jurisdictional amount of the refunds ($5,631,624) should be returned to ratepayers through PNM’s FPPCAC. F. CREDIT CARD PROGRAM REGULATORY ASSET PNM requests approval of a regulatory asset for costs incurred to implement a recurring credit card payment program. Creation of the regulatory asset would allow PNM to defer collection of the costs until its next rate case, when it would propose to amortize and recover the deferred costs. PNM would record carrying charges for the regulatory asset based on the approved weighted average cost of capital in this case. PNM does not seek to recover any costs of the credit card program in this case. Ortiz Direct 55. Customers currently pay a $2.95 transaction fee for each payment made by credit card to PNM. PNM proposes to eliminate this fee for customers who sign up for automatic and recurring payments. For the 12 months ending March 31, 2015, 7.9% of PNM customer payments were made using credit or debit cards. Of the customer payments made using credit or debit cards, 9% were made by PNM customers who identified themselves as low income through the Low Income Heating Energy Assistance Program. Ortiz Direct 55-56. Credit card vendors have offered a discounted price for recurring credit card payments, and PNM expects that vendors will offer PNM a transaction fee of $1.50 for recurring customer credit card payments. Ortiz Direct 56. PNM expects that not charging the transaction fee to customers who sign up for automatic, recurring payments will cost PNM between $360,000 and $630,000 annually, based on a projected participation level of between 4% and 7% of customers. Ortiz Direct 57. Recommended Decision Case No. 15-00261-UT 134 Mr. Ortiz said that PNM probably would not implement the credit card program if the PRC does not approve a regulatory asset because shareholders would have to absorb the cost until the next rate case. Tr. (4-11-16) 248-49. Staff and the AG oppose PNM’s request for approval of a regulatory asset for costs associated with the Credit Card Program. Crane Direct 44-46; Gunter Direct 23-25. PNM’s request for approval of a regulatory asset for costs incurred to implement a credit card program should be denied. Accepting credit card payments is a convenience to those customers who pay their bills via credit card, and the cost associated with accepting credit card payments should be borne by only those customers who pay their bills via credit cards. Application of Wisconsin Power & Light Co. for Authority to Increase Rates, 242 P.U.R. 4th 193, 2005 WL 1861731, * 39 (Wisc. P.S.C. 7-19-05); cf. Park Water Co., Resolution No. W-4936, 2013 WL 265506, *4 (Cal. P.U.C. 1-10-13) (recovery of costs of credit card program from all ratepayers should not be permitted because utility did not show that option offered any net savings as required by statute). G. REGULATORY LIABILITY TO RECORD TIMING DIFFERENCE RELATING TO AROS AND COAL MINE RECLAMATION OBLIGATIONS An Asset Retirement Obligations (ARO) is a liability resulting from a legal obligation to retire or decommission a long-lived tangible asset. FERC Order No. 631, ¶ 2. An ARO can impose a significant liability that may not come due for many years. The wide disparity in how these liabilities were reported, or not, on balance sheets lead the Financial Accounting Standards Board (FASB) to issue guidelines on how AROs are to be calculated and recorded. FASB 143 was adopted into the Uniform Systems of Account by FERC in 2003 in Order 631. The PRC allows utilities to incorporate the cost of decommissioning plants into their rates by allowing them to recover costs associated with decommissioning over time before the retirement or obligation occurs. It is an expense paid by ratepayers for a cost to be incurred by a utility in the future related to plant in service now. Tr. (4-27-16) 3023-24 (Crane). Recommended Decision Case No. 15-00261-UT 135 Unfortunately, none of the witnesses included a discussion of the actual FASB approved ARO methodology in testimony. Mr. Monroy’s discussion of accretion expenses related to AROs is sufficiently vague to question whether it includes the original cost estimate for decommissioning, which it should not. Monroy Direct p. 52. AG witness Crane stated that decommissioning expenses can be recovered in two ways: via an accretion method or via a straight-line method. She said that, under the accretion method, decommissioning costs increase over time so that ratepayers in earlier years pay less than ratepayers in later years. Under a straight-line basis, decommissioning costs are the same annually. Crane Direct 47; Tr. (4-27-16) 3023-24 (Crane). However, the text of FASB 143 does not appear to discuss separate straight line or accretion methods of recovering ARO costs. On the contrary, it appears that the annual ARO expense is a combination of a straight line depreciation expense of the current cost of decommissioning and an accretion expense needed to cover added cost generated by the time value of money and inflation. Edison Electric Institute on FASB 143 (Oct 2002) p.14 – 24 Ms. Crane further said that PNM is currently using “the accretion method,” but provided no source for PNM’s authority to use the method, nor an explanation of how the method works. Ms. Crane said that the accretion method is the accounting treatment required under Generally Accepted Accounting Principles (GAAP), but not required to be used for ratemaking purposes. Crane Direct 47. However, the Hearing Examiner was unable to find any reference to an “accretion method” in GAAP reference guides. PNM proposes to change to the straight-line method of recovering decommissioning expenses related to AROs for its generating facilities, excluding Palo Verde.31 Again, the Hearing Examiner was unable to find any reference to a “straight-line method” in GAAP. These generating facilities are San Juan, Luna, Lordsburg, Afton, Reeves, Four Corners and Algodones. Exh. HEM-4 WP OA, p.1. 31 Recommended Decision Case No. 15-00261-UT 136 It is troubling that the parties appear to be claiming use of methodologies not actually included, much less approved, by FASB, FERC, or the PRC. It is as if the parties forgot to mention that the original estimate of the decommissioning cost is to be depreciated over the life of the associated asset in equal annual expenses. So it is unclear whether the references to accretion costs include the original decommissioning cost estimate, which should be depreciated.32 Aside from the issue of whether there are separate straight line and accretion methods for collecting ARO funds, the parties contest whether an accreting expense can or should be normalized into equal annual amounts for rate design purposes. PNM proposes to recover decommissioning expenses on a straight-line basis because, under the accretion method, these expenses increase as the balance of the ARO liability grows. PNM argues that recovering these expenses on a straight-line basis is more equitable to ratepayers because it properly matches cost causation with receipt of benefits. Monroy Direct 52. For the Test Period, a change in recovery from the accretion method to the straight-line method expense for generating facilities excluding PV would increase PNM’s decommissioning 32 The ARO annual expense is derived via a complex formula that estimates removal, decontamination, and remediation costs for the ARO, then projects that cost out to the removal date specified by the obligation. The estimated future cost is then discounted back to the date the associated asset was acquired or the ARO was instituted, using a credit-adjusted risk-free discount factor. This discounted cost (the removal cost, not the asset itself) becomes a depreciable asset and is depreciated over the life of the asset and credited to Account 108 the Accumulated Reserve for Depreciation in equal monthly accruals. Like other depreciation expenses, it is calculated and charged as a straight-line equal accrual monthly amount. See Financial Accounting Standards Bd., Statement of Financial Accounting Standards No. 143, ¶¶ A21, B32. Meanwhile, the build-up of cost associated with the time value of money and inflation, the accretion in cost, is treated as an operating expense and charged to Account 411.10. The accretion cost is the difference between the depreciation expense and the discounted present value of the ultimate cost of removal. The accretion cost climbs over time by the credit-adjusted risk-free growth factor. See id., ¶¶ B56-B58. A third component of the ARO annual expense is the amortization of the regulatory asset created by the accumulated past period depreciation and accretion costs incurred but not recovered prior to the establishment of the ARO expense. The regulatory asset accruals are split between depreciation and operating cost accounts. Recommended Decision Case No. 15-00261-UT 137 expense from $423,647 to $2,291,312, or by $867,665, for generating facilities excluding PV. PNM Exh. HEM-4 WP OA, p.5, row 8, to Monroy Direct. In connection with its proposal to change to a straight-line recovery method, PNM asks the PRC to approve creation of a regulatory liability in the amount of $1,423,647, to recognize the timing difference between recording the expenses on a straight-line basis and the accretion expense otherwise applicable under GAAP accounting. Monroy Direct 52-53; PNM Exh. HEM-4 WP RA, p.4 (row 23) to Monroy Direct. PNM’s coal mine reclamation obligations are not accounted for as AROs because PNM does not own the coal mines supplying SJGS. However, PNM currently recovers its coal mine reclamation costs similarly to how it recovers its decommissioning expenses related to AROs for its generating facilities, excluding PV. PNM proposes to change its recovery of coal mine reclamation expenses to a straight-line method as well. Monroy Direct 42. For the Test Period, a change in recovery of coal mine reclamation expense from the GAAP method to straight-line would increase this expense from $1,005,761 to $1,695,715, or by $689,954. Monroy Direct 67; PNM Exh. HEM-4 WP CMD, p.4, to Monroy Direct. In connection with its proposal to change to a straight-line recovery method for coal reclamation expenses, PNM asks the PRC to approve creation of a regulatory liability in the amount of $663,941. PNM Exh. HEM-4 WP CMD, p.2, to Monroy Direct. The AG opposes PNM’s proposal to change to a straight-line recovery method for both production plant decommissioning costs and underground mine decommissioning costs and argues that the accretion method better promotes intergenerational equity because its lower decommissioning costs in early years offset the higher return on rate base cost of the asset in earlier years. The AG also expresses concern that, under a straight-line recovery method, PNM would receive more revenues earlier that “are supposed to be earmarked for a long-term future Recommended Decision Case No. 15-00261-UT 138 obligation, without any requirement that it actually set those funds aside in the interim.” Crane Direct 47-49. PNM responds that the AG is conflating return of a capital investment (depreciation or accretion) with return on a capital investment. PNM argues that decommissioning expenses should be recovered in the same way as depreciation expense, i.e., on a straight-line basis. Monroy Rebuttal 29-30. ABCWUA also opposes PNM’s proposal to change to a straight-line recovery method for both production plant decommissioning costs and underground mine decommissioning costs and argues that it would overcharge current ratepayers compared to future ratepayers. Under a straight-line recovery, the annual decommissioning expense for generation facilities excluding PV would be $767,939. ABCWUA argues that $767,939 in year 2016 dollars is not the same as $767,939 in year 2015 dollars because a current dollar is worth several times a future dollar, so using straight-line recovery would overcharge current ratepayers. ABCWUA characterizes PNM’s proposal as “eliminating the time value of money.” Dunkel Direct 43. PNM correctly responds that its proposal does take into account the time value of money by proposing approval of a regulatory asset, which reduces PNM’s return on rate base and recognizes current customers’ contributions to the future cost. PNM explains, “Because this credit to rate base is based on a return at current dollars, customers are compensated for the time value of the advanced funds contributed.” Monroy Rebuttal 31. PNM’s proposal raises five concerns: 1) whether PNM is currently depreciating the original decommissioning cost estimate over the life of the associated asset; 2) whether PNM is estimating and allocating only the accretion costs associated with its ARO in an accreting manner; 3) whether normalization of the accreting costs into equal annual expenses is permitted under FASB and FERC approved accounting procedures; 4) whether the creation of a regulatory asset to record the timing differences between straight line and accreting expenses Recommended Decision Case No. 15-00261-UT 139 protects customers by withholding the sum from rate base and compensates for the time value of money; and 5) whether a separate account of some kind should be created to ensure that the funds collected over time are in fact available at the time they are needed. PNM’s proposals to change its methods of recovery decommissioning expenses related to AROs for its generating stations, excluding PV, and coal mine reclamation costs, and its related requests to create regulatory liabilities, should be rejected. In its next general rate case filing, PNM should make clear that the original cost estimate of the decommissioning costs is indeed being depreciated over the life of the associated assets, and that the accreting portion of the decommissioning costs include only the time value of money and inflation effects, i.e., the FASB methodology. The question of whether a separate decommissioning funds should be established was obliquely raised by Crane but not specifically requested. Since no party proposed a solution, none is offered here.33 H. REGULATORY LIABILITY TO RECORD EXCESS DEFERRED STATE INCOME TAXES PNM requests approval of a regulatory liability to record estimated excess deferred state income taxes. A phased-in reduction in the New Mexico corporate income tax rate has created a difference between the deferred state income taxes previously accrued in PNM’s books, which were calculated based on the prevailing tax rate, and the deferred taxes that will be paid out at the lower tax rates in effect after 2014. As of December 31, 2014, PNM estimated the amount of excess deferred income taxes to be $17.4 million. The final amount of the excess deferred income taxes depends on the actual ADIT balances in 2018, including ADIT accruals and reversals during the 2014-2018 phase-in period. Therefore, the final amount will not be known with certainty until after the 2018 tax year. PNM proposes to record the estimated amount as a regulatory liability and, after the amount becomes fixed in 2018, amortize and return the 33 See Accounting, Financial Reporting, & Rate Filing Requirements for Asset Retirement Obligations, Order on Rehearing, 104 FERC ¶ 61,183 (7-31-2003) (issue of external funding of amounts collected in rates for AROs should be resolved on a case-by-case basis in rate cases). Recommended Decision Case No. 15-00261-UT 140 balance to customers over a time period that reasonably reflects the period over which the actual deferred benefit of the lower rate will be realized by PNM. Harland Direct 38-40. Instead of waiting until the amount of excess deferred taxes becomes known, NMIEC Witness Gorman proposes that PNM start amortizing the estimated future excess deferred state income taxes over a five-year amortization period and include the annual amortization amount as a credit in the Test Period revenue requirement. Based on PNM’s $17.4 million estimate, Mr. Gorman concludes that his proposed adjustment would decrease ADIT by $3.5 million annually and, with a tax gross-up, would decrease the Test Period revenue requirement by approximately $5 million. Gorman Direct 81-82. PNM witness Harland persuasively rebutted Mr. Gorman’s proposal. Mr. Gorman’s proposal dismisses the fact that the actual excess deferred state income taxes are not yet known and measurable. Because of the long phase-in period, the actual amount of excess deferred taxes could vary substantially (up or down) from the amount currently estimated. Timing is a critical factor in determining the correct amount. Normal business operations, changes in the business, changes in federal or state law, and changes in the economy could substantially affect the amount of actual excess deferred taxes determined upon full implementation of the rate reduction. Harland Rebuttal 36-37. PNM will not realize the benefit of the state corporate income tax reductions until the phase-in is complete and the resulting deferred taxes related to the phase-in period are paid out, or reversed, over time. In a future rate case, PNM plans to propose that the Average Rate Assumption Method (ARAM), or a reasonable facsimile, be used to determine the timing of the return to customers. ARAM is the mechanism that the IRS required for the excess deferred income taxes created by the federal income tax rate reduction in the Tax Reform Act of 1986. Harland Rebuttal 40. Such a mechanism protects customers because the ADIT offset to rate base is not reduced by the effect of the rate change until the excess deferred income taxes are Recommended Decision Case No. 15-00261-UT 141 returned to customers through a reduction to income tax expense. Additionally, it protects the PNM by matching the timing of the return of the excess deferred income taxes to customers with the timing of the actual cash benefit received by PNM as the ADIT reverses at the future lower rate. Id. at 40. Customers will not be harmed by delaying the return of the excess deferred income taxes until the amount is known and measurable. Just as with any other ADIT, excess deferred taxes reduce rate base because they are a component of ADIT. This compensates customers for the difference in timing between when tax expense is recovered from customers and when it is paid out by PNM. The excess deferred taxes will only be removed from the ADIT in rate base when they are actually returned to customers. Harland Rebuttal 39. Therefore, customers are not harmed by any delay. PNM’s request for approval of a regulatory liability to record estimate excess deferred state income taxes should be granted. I. REGULATORY ASSET FOR STATE NET OPERATING CARRYFORWARD IMPAIRMENT LOSS PNM requests approval of a regulatory asset of $2,145,449 associated with the impairment of Accumulated Deferred Income Taxes (ADIT) associated with state Net Operating Loss (NOL) carryforwards. The impairment of the state NOL carryforward occurred in the Base Period when PNM was required to recognize on its books its inability to use all of its New Mexico NOL carryforwards before expiration of the statutory five-year carryforward period. The write-off of these impaired NOL carryforwards reflects a permanent book/tax timing difference in the accounting for deferred income taxes and results in a reduction to rate base in the Test Period revenue requirement. PNM is not seeking to include this regulatory asset in rate base. PNM requests authority to amortize and recover the impairment in the revenue requirement over two years. Harland Direct 35. Recommended Decision Case No. 15-00261-UT 142 In the early years of an asset, accelerated depreciation creates larger tax deductions than book expenses. For book purposes, the tax savings are deferred, not saved. The tax savings are credited to an account called Accumulated Deferred Income Tax (“ADIT”), which is a liability account that reduces rate base. The cash savings is viewed as a cost-free source of capital to a utility. Use of accelerated depreciation can create not only an ADIT liability, but also an ADIT asset if a utility has negative taxable income as a result of taking large tax deductions created by accelerated depreciation. In this situation, the utility experiences a Net Operating Loss (“NOL”), which may be carried forward to reduce taxable income in future periods. A utility records the NOL as an ADIT asset, which increases rate base. 2010 PNM Rate Case, Recommended Decision 67. In 2010, 2011, 2012, and 2014, PNM incurred NOLs because it generated deductions which exceeded its taxable income. Monroy Direct 19. When carried forward, the NOL is a temporary book/tax difference for which an ADIT asset must be recorded. In a revenue requirement study, this ADIT asset operates as an offset against the ADIT liability in the rate base to reflect that the company has not yet been able to realize that portion of the cash tax benefit represented by the ADIT liability balance. The sum of the ADIT liability created by the bonus depreciation and the offsetting ADIT asset created by the NOL carryforward represents the cash tax benefits that were actually received by the company. Harland Direct 19. Under New Mexico law, NOLs that were incurred before 2013 can only be carried forward five years. If the taxpayer does not have sufficient taxable income against which to offset the entire NOL carryforward, the unused portion expires and cannot be realized. This creates a permanent book/tax difference because the tax benefit of the unused deduction will never be realized. Under GAAP, once it is determined that it is unlikely that an NOL carryforward will be used before expiring, the company must impair, or write-off, the portion that it expects to expire unused. Harland Direct 35-36; Harland Rebuttal 9. Recommended Decision Case No. 15-00261-UT 143 Because of the repeated extensions of bonus tax depreciation, and especially the 2014 retroactive extension in December 2014, PNM determined that it would not be able to use all of the state NOL carryforward generated in 2010 before it expired in 2015. At that time, PNM was required by GAAP to impair the NOL carryforward by recording an impairment reserve against the portion of NOL carryforward ADIT asset related to the potentially unusable portion of the NOL carryforward. This increased 2014 income tax expense by $2,145,449 and created a reserve (liability) in ADIT that offset the NOL carryforward ADIT asset by the same amount. This effectively reduces the NOL carryforward ADIT asset to the amount of future tax savings the Company expects to realize. Harland Rebuttal 9. The NOL impairment was recorded at the corporate parent level and was allocated between the members of the consolidated group which contributed to the loss. Of the NOL total impairment loss, PNM seeks recovery of the 68.57% attributable to PNM’s losses, with no return on the unamortized balance, and with the remainder borne by shareholders. Harland Rebuttal 16-17. As a result of this write-off, PNM reduced the NOL carryforward ADIT tax asset in its Test Period revenue requirement by $2,145,449, the amount of the impairment loss. PNM reduced the NOL carryforward ADIT asset (and rate base) included in the cost-of-service by the full amount of the impairment reserve. With these adjustments, the ADIT asset reflects only the tax benefit of the NOL carryforward the Company expects to realize in the future. Harland Rebuttal 10. PNM argues that it should be allowed to amortize, and collect from ratepayers, the $2,145,449 because its ratepayers have benefitted from the bonus tax depreciation that created the impairment loss. It says that customers should not receive the benefit of bonus tax depreciation without also bearing the related expenses. Recommended Decision Case No. 15-00261-UT 144 PNM says that bonus depreciation provides a significant benefit to both shareholders and customers because the large amounts of ADIT created from bonus depreciation provide a source of government-supplied, cost-free capital to the utility. The utility benefits to the extent it avoids having to access the capital markets to raise capital to support its utility investments by issuing debt or equity securities, and customers benefit through the ADIT reduction to rate base in rate cases. Tr. (4-20-16) 1891 (Harland). To demonstrate the net benefit to customers in this rate case, PNM Witness Harland performed a cost-benefit analysis. Exh. MFH-1 to Harland Rebuttal. In that analysis, bonus tax depreciation, net of all costs including the impairment loss amortization, is shown to provide a substantial benefit to current and future customers. Specifically, the ADIT liability resulting from bonus depreciation, net of all costs, including the impairment loss and a return on the entire NOL carryforward ADIT asset, provides a revenue requirement benefit to customers of almost $2 million in the Test Period. This benefit will grow in the future as the NOL carryforward is used and the NOL carryforward ADIT asset is eliminated. Harland Rebuttal 5. This analysis also shows that PNM has recorded almost $235 million of tax depreciation ADIT liability related to bonus tax depreciation. Of this amount, $108 million is the incremental increase in the ADIT liability above what would have been created if PNM used normal MACRS tax depreciation, far in excess of the $83 million of NOL carryforward ADIT asset recorded.34 In other words, the bonus depreciation tax deductions have far exceeded the deductions that are being carried forward as NOL. ABCWUA Witness Dittmer, AG Witness Crane, and Staff Witness De Cesare all contend that the impairment loss on the New Mexico NOL carryforward should not be recovered from In the long-term, this NOL carryforward ADIT asset will reverse, and a larger benefit will inure to customers from the bonus tax depreciation ADIT liability. PNM-44 (Harland) at 11:12-14. 34 Recommended Decision Case No. 15-00261-UT 145 customers, and that an adjustment should be made to remove the NOL impairment loss amortization from the Test Period revenue requirement. ABCWUA Witness Dittmer agrees that ratepayers benefit from bonus depreciation, but argues that shareholders have benefitted to a far greater extent than customers from the bonus tax depreciation deductions taken from 2011 through 2014, the period between PNM’s last rate case and the time of the impairment. Dittmer Direct 22. Mr. Dittmer’s claim that PNM shareholders have benefitted in “multiples” greater than customers from the bonus tax depreciation is based on his personal assessment of what has transpired since PNM’s last rate case in 2010. Tr. (4-25-16) 2774. He states that “this is the first rate case wherein ratepayers will significantly and directly benefit from bonus depreciation elections taken in calendar years 2011 through 2014.” Dittmer Direct 22. While Mr. Dittmer argues that “PNM’s shareholders have benefitted to a larger extent from Bonus Depreciation than ratepayers,” he has not performed any calculation to support this claim. Tr. (4-25-16) 2772. Mr. Dittmer’s argument should be rejected. Generally, events occurring between Test Periods in rate cases are ignored. “[T]here are things that happen in intervening periods all the time that do not get captured going forward unless there is a special deferred accounting mechanism that gets established.” 2010 PNM Rate Case, Recommended Decision 69. AG witness Crane makes two arguments for disallowing the NOL impairment loss amortization: (1) she suggests that the customer benefit from ADIT has been “eliminated” by the NOL carryforward; and (2) she contends that the NOL impairment loss amortization should be disallowed because customers get no income tax expense benefit for bonus tax depreciation. Crane Direct 50. Ms. Crane’s arguments should also be rejected. The fact that PNM is currently in a NOL situation and is not making any tax payments to the IRS does not change the proper analysis Recommended Decision Case No. 15-00261-UT 146 when income tax normalization, which the Commission has long adopted, is considered. The benefit of bonus tax depreciation has not been eliminated by the NOL carryforward. The NOL carryforward ADIT asset offsets only the portion of the bonus tax depreciation ADIT liability that has not yet been realized. It does not eliminate the total benefit because it does not offset 100% of the bonus tax depreciation ADIT liability. PNM’s Initial Posthearing Brief 147. This is shown in Mr. Harland’s cost-benefit analysis, where the bonus tax depreciation ADIT liability, net of foregone MACRS depreciation ADIT, is ($108,181,079) in the Test Period, and the NOL carryforward ADIT asset is $83,333,832. Customers are receiving a rate base reduction of almost $25,000,000 from the incremental bonus tax depreciation above and beyond the reduction from normal MACRS tax depreciation. The ADIT benefit is not being eliminated. Additionally, the bonus tax depreciation benefit to future customers will increase as the NOL carryforward is used and the NOL carryforward ADIT asset disappears, leaving the entire depreciation ADIT liability as a reduction to rate base. Harland Rebuttal 17-18. Ms. Crane’s complaint that customers get no income tax expense benefit from bonus depreciation is a complaint about income tax normalization, which is required by federal law. 2010 PNM Rate Case, Certification of Stipulation 66. The alternative to full income tax normalization is the flow-through method, where the utility’s tax payments, as opposed to its tax liability, are the basis for the Test Period tax expense included in the revenue requirement. With income tax normalization, on the other hand, the benefit of bonus depreciation flows through to customers in the form of a reduction to rate base. While customers receive no direct reduction in income tax expense as a result of bonus tax depreciation, they do receive a substantial reduction in equity and debt return on rate base and an indirect income tax expense benefit thereon. This is shown in Mr. Harland’s cost-benefit analysis, where the Test Period reduction in return on rate base from the rate base effects of bonus tax depreciation, net of the NOL carryforward, is $2,038,190, and the direct income tax benefit thereon is $836,540. Harland Recommended Decision Case No. 15-00261-UT 147 Rebuttal 18-19 & Exh. MFH-1 to Harland Rebuttal. Moreover, the fact that the benefit comes from a reduction in rate base or a reduction of recoverable expenses misses the point. What is relevant is that customers have benefitted significantly from the bonus depreciation that gave rise to the state NOL carryforward that became impaired. Staff Witness De Cesare argues that because the NOL expired and is unused, it had no benefit to customers and, therefore, recovery of the cost should not be allowed. De Cesare Direct 25. Mr. De Cesare also contends that risk of NOL carryforward expirations is accounted for in PNM’s authorized rate of return. Staff Witness De Cesare addresses only the impairment loss without considering bonus tax depreciation in concluding that, because the NOL expired unused, it had no benefit to customers. The cost of the NOL impairment loss cannot be addressed without addressing the benefit to customers from this underlying bonus tax depreciation. All factors creating the impairment must be addressed in totality in order to determine if there was a net benefit to customers resulting from the decision to take bonus tax depreciation. As stated above, bonus tax depreciation, net of all costs, including the NOL impairment loss, creates a substantial benefit for customers in this case and for many years following. Mr. De Cesare’s argument that risk of NOL carryforward expirations is accounted for in its authorized rate of return is also without merit. The impairment loss is a direct cost of PNM’s decision to take bonus tax depreciation, which provides a substantial net benefit to current and future customers. It is no different than any other recoverable cost incurred in the provision of electric service to customers. Additionally, as stated earlier, PNM reduced the NOL carryforward ADIT asset by the entire PNM portion of the NOL impairment loss, and is not seeking a return on the unamortized portion. Allowing creation of a regulatory asset for the impairment loss is consistent with the PRC’s decision in the 2010 PNM Rate Case, in which the PRC ordered PNM to adjust its revenue Recommended Decision Case No. 15-00261-UT 148 requirement to reflect not only the ADIT liability created through using bonus depreciation, but the ADIT asset created as a result of the NOL and PNM’s loss of use of the Domestic Production Activities Deduction (DPAD), which is only available to reduce current taxable income and cannot be carried forward. 2010 PNM Rate Case, Recommended Decision 67. PNM’s request to create a regulatory asset in the amount of $2,145,449 associated with the impairment of Accumulated Deferred Income Taxes (ADIT) associated with state Net Operating Loss (NOL) carryforwards, should be granted and PNM should be allowed to amortize the amount over two years. XX. EXPENSES A. DEPRECIATION RATES PNM proposes changes to its depreciation rates for its production, transmission, distribution, and general plant in accordance with a depreciation study supported by witness Dane Watson of the Alliance Consulting Group, which provides depreciation consulting to the utility industry. NMIEC and ABCWUA oppose many of PNM’s proposed depreciation rates. ABCWUA’s depreciation witness is William Dunkel of William Dunkel and Associates. NMIEC’s depreciation witness is Brian Andrews, a consultant with Brubaker & Associates. PNM’s total Base Period depreciation expense is $90,491,546. Exh. HEM-3 COS Base Allc, p.10, row 482, column H. Its proposed total Test Period depreciation expense is $126,885,928. Exh. HEM-3 COS Test, p.12, row 482, column J. This represents a $36,394,382 increase in total depreciation expense. PNM says that $20,602,731 of this increase arises from its proposed changes in depreciation rates. PNM’s Initial Posthearing Brief 151 n.154. NMIEC recommends reducing the depreciation rates for 10 transmission and distribution accounts with a 2013 balance greater than $75 million, which reduces the Test Recommended Decision Case No. 15-00261-UT 149 Period depreciation expense by $13.3 million, to $113.6 million. Andrews Direct 3. Mr. Dunkel recommends a $20.3 million reduction in PNM’s requested depreciation expense, to $106,519,790. ABCWUA’s Initial Posthearing Brief 55; Exh. JRD-3, p.13, to Dittmer Direct. 1. PNM’S EXISTING DEPRECIATION RATES PNM’s existing depreciation rates, which differ for PNM North and PNM South assets35, were approved in multiple cases. Most of the current PNM North depreciation rates took effect in 2004, and are based on a study filed by PNM in 2003 that developed depreciation rates as of December 31, 2002. Tr. (4-20-16) 1999 (Peters). The 2003 study is PNM Exhibit 23. Some of the depreciation rates that took effect in 2004 have been revised, not as the result of approval of a new study, but in isolation. Revised depreciation rates for the SJGS took effect in 2006 based on an extension of the life of that plant to 2053. Revised depreciation rates for the Four Corners Plant took effect in 2007 based on a 2031 retirement date for that plant. Revised depreciation rates for Reeves Generating Station and PV took effect in 2009. Tr. (4-2016) 1999-2000 (Peters); Revised Appendix D to Watson Direct. The revised depreciation rates for PV reflect the Nuclear Regulatory Commission’s then-expected grant of an application for a PV license extension to 2046. 2008 PNM Rate Case, Final Order 33-34. Current depreciation rates for PNM South assets took effect in 2006. Watson Direct 11; Tr. (4-20-16) 2000-01 (Peters). 2. PNM’S PROPOSED DEPRECIATION RATES — IN GENERAL PNM proposes in this case to consolidate depreciation rates for PNM North and PNM South. PNM’s current and proposed generation plant retirement dates are in Revised Appendix B to Mr. Watson’s Direct Testimony. Tr. (4-20-16) 2001-02 (Peters). See pages 1-2, 31-32, and 131-136 of the Certification of Stipulation in the 2010 PNM Rate Case for history of the PNM North and South service territories. 35 Recommended Decision Case No. 15-00261-UT 150 PNM supports its proposed depreciation rates with a Depreciation Study that reflects assets recorded in PNM’s books as of December 31, 2013, plus post-2013 generation additions. Watson Direct 9-11. PNM’s Test Period depreciation expense is based on PNM’s proposed depreciation rates from the 2013 Study and monthly asset balances for the Test Period ending September 30, 2016. Watson Direct 4. Mr. Watson estimates that adoption of PNM’s proposed depreciation rates would increase PNM’s annual depreciation expense for existing and new plant by $35.1 million, from $99 million to $134 million. Watson Direct 2, 4. This increase reflects: x x x x x a $8.7 million increase in Production a $3.1 million increase in Transmission a $3.6 million increase in Distribution a $9.4 million increase in General $10.3 million for Rio Bravo, La Luz, 2014 and 2015 Solar, and the acquisition of 64.1 MW in PV 2.36 PNM does not currently recover depreciation expense for these assets. Tr. (4-13-16) 790 (Watson). Watson Direct 17-18. 3. PUBLIC UTILITY DEPRECIATION CONCEPTS PRC Rule 17.3.340.10 requires public utilities to record their plant investment activity in individual plant accounts as set forth in FERC’s Uniform System of Accounts and to set depreciation rates by account or subaccount. Additions, retirements, transfers, adjustments, and balances to individual plant accounts must be recorded and the records maintained. Depreciation, as defined by FERC, is “the loss in service value not restored by current maintenance, incurred in connection with the consumption or prospective retirement of electric plant in the course of service from causes which are known to be in current operation . . . .” Andrews Direct 4. This loss in service value occurs over several accounting periods and 36 These assets total $451.5 million in investment. Watson Direct 10. Recommended Decision Case No. 15-00261-UT 151 represents the consumption of the investment over time. General accounting rules provide for the recovery of this loss in value through periodic depreciation charges. Because public utilities are so heavily capital intensive, depreciation expense is often a utility’s greatest expense and a major driver of the revenue requirement. Conceptually, depreciation is a simple mathematical exercise of dividing the investment by some number representing estimated life; in fact, deriving depreciation rates and charges is often shrouded in complex analyses, uncertain future values, imperfect databases, and a substantial amount of judgment on the part of the analyst. Such are the issues in this docket. Public utility property accounts are made up of hundreds of thousands of transactions that record the build-up of plant in service, as well as the retirements, transfers, costs of removal, and salvage over time. Distilling this data into a coherent life estimate requires various actuarial analysis methodologies that convert the raw data into average service life and average remaining life estimates. Arcane differences in the type, quantity, and quality of the data, as well as the pattern of retirement dispersion, can give rise to substantial differences in life estimates and annual depreciation accruals. PRC Rule 17.3.340.11 allows utilities to use any reasonable acceptable method for estimating service lives, which includes analysis of plant mortality by group accounts. However, it suggests that utilities use recognized methods, which it lists as Actuarial Methods, Simulated Methods and the Life Span Method. The Rule says, “Where utilities lack sufficient records to develop actuarial data, the use of simulated methods . . . are recommended.” 17.3.340.11(C)(1) NMAC. 4. PRODUCTION PLANT PNM’s production plant is made up of steam and nuclear generating facilities and various gas powered generating facilities. Parties contest the proposed depreciation rates for only three power generating stations – Four Corners, Luna, and the 2011 vintage Solar unit. Recommended Decision Case No. 15-00261-UT 152 a) Four Corners Termination Date The remaining life of the Four Corners Plant is derived using the Life Span Method, which is a method of determining the service life of assets that are expected to be retired at one time as a single unit. Mr. Watson estimated terminal retirement dates for the various generating units based on consultations with PNM management, financial, and engineering staff. Watson Direct 20. No one challenges Mr. Watson’s use of the Life Span Method to analyze the life characteristics of Production Plant, but the termination date used in the methodology is challenged. ABCWUA disputes PNM’s choice of 2031 as the terminal date of the Four Corners Production Plant (Exh. DAW-2, Revised Appendix D, to Watson Direct), and recommends 2041 as the terminal date. Dunkel Direct 6. Mr. Watson said that he relied “specifically and solely on the company to provide that 2031 number to me.” Tr. (4-13-16) 727. ABCWUA is correct that the preponderance of the evidence shows that 2041 is the appropriate retirement date. First, Mr. Watson said that he relied on discussions with PNM personnel to select 2031. Watson Direct 20. However, Mr. Watson’s own Interview Notes say, under the heading, “Four Corners”: “Use 2041 for retirement date for [Unit] 4” “Retirement date is 2041, in service in the early 1970s (both 4 and 5)” Under the heading, “Recap,” the Notes further say: “Four Corners extend date 2041.” Exh. WDA-2 at 5. Second, at the hearing, ABCWUA’s attorney showed Mr. Watson an August 12, 2014 PowerPoint presentation to the Rate Case Policy Team titled “Depreciation Study Update — Recommended Decision Case No. 15-00261-UT 153 History and Current Assumptions.” On the slide titled “2013 Assumptions — Production Plant,” a bullet says: x Four Corners — Decommissioning date updated to 2041 Corresponds to lease with TNN and approximates APS date of 2038 APS intends to extend beyond 2038 in next case ABCWUA Exh. 12. Third, PNM’s Form 10-K for the year ending December 31, 2015, says: The Four Corners owners executed amendments to the agreements governing the operations of Four Corners that would extend those agreements until July 2041. The amendments are expected to become effective upon APS’ purchase of the ownership interest of EPE, which is anticipated to occur in July 2016. Staff Exh. 1, p.27. Fourth, PNM’s July 2014 Integrated Resource Plan says that the retirement date for the Four Corners’ Units is “[a]fter 2033.” Exh. WDA-3 to Dunkel Direct. b) Luna and Solar-2011 After initial disagreements over the terminal year for the Luna Generating Plant and the Solar 2011 compressors, PNM revised its depreciation study to reflect adjusted termination dates for these facilities. 5. TRANSMISSION AND DISTRIBUTION (T&D) PLANT PNM’s Depreciation Study relies on a Simulated Plant Record (SPR) Method of estimating the life of PNM’s Transmission & Distribution (T&D) Plant. Watson Direct 19. For the T&D Accounts, Mr. Watson performed both an actuarial analysis and SPR analysis to determine average service lives; however, Mr. Watson relied only on the SPR Analysis for his T&D depreciation rate recommendations. Tr. (4-13-16) 777 (Watson). Witnesses Dunkel and Andrews oppose Mr. Watson’s use of the SPR method, arguing that the actuarial method is the preferred approach and could have been used. Based on critical flaws in Mr. Andrews’ and Dunkel’s Testimonies, as described below, little weight should be given to their arguments and Recommended Decision Case No. 15-00261-UT 154 Mr. Watson’s depreciation rate recommendations, as described in the following discussion, should be adopted. a) Actuarial Analysis Actuarial analysis defines the retirement characteristics of a subject population by examining the age at which units in the population retire from service. Actuarial analyses rely on, and require, both the installation year and the retirement transaction year so as to construct “aged data” or “dated retirements,” which show the age of each asset upon retirement. If the year of installation (vintage) and year of retirement are known for most utility assets over time, the actuarial method is “generally considered the preferred approach.” Public Utils. Depreciation Practices 111. The actuarial analysis produces an Original Life Table (OLT) which, when graphed, illustrates the pattern of plant survivorship over the lifespan of the typical group unit. The OLT is then compared to a series of prototype curves for use in forecasting future retirements and average service lives. Much of the research establishing actuarial analysis was done at the Engineering Research Institute of Iowa State University in the 1910s through 1930s. Statistical Analysis of Industrial Property Retirements (referenced in Public Utils. Depreciation Practices 124). These studies discovered patterns of retirements that, when graphed, resemble ski slope shaped survivorship curves. Being survivorship curves, they all start out at 100% surviving and end with 0% surviving. The prototype curves that came out of the Iowa studies all presume and enforce a retirement trajectory that eventually retires 100% of the plant. If no age groups have completed the full life cycle of all members, the survivor curve will appear to be a “stub” of a curve. For example, if the oldest vintage is 60 years old and some plant is still in service, the survivor curve will reflect that the curve has not yet reached zero surviving, i.e., it will be a stub curve. Comparing the stub OLT curve to the Iowa Curves can be Recommended Decision Case No. 15-00261-UT 155 problematic depending on the percent of survival achieved. Public Utils. Depreciation Practices 120. It follows that when the data indicate few retirements relative to the initial installations, the survivorship percentage will likely be high over an extended period. The OLT curve will appear as a high plateau stub curve over many age groups. Similarly, if data is missing from the analysis, such as several experience years, the effect is as if the retirements for that period did not take place, which places the data in a stub curve condition. Watson Rebuttal 4. The original installations are still incorporated but the missing retirements are not. The result is a high survivorship ratio over an extended period. If the missing data is from the account’s early years, the effect is as if the account had no retirements of young plant. Again, the OLT curve will appear as a high plateau stub curve over many age groups. For all three of these examples (stub curves, few retirements, and missing data), the survivorship curve for the OLT will appear as a plateauing line with a high surviving range, i.e., a ski slope that never quite reaches the slope. A forecast of survivorship based on data suggesting a high plateau of survivorship will naturally result in a prediction of high survivorship over the long run, i.e., a longer average service life leading to a lower depreciation rate. Given (1) the missing data; (2) the predilection of missing data to generate longer average service lives; (3) the “ludicrously” long service lives forecast by the actuarial method; and 4) even the advocates of actuarial analysis, in the end, abandoned it, the actuarial model raises sufficient concerns as to its accuracy to nullify its reliability in this case. b) Missing Data In this docket, the available actuarial data covers only the last ten years of annual retirement transactions. While these retirements are associated with plant installed over a 65year range, it represents only the last ten years of activity over that range. All of the previous Recommended Decision Case No. 15-00261-UT 156 retirements from 1948 through 2002 are missing — as if they never occurred. Consequently, it follows that the actuarial construction of the OLT derives a survivorship percentage significantly higher for each age group than it would if all those retirements had been incorporated. This creates a high plateau survivor curve over a long range of age groups and mathematically produces a high average service life forecast, as shown in Exhibit BCA-1 to Mr. Andrews’ Direct Testimony. Mr. Watson acknowledged that, in the 2010 PNM Rate Case, PNM filed a depreciation study, performed by a different firm, in support of new depreciation rates that relied on an actuarial analysis for the T&D accounts. That depreciation study was based on a 1994-2009 (15 years) experience band, five years longer than the experience band available to Mr. Watson. Andrews Direct 12. But Mr. Watson said that the data from 1994-2002 used in that study could not be verified or trusted, to either PNM’s books or its FERC Form 1 filing, and that only data from 2003-2009 could be verified. For that reason, PNM and Alliance decided to rely only on 20032009 data for the actuarial analysis. Watson Rebuttal 25; Tr. (4-13-16) 777 (Watson). Mr. Watson said that there simply was not enough data. Exh. DAW-2, p.9, to Watson Direct. Where Mr. Watson believed there was sufficient aged data to obtain reliable results, as in the general plant accounts, he did use actuarial analysis. Watson Rebuttal 14, 16. While Mr. Watson did not quantify how many years of experience are required, when pressed by the Hearing Examiner at the hearing, he did state that ten is not enough for a 50 to 60-year average service life asset. Tr. (4-13-16) 764. Mr. Dunkel opposes Mr. Watson’s use of the SPR Method, explaining, “In this proceeding there is no valid reason to create, and rely upon, ‘simulated’ plant records, because we have the actual PNM ‘aged’ plant record data.” Dunkel Direct 21. Mr. Dunkel asserts that the 2003-2013 experience band is sufficient to produce a reliable survivor curve analysis. Recommended Decision Case No. 15-00261-UT 157 Dunkel Direct 22; Tr. (4-13-16) 2723. He said that the most recent practices are most relevant for estimating lives in the future. Dunkel Direct 22. The question of whether the amount of data in hand is sufficient, or not, to complete and reliably use the results of the survivor curve studies is central to the depreciation arguments in this docket. Tr. (4-13-16) 753 (Watson). Unfortunately, there is no defined number that answers to this question. Rather, the answer, to a great extent, is informed by a depreciation analyst’s judgment. Probably for this reason, there is little guidance in literature. None of the witnesses provided guidance as to the threshold of how much data is necessary to provide a reliable actuarial analysis. Nor did any witness provide guidance as to why the lack of data might or might not influence the results of the study. c) Telling Departures It is telling that despite their assertions that PNM’s actuarial data was adequate to perform a reliable actuarial survivor curve study, both witnesses Dunkel and Andrews departed from the actuarial analysis in selecting an average service life and survivor curve with which to develop the actual depreciation rates. Tr. (4-25-16) 2739- 2740 (Dunkel), 2823 (Andrews). In no instance did Mr. Dunkel or Mr. Andrews accept the best fit results of the actuarial analysis as appropriate to use in the calculation of the T & D depreciation rates. Instead, chose another service life estimate. Depreciation literature supports the heavy reliance on the depreciation analyst’s expert opinion and other non-statistical factors in selecting the actual remaining life to be used in the depreciation calculations. However, the reasoning behind witnesses Dunkel’s and Andrews’s selections leaves much to be desired. More often, they simply stated that the indicated service life was unreasonable, and then inserted their own service life estimate with no supporting discussion. Recommended Decision Case No. 15-00261-UT 158 Mr. Andrews’ justification in account after account was that his was a “good middle ground” selection without, generally, comment on why the actuarial best fit should be rejected. Mr. Dunkel used just one account as an example of his analysis — Account 364: Poles, Towers, and Fixtures. Dunkel Direct 15. He concluded that, “For the reasons discussed in this testimony, I recommend a 65-year average service life . . . .” Dunkel Direct 16. But nowhere in his testimony does he support the 65-year selection, which flies in the face of the result of the actuarial study’s indication that a 98.89-year average service life is the appropriate result. Exh. BCA-1, p.5, to Andrews Direct. As to all the other accounts where the selected service life differs from the actuarial study results, Mr. Dunkel offers no support whatsoever. Mr. Andrews’ actuarial analysis of Account 366 (Distribution Conduit) is illustrative. Andrews Direct 46. Mr. Andrews found, and Mr. Watson agreed, that the best fit to this Account using an actuarial analysis produces an ASL of 615 years, meaning that some conduit would last 1,325 years. Mr. Andrews deemed the 615-year ASL “unreasonable,” while according to Mr. Watson, this result “is ludicrous.” Tr. (5-13-16) 779. Mr. Andrews said that when an actuarial analyses produces a long life such as 615 years: That’s where the judgment would come into play, and you would have to use experience and good judgment to determine a life that’s reasonable for the property in that account. .... Well, that’s where the judgment comes into play and you don’t rely 100% on the results from the actuarial analysis. Tr. (4-25-16) 2823, 2825. As noted above, the literature gives wide sway to the expert opinion of the practitioner, but neither Mr. Dunkel nor Mr. Andrews provides even a shred of expert opinion for their selection and in fact deride Mr. Watson for giving credence to the background information provided by company operation personnel. Andrews Direct 11, 21-22, 25. Yet Mr. Andrews conceded that “informed judgment” must be used to select the appropriate survivor curve, based Recommended Decision Case No. 15-00261-UT 159 on analysis of actual data and interviews with PNM employees. Andrews Direct 18. But he does not do so himself. The actuarial approach may well be the better methodology but witnesses Dunkel and Andrews don’t appear to have actually applied the actuarial method in this case. They discuss the method at length, but in the end, their recommendations are not based on any actuarial study results. Nor do they support their departures from the actuarial study indications. Consequently, their recommendations should be given little weight. d) Simulated Plant Record Analysis SPR analysis is a commonly accepted approach used to determine the mortality characteristics of utility property. It is used when vintage transactional data is unavailable or limited. Watson Direct 9. It requires, and relies on, for each account, the plant additions for each year and the ending year balance. Tr. (4-13-16) 766 (Watson). PNM’s 2003 Depreciation Study used the SPR analysis for T&D accounts. Tr. (4-13-16) 763 (Watson); PNM Exh. 23, pp. 35, 9 n.5, In SPR, an Iowa Curve and an average service life are selected as a starting point and applied to the actual annual additions to produce a series of annual balance totals for each account. Exh. DAW-2, p.9, to Watson Direct. The balances are compared with the actual surviving balance using graphical and mathematical analysis. The Iowa curve and ASL that best fit the data are used to develop the average remaining life for the depreciation rate calculations. Test statistics are used to gauge the accuracy of the predictions. The primary statistic is the Conformance Index (CI), a sum of squares test that magnifies differences and points toward the best fit curve. Exh. DAW-2, p.11, to Watson Direct. e) Mr. Watson’s Analysis Mr. Andrews alleged that Mr. Watson’s recommended curves are not the best fit even under the SPR analysis. He said that Mr. Watson’s SPR analysis produced curves with higher Recommended Decision Case No. 15-00261-UT 160 CIs and more acceptable Retirement Experience Indexes than the curves selected by Mr. Watson. Tr. (4-25-16) 2836-40. Mr. Andrews said that many of Mr. Watson’s life-curve combinations are different from the combination that ranks highest for a particular account. NMIEC says that PNM’s proposed lives are not supported by actual data but are largely based on informal conversations with PNM personnel. Andrews Direct 12-13, 21-22, 25, 29, 33, 37, 41. Depreciation literature is replete with admonitions that the depreciation analyst must incorporate other information into derivation of service lives. Mr. Watson prefaces his study with just such a statement: Any depreciation study requires informed judgment by the analyst conducting the study. A knowledge of the property being studied, company policies and procedures, general trends in technology and industry practice, and a sound basis of understanding depreciation theory are needed to apply this informed judgment. In this depreciation study, judgment was used in areas such as survivor curve modeling and selection, depreciation method selection, simulated plant record method analysis, and actuarial analysis. Where there are multiple factors, activities, actions, property characteristics, statistical inconsistencies, property mix in accounts or a multitude of other considerations that affect the analysis (potentially in various directions, judgment is used to take all of these considerations and synthesize them into a general direction or understanding of the characteristics of the property. Individually, no one consideration in these cases may have a substantial impact on the analysis, but overall, the collective effect of these considerations may shed light on the use and characteristics of assets. Judgment may also be defined as deduction, inference, wisdom, common sense, or the ability to make sensible decisions. There is no single correct result from statistical analysis; hence, there is no answer absent judgment. Exh. DAW-2, pp.12-13, to Watson Direct. Mr. Watson has given himself wide latitude to depart from the statistical analysis alone. And no one contested his right to do so. Mr. Watson’s testimony provides ample evidence that his statistical analysis was tempered with information provided by PNM’s operating personnel. Watson’s departures from the mathematical indications of the SPR model appear fully supported and reasonable. Recommended Decision Case No. 15-00261-UT 161 f) Banding Placement bands are the number of years of plant installations included in the data. A placement band tells you how much of your plant history is included in the analysis. Experience bands are the number of years of transactions available relating to the placement band installations. The experience band tells you what happened to the placement band assets. Mr. Dunkel consistently conflated the relevance of placement data and experience data in explaining his actuarial analysis. At the hearing, when asked whether he agreed that PNM’s aged database was from 2003-2013, Mr. Dunkel responded, “No.” Mr. Dunkel indicated that the aged database included the data in the placement bands. Tr. (4-25-16) 2733-34. Mr. Watson explained that assets with lives longer than 40 or 50 years will not start to retire in a statistically meaningful way until much closer to their average lives than at age 11. Because so few of such assets would be retired in their first 11 years, he believes that using any retirements from the first 11 years to predict the average lives of the assets would be “very difficult or meaningless[.]” Watson Rebuttal 17. When asked whether Mr. Dunkel wasn’t on “the high end” of the three, five or ten year bands, Mr. Watson responded, “No,” that Mr. Dunkel (i) was able to analyze only one band; (ii) was not able to use rolling bands; and (iii) lacked data to analyze the conduit account through 40 to 50 years of 10-year bands. Tr. (4-13-16) 690, 694. Other than noting that the 2003–2013 set of data is a subset band of the total possible data set, there was little discussion of any need to segment the data. There is nothing in the record supporting an affirmative need to examine a subset, or band, of retirement data, as suggested in Public Utilities Depreciation Practices. g) Mr. Dunkel’s Inconsistencies Recommended Decision Case No. 15-00261-UT 162 Mr. Dunkel’s statements regarding the scope of the data, the preferred survivor curve methods, the sufficiency of the amount of data, and his incorporation of data and methods in his analysis were often incorrect, misleading, or wrong. Mr. Dunkel’s testimony on the number of years of aged data that he used in his actuarial analyses is inconsistent and confusing. He claimed to have experience band data from 1994 through 2013 that could have been used but apparently was not; that were integrated in his mind, but he could not explain how; and that the most recent experience was the most important anyway. Exhibit 10 to Mr. Dunkel’s Direct Testimony, which shows his proposed survivor curves for 33 accounts, is inconsistent with the text of his prefiled testimony and his testimony at the hearing. Exhibit 10 shows that for 11 accounts, Mr. Dunkel used experience bands37 other than 2003-2013, and five of these 11 experience bands start earlier than 1994, which is the earliest of the aged data that Mr. Dunkel said was available. The beginning dates for these five experience bands are 1946, 1982, 1993, 1957, and 1958.38 When then asked whether the 11 years of aged data was sufficient to perform an actuarial analysis, Mr. Dunkel said that it was sufficient when combined with the placement data going back to 1946. He said, “Eleven is fine for a mass account.” Tr. (4-25-16) 2729. But Mr. Watson countered that short experience bands only work for short-lived assets where you can capture more of the experience of the life cycle. Tr. (4-13-16) at 764-765. Longer-lived assets will not experience enough of their life-cycle in a short experience band data set. In his prefiled Direct Testimony, Mr. Dunkel said that he used a 2003-2013 experience band. Dunkel Direct 22 (“We had the aged data needed to observe what happened to all pole Exhibit 10 refers to experience bands as “activity years.” Account 352: 2005-2013; Account 354: 2004-2013; Account 357: 1983-2013; Account 359: 19462013; Account 369.1: 2004-2013; Account 370: 2013-2013; Account 371.1: 1982-2013; Account 390.1: 1993-2013; Account 390.2: 1957-2013; Account 392.4: 1997-2013; and Account 397.0: 1958-2013. 37 38 Recommended Decision Case No. 15-00261-UT 163 investments during the most recent 11 years (Experience Band 2003-2013).” He emphasized that recent aged data is most relevant and that he did not use older aged data: Experience data prior to 2003 was available and could have been included if it was considered relevant to the future. For example, the Actuarial analysis in the PNM 2009 study included experience data for the years 19942009 and placement data for 1946-2009 for Account 364. PNM provided the aged data file from the 2009 study in response to ABCWUA 2-5. Dunkel Direct 22 n.52 (emphasis added). It is puzzling that Mr. Dunkel criticized Mr. Watson for not using available data but then ignored a significant amount of data himself, with no explanation. At the hearing, Mr. Dunkel said that, in addition to the 2003-2013 aged data, there was a second band of aged data from the 2009 Depreciation Study, starting in 1994. Tr. (4-25-16) 2722-26. When then asked how his graphs integrate the two experience bands, Mr. Dunkel said: I didn’t integrate them in the graph. I show one graph that is from the newer set of data. I have another graph from the older set of data. I integrated them in my mind. . . Tr. (4-25-16) 2726. When then asked where in his written testimony he described the process he used to integrate the two experience bands, Mr. Dunkel said, “I relied primarily on the most recent data” because that data is most relevant. Tr. (4-25-16) 2727. Mr. Dunkel’s testimony was confusing as to what data he actually used, and whether he used all of it or not. Moreover, in the end he abandoned the actuarial study he was trying to defend to substitute his own service life estimates, which in some cases were hundreds of years shorter than the indicated service lives, with no support for why he deviated from the actuarial model in account after account after account. 6. NET SALVAGE FOR ACCOUNT 373 (STREETLIGHTING) Net salvage is the difference between the gross salvage (what the asset was sold for) and the removal cost (cost to remove and dispose of the asset). Watson Direct 26. Salvage and cost of removal are built into depreciation rates by a net salvage factor which is calculated by Recommended Decision Case No. 15-00261-UT 164 dividing actual salvage value and actual cost of removal by the original cost of the retired asset. Therefore, the net salvage factor is a ratio of the original cost of the asset retired. Accounting for Pub. Utils., § 6.07. Negative net salvage occurs when the cost of removal exceeds the salvage value for property retired. The current net salvage factor for Account 373 is 60% for PNM North and 20% for PNM South. Mr. Watson recommends changing the net salvage factor to -10% or negative 10%. Exh. DAW-2, p.91, to Watson Direct. Mr. Dunkel recommends changing the net salvage factor to positive 35% to reflect including third party reimbursements to PNM as gross salvage. Mr. Dunkel argues that most of the $161,137 that Mr. Watson excluded from gross salvage is third-party reimbursements. Dunkel Direct 36-38. Third party reimbursements include payments received by PNM for damages for loss to property, such as insurance proceeds. Mr. Dunkel argues that all third-party reimbursements should be added to any gross salvage value, thereby increasing the accumulated depreciation account balance, decreasing the value of the asset in rate base and decreasing depreciation expense. PNM argues that the third-party reimbursements for Account 373 should be credited as offsets to new plant in service as contributions in aid of construction, which also has the effect of reducing rate base. PNM’s Initial Posthearing Brief 176. Deciding this issue depends on whether the third-party reimbursements for streetlights are considered payments to PNM for (1) the retirement-related cost to remove and dispose of the asset being retired; or (2) the construction-related cost to install a new asset. PNM’s Initial Posthearing Brief 174. Third-party payments are calculated based on the cost of the new assets. Watson Rebuttal 121. This supports concluding that the third-party payments for damages to streetlight Recommended Decision Case No. 15-00261-UT 165 poles should be applied against the cost of replacing and installing the new asset, not the cost of retiring the old asset. This conclusion is consistent with two other state commission decisions that rejected including reimbursements in gross salvage and instead treated them as CIAC and offsets to plant in service. Central Ill. Light Co., 230 P.U.R.4th 17, 2003 WL 23145348, *10 (Ill. Comm. Comm’n 10-17-03); Delmarva Power & Light Co., 1986 WL 1299798 (under “Net Salvage”) (Md. P.S.C. 10-23-86). In the latter case, the Maryland Public Service Commission said: Prior to 1983, DP & L included these proceeds in gross salvage and in that year, DP & L changed its accounting procedures and began to apply the proceeds to the cost of new plant additions. I find nothing unreasonable and will accept DP & L’s new procedure regarding the accounting of reimbursements. To include such in salvage distorts normal salvage occurring at end of the item’s useful life due to wear and tear or technical obsolescence. Furthermore, the revised accounting has the benefit of reducing net plant and resultant rate of return requirement. Mr. Dunkel’s recommendation to include third party reimbursements for assets in Account 373 as gross salvage is rejected. This conclusion is not precedent that third party reimbursements should always be treated as CIAC, as PNM seems to acknowledge. See PNM’s Initial Brief in Chief 174-77 (recognizing that reimbursements may be for retirement-related costs or for construction-related costs). B. AMORTIZATION PERIOD FOR RESERVE DIFFERENCES ON AMORTIZED ACCOUNTS As part of the conversion of certain general plant accounts from “depreciable” to “amortizable” accounts, PNM proposes to amortize reserve differences totaling ($21.1) million over 3.3 years, in the amount of ($6.3) million annually. The depreciation reserve account is a contra asset account which reduces the net book value of an asset by the amount of depreciation expense. It is also referred to as “accumulated depreciation.” Public Util. Depreciation Rates 46. A reserve deficit is the difference between the book reserve and the reserve that should have accumulated under prospective depreciation rates. Id. 188-89. Recommended Decision Case No. 15-00261-UT 166 ABCWUA does not oppose converting the accounts from depreciable to amortizable accounts, but opposes PNM recovering the annual amortization amount after 3.3 years, after PNM has recovered the total ($21.1 million) deficit. ABCWUA recommends that recovery of the amortization expense end automatically when PNM has recovered the ($21.1 million) deficit. Watson Direct 29-32. ABCWUA’s recommendation should be rejected. Specific costs for which recovery is approved through amortization, such as rate case expenses, are not tracked and amortization is not stopped once the full cost is recovered. Additionally, the risk of over-recovery of the challenged amortization is minimized given PNM’s repeated statements that it plans to file a new rate case in time for new rates to take effect by 2018. PNM’s Initial Posthearing Brief 179. C. RATE CASE EXPENSES PNM initially requested $4,034,109 for recovery of rate case expenses, to be amortized over two years. In Rebuttal Testimony, PNM updated and decreased this requested amount to $3,790,023, based on actuals through December 31, 2015 and revised estimates. Monroy Rebuttal 46 & Exh. HEM-1 to Monroy Rebuttal. TKH WDEOH EHORZ VKRZV 310¶V UHTXHVWHG UDWH FDVH H[SHQVHV OLVWHG E\ RXWVLGH FRQVXOWDQW Recommended Decision Case No. 15-00261-UT 167 6FRWW 0DGGHQ ,QF DVVLVWHG 310 ZLWK FRRUGLQDWLRQ DQG GRFXPHQWDWLRQ RI FDSLWDO EXGJHW GRFXPHQWDWLRQ DQG UHYLHZHG DQG YDOLGDWHG IXQFWLRQDOLW\ RI 310¶V FRVW RI VHUYLFH PRGHO 310 ZLWQHVV +HYHUW LV 0DQDJLQJ 3DUWQHU RI 6XVVH[ (FRQRPLF $GYLVRUV DQG VXSSRUWHG 310¶V UHTXHVWHG UHWXUQ RQ HTXLW\ Willis Towers Watson (Towers) is PNM’s outside consultant for evaluating PNM’s pension and benefits programs and related accounting. Towers perf0rmed the actuarial studies on pensions and benefits to support PNM’s Test Period expenses and assisted PNM in responding to discovery. Additionally, Gene Wickes of Tower provided Rebuttal Testimony in response to issues raised by Intervenors on PNM’s prepaid pension asset. Alliance Consulting developed the depreciation study that supports PNM’s new proposed depreciation rates, and Dane Watson was PNM’s depreciation witness. Price Waterhouse Coopers, LLP developed PNM’s lead-lag study. KPMG, LLP performed the independent review required by PRC Rule 530 Schedule Q-6. The Brattle Group developed PNM’s sales and customer forecasts. Amerisco, Inc. assisted in developing PNM’s forecasts for base load, weather impacts on load, and other factors affecting load. Christensen & Associates provided testimony supporting PNM’s Revenue Balancing Account proposal. Monroy Rebuttal 37-41. “[R]ate case expenses are one aspect of a utility’s operating costs and are recoverable in a general rate proceeding.” PNM Gas Servs. v. New Mexico Pub. Util. Comm’n, 2000-NMSC-012, ¶ 68, 129 N.M. 1. However, a utility must meet a “heightened burden” of demonstrating the reasonableness of its requested rate case expenses. 2000-NMSC-012, ¶ 73. By enacting §62-313(B), the New Mexico Legislature effected a change in policy for recovery of rate case expenses Recommended Decision Case No. 15-00261-UT 168 from the Commission’s general rule that unless an expense is challenged it is presumed to have been reasonably incurred. 2000-NMSC-012, ¶ 73. The AG argues that PNM has not shown that its rate case expenses are reasonable and rec0mmends that PNM be allowed to recover only 50% of its requested rate case expenses. The AG also recommends that the PRC exclude the unamortized balance of rate case expenses from PNM’s rate base because PNM should not be permitted to “turn rate case costs into a profit.” Crane Direct 43. The AG’s recommendation’s to reduce PNM’s requested rate case expenses by 50% should be rejected because it is not cost-based. See Attorney General v. New Mexico Pub. Regulation Comm’n, 2011-NMSC-034, ¶¶ 12, 18 (rates under PUA must be cost-based). The AG’s recommendation to exclude the unamortized balance of rate case expenses from PNM’s rate base should be denied because the PRC has allowed such recovery, Case No. 06-00210-UT, Final Order 34-35, ¶ 101 (6-29-07), and PNM was provided no notice that such recovery might be denied because of a change in policy. Staff recommends that the PRC disallow rate case expenses for PNM’s use of outside experts and consultants, namely for Sussex Economic Advisors, LLC, Christensen & Associates, Brattle Group and Scott Madden. Staff argues that PNM should have trained in-house professionals that can perform these services. Staff also recommends that the unamortized balance of rate case expenses be amortized over three years instead of two years. De Cesare Direct 14-17. Staff’s recommendation to disallow rate case expenses for PNM’s use of outside experts and consultants should be rejected. The PRC has consistently allowed utilities to recover through rate case expense the cost of using outside experts and consultants, and the New Mexico Supreme Court has allowed recovery of consulting fees and outside legal representation. Petition of PNM Gas Servs., 2000-NMSC-012, ¶ 76, 129 N.M. 1 (utility “reasonably and Recommended Decision Case No. 15-00261-UT 169 prudently incurred substantial costs for consulting fees”). PNM was provided no notice that it might be denied recovery of the cost of using such outside experts and consultants merely because they are outside experts and consultants. Additionally, Staff did not make an offsetting adjustment to PNM”s cost of service to reflect the cost of added staff and training to perform the services in-house. Monroy Rebuttal 35. Staff’s recommendation to change the amortization period from two to three years should also be rejected in light of PNM’s repeated statements that it will file a new rate case perhaps as soon as December 2016 for new rates to take effect in 2018. Monroy Rebuttal 45. No party or Staff has argued that PNM’s requested rate case costs exceed its actual and estimated costs. The magnitude of this case is sweeping. In total, there were nearly four weeks of hearing. PNM’s 371-page Initial Brief in Chief is not excessive. PNM’s requested rate case expenses of $3,790,023 should be approved, to be amortized over two years. D. NUCLEAR DECOMMISSIONING TRUST FUND CONTRIBUTIONS Part of the expense associated with a nuclear power plant is the cost of safely dismantling, decontaminating and disposing of the unit at the end of its service life. PNM currently funds $2.6 million annually for PV Units 1 and 2 based on an IRS dictated method. The contributions and earnings are placed into funds for investment. PNM established a Master Trust for collectively investing the assets of the tax-qualified and tax nonqualified nuclear decommissioning reserve funds for PV. The Master Trust is divided into six funds. Eden Direct 30-32. PNM’s Test Period expense for decommissioning PV Units 1 and 2 is $2,591,419. Exh. JRD, p.6, to Dittmer Direct. This expense is based on a study for the IRS to determine the maximum allowable tax-deductible contribution. Dittmer Direct 23. Recommended Decision Case No. 15-00261-UT 170 The most recent PV decommissioning study was performed in 2011 and used a 2007 base year. It was updated in 2013 to estimate “today’s” decommissioning costs. Dittmer Direct 24. Based on the 2013 cost update, PNM’s estimated share of decommissioning costs for PNM’s PV1 capacity is $87.6 million and $85.2 million for PNM’s PV2 capacity. As of June 30, 2015, the balance of PV1’s Nuclear Decommissioning Trust was $78.7 million, making it 89.8% funded on a pre-tax basis. As of June 30, 2015, the balance of PV2’s Nuclear Decommissioning Trust was $85.2 million, making it 100% funded on a pre-tax basis. Eden Direct 32; Dittmer Direct 25. Since the 2011 study, the NRC extended the PV1 operating license from 2024 to 2044 and the PV2 operating license from 2025 to 2045. Dittmer Direct 24. The 2011 study assumed an 8.41% earnings rate for funds in the Unit 1 Decommissioning Trust and an 8.35% earnings rate for funds in the Unit 2 Decommissioning Fund. It assumed a 5% cost escalation rate. In recent years, actual three and five-year pre-tax returns through December 31, 2015, are 8.1% and 8.6%, and actual cost escalation was less than 5%. Dittmer Direct 24, 27 n.6; Eden Rebuttal 17. In response to an ABCWUA discovery request to update the PV cost study to reflect current fund balances, the 20-year operating license extensions, and an appropriate return rate, PNM said that it could not perform the update. ABCWUA witness Dittmer then did the requested update himself and concluded that the Unit 1 Trust needs to earn an annual average 5.4% rate to have adequate funds at retirement without more contributions. The Unit 2 Trust needs to earn an annual average 5% rate (the cost escalation rate) to have adequate funds at retirement without more contributions. Dittmer Direct 27 & n.27. PNM did not challenge these numbers. See Eden Rebuttal 13-20. Recommended Decision Case No. 15-00261-UT 171 Based on this evidence, ABCWUA recommends that PNM cease collecting from ratepayers any decommissioning costs, describing the current funding status as “very good.” Dittmer Direct 25. PNM’s argues that its recovery of decommissioning expenses should continue because of the risk of a market downturn and changes to decommissioning cost estimates. Eden Rebuttal 14, 16. ABCWUA’s recommendation that PNM cease contributing to the decommissioning funds should be followed. Requiring PNM to stop funding the PV1 and PV2 decommissioning funds is not reckless. In fact, the PRC recently approved EPE’s decision to withdraw its request to recover its PV decommissioning costs in response to intervenor testimony. Case No. 1500127-UT, Recommended Decision 154 (2-16-16), adopted in relevant part by Final Order Partially Adopting Recommended Decision (6-8-16); Dittmer Direct 29. If in the future it appears that fund balances may be insufficient to fund decommissioning costs, rate recovery for decommissioning expenses can be reinstated. Dittmer Direct 28. In fact, given that PNM may file its next rate case as soon as December 2016, this issue can be reviewed in the near future. E. FOUR CORNERS COAL CONTRACT PNM includes in its cost of service a Test Period expense of $19.5 million to recover costs of its new Coal Supply Agreement (CSA) for the Four Corners Plant. PNM has not annualized this expense and it would increase by $3.6 million if the expense were annualized. Tr. (4-14-16) 935 (Taylor). The Four Corners Plant is operated by Arizona Public Service (APS). The CSA was negotiated between APS and the Navajo Nation’s coal company and was approved by PNM and the other Plant owners. PNM approved the contract because it found that it was cost effective and the best option available. Tr. (4-14-16) 931 (Taylor). Recommended Decision Case No. 15-00261-UT 172 NEE argues that PNM’s coal supply costs for the Four Corners Plant should be eliminated from PNM’s cost of service because, at the time that PNM approved the CSA in late 2013, it had not performed any analysis to show whether Four Corners was its most costeffective resource. Tr. (4-28-16) 3329 (Van Winkle). CFRE agrees. CFRE’s Posthearing Response Brief 9-10. CCAE argues that “PNM has simply not provided any evidence that its decision to extend its investment in Four Corners, including entering into a new coal contract, was prudent.” CCAE’s Initial Posthearing Brief 52. While this case is the appropriate case to address, if challenged, the reasonableness of the cost of the CSA, it is not the appropriate case to address the reasonableness of PNM’s inclusion of the Four Corners Plant in its resource portfolio. The appropriate case to address inclusion of the Four Corners Plant in PNM’s resource portfolio was Case No. 13-00390-UT, in which PNM requested issuance of a certificate of public convenience and necessity (CCN) for 134 MW of San Juan Unit 4, another coal plant. In that case, NEE did in fact object to issuance of that CCN “because any further acquisition of coal is unreasonable and imprudent in the extreme especially when feasible renewable alternatives exist today that meet customer needs and are better on every measure.” Case No. 13-00390-UT, Final Order 12, ¶ 27. The PRC rejected this argument, finding in part that issuance of the CCN was “the result of two years of thorough analysis and investigation of a well-developed Record.” The PRC further said, “The recommended replacement resource have already been built, saving ratepayers millions of dollars.” Id., ¶¶ 30-31. In addition to directly challenging PNM’s inclusion of coal in its resource portfolio as a basis for exclusion of the CSA from cost of service, NEE’s Initial Posthearing Brief 29-32, 33-38, NEE also argues that PNM did not meet its burden of proof, stating that the CSA “includes a 128% increase in coal costs and binds ratepayers for 15 years.” Id. 32-33. As stated above, a challenge to the reasonableness of the cost of the CSA is properly within the scope of this rate Recommended Decision Case No. 15-00261-UT 173 case. However, NEE did not submit relevant evidence that the cost of the CSA is unreasonable. Rather, NEE’s argument that the cost is unreasonable is based on its objection to coal as a resource. The Four Corners Plant has a CCN and is part of PNM’s resource portfolio: that is a given for purposes of this case. PNM needs a coal supply to operate that Plant. NEE has provided no evidence that, given that the Four Corners Plant is currently part of PNM’s resource portfolio, the cost of the CSA is unreasonable. F. PAYROLL TAX EXPENSE PNM uses a rate of 8.79% to calculate corporate payroll taxes in the Test Year revenue requirement. This percentage is based on the effective rates of social security, Medicare and federal and state unemployment taxes, and accounts for applicable wage-based limitations on these taxes. AG Witness Crane contends that the Company should utilize the statutory payroll tax rate of 7.65% instead of 8.79%. Ms. Crane’s proposed 7.65% rate is the sum of the effective social security tax rate of 6.2% and the Medicare tax rate of 1.45%. Ms. Crane asserts that the statutory tax rate is generally applied in utility rate proceedings to determine the payroll taxes associated with salary and wage expense. Crane Direct 68. . The 7.65% rate proposed Ms. Crane disregards the estimated 0.6% of federal unemployment taxes (applied to wages up to $7,000) and 4.8% of state unemployment taxes (applied to wages up to $23,400). Monroy Rebuttal at 54. Because PNM is required to pay these taxes on labor dollars, it is wholly appropriate to include them in payroll tax expenses. Accordingly, the downward adjustment to PNM’s payroll tax expense as recommended by Ms. Crane is unreasonable and should be rejected. G. NON-QUALIFIED RETIREMENT PLAN EXPENSE PNM is seeks recovery of $1,054,262 for the cost of its Non-Qualified Retirement Plans. These plans provide supplemental retirement benefits for key executives in addition to the normal retirement programs provided by PNM. They are non-qualified plans because the Recommended Decision Case No. 15-00261-UT 174 Internal Revenue Service they are not eligible for favorable tax benefits that are available for qualified plans. Crane Direct 69. The AG argues that PNM should not be allowed to recover this expense because it represents excess benefits that should be paid for by shareholders. Crane Direct 70. The AG raised this same argument in the 2015 EPE Rate Case, and the PRC rejected the argument, finding that EPE submitted sufficient evidence that its non-qualified retirement plans were reasonable programs designed to help EPE compete for executive employees. However, the PRC disallowed one-half of EPE’s claimed expense because the evidence showed that one component of the benefit was based on the financial performance of EPE, and EPE had not quantified that amount. 2015 EPE Rate Case, Final Order 47-50, ¶¶ 101-07. PNM, like EPE, has submitted sufficient evidence to support recovery of its NonQualified Retirement Plan expenses. See Rebuttal 10-11. No Intervenor or Staff has identified any evidence showing that the benefits under PNM’s Plans are tied to PNM’s financial performance. Therefore, PNM’s requested recovery of the cost of its Non-Qualified Retirement Plan expenses should be approved. H. DUES AND FEES Staff recommends disallowing recovery of all of PNM’s Test Period amounts for dues and fees, arguing that Henry Monroy’s Direct Testimony lacked detailed support of the type of dues and fees paid by PNM. De Cesare Direct 12, 23. In response to Staff’s objection, Mr. Monroy attached Exhibit HEM-2 to his Rebuttal Testimony, which lists the dues and subscriptions recorded to the dues, fees, and fines cost type and identifies their purpose. Many of the dues are for membership in community organizations, including, for example, dues for the Rotary Club of Silver City, Santa Fe Lodgers Association, Village of Tularosa, and multiple chambers of commerce. Exh. HEM-2 to Monroy Rebuttal. In Recommended Decision Case No. 15-00261-UT 175 support of recovering these expenses, Mr. Monroy said, “PNM belongs to various organizations to facilitate maintaining strong community relations.” Monroy Rebuttal 63. Staff’s recommendation should be adopted for the reason given by Staff and because some of PNM’s claimed dues cannot be recovered from ratepayers under Rule 17.3.350.10(A), which says: Contributions or donations to any political candidate, any political party, any religious cause, or any social, recreational, or fraternal organization shall not be allowed as a cost of service. Paragraph B of Rule 17.3.350.10 allows “dues in professional or trade associations and subscriptions to publications” as a cost of service provided these dues contribute to the professional education and standing of the utility’s employees for which the dues shall be paid. Paragraph C of Rule 17.3.350.10 says that contributions, donations, dues, subscriptions and membership fees other than those included in Paragraphs A and B “will not be included in the determination of cost of service unless a utility affirmatively demonstrates that such expenditures are reasonable.” Paragraph C further says, “Maintenance of corporate good will or good corporate citizenship is an insufficient reason for inclusion of these expenditures in determining a utility’s cost of service.” In Case No. 1787, Gas Company of New Mexico sought to recover in its cost of service dues for civic activities and club and non-industry groups. The PRC said that the identically worded PRC Rule that was the predecessor to 17.3.350.10(A) was alone sufficient to preclude these expenses. The PRC further said that, in the event that the predecessor to Rule 17.3.350.10(A) did not apply, the identically worded predecessor to 17.3.350.10(C) NMAC would also preclude the expenses. Case No. 1787, Recommended Decision 60. In disallowing the costs, the PRC said: We note that [17.3.350.10 NMAC] was designed to avoid just this kind of dispute. We hope the Company will not seek to include such expenses in the future unless they are prepared to make a credible showing of why such expenses are allowable under [17.3.350.10 NMAC]. Recommended Decision Case No. 15-00261-UT 176 Recommended Decision 61 (10-11-83).39 Similarly, in Case No. 2462, the PRC, relying on the predecessor to Rule 17.3.350.10(C), excluded from the cost of service, dues paid by Hobbs Gas Company to Industrial Development of Lea County, the Chamber of Commerce and the Fee and Public Land Association, even though the PRC had previously allowed recovery of these dues. Final Order 10 (4-30-93). PNM’s Test Period expenses for dues and fees, in the total amount of $1,165,357, should be disallowed. Even if some of these expenses fall into the category of “dues in professional or trade associations and subscriptions to publications,” PNM has not identified which expenses fall into that category. I. ADVERTISING EXPENSE PNM witness Larsen sponsored an exhibit showing PNM bill inserts for which it seeks cost recovery. One of the inserts is an advertisement for “Cool Comfort Loans,” which promotes low interest loans for customers switching to refrigerated air conditioning. The insert says, “It’s easier than ever to enjoy a cool and comfortable summer by converting your swamp cooler to refrigerated air conditioning.” It then lists benefits of refrigerated air conditioning. Exh. RL-3, p.8, to Larsen Direct. Mr. Larsen said that PNM included the bill stuffer in bills “at least a couple times.” The cost of this bill stuffer was $9,800. Tr. (4-20-16) 1895-97 (Larsen). Commission Rule 17.3.350.9(C)(1) prohibits recovery by a utility of advertising expenses that “[p]romote increases in the usage of energy or utility services.” Refrigerated air conditioning uses more electricity than swamp coolers. Tr. (4-20-16) 1897 (Larsen). The Final Order in Case No. 1787 says, under the heading “Specific Activities, Club and Non-Industry Dues and American Gas Association Expenses,” that [t]he Commission agrees with the Hearing Examiner’s analysis of this issue and adopts his recommendation.” Final Order 17 (10-26-83). 39 Recommended Decision Case No. 15-00261-UT 177 WRA recommends excluding $19,600 in advertising expense for the cost to PNM of including the “Cool Comfort Loans” insert in two mailings because the advertisement encourages additional electricity use. WRA’s Initial Posthearing Brief 47. WRA’s recommendation should be adopted. J. EMPLOYEE MEALS AND EMPLOYEE EXPENSES Staff’s arguments for excluding from the cost of service all amounts for employee meals and employee expenses, De Cesare Direct 9-11, 17, 21, lack support and should be rejected. The AG recommends disallowing 50% of meals and entertainment expense because “the IRS typically limits recovery of meals and entertainment expenses to 50% on the basis that a portion of these expenditures are not appropriate deductions for federal tax purposes.” Crane Direct 73. The PRC rejected this same argument made by the AG in Case No. 07-00319-UT, Recommended Decision 151-52 (7-31-08), adopted in relevant part by Final Order (10-14-08), and it should be rejected here. K. PALO VERDE INCENTIVE COMPENSATION ABCWUA proposes to eliminate from Base Year recorded operating results, PNM’s share of incentive compensation expense for PV Units 1 and 2 (10.3%) that was accrued in 2014 as a result of Arizona Public Service Company — the operator of PV — achieving a pre-established level of financial earnings. Dittmer Direct 50-51; Tr. (4-18-16) 1583 (Olson). ABCWUA calculates the amount of this exclusion to be $581,546. Exh. JRD-3, p.17 to Dittmer Direct. The PRC recently denied El Paso Electric Company’s request to recover this same expense in its cost of service based on the PRC’s general policy of excluding financially-driven incentive compensation. The PRC said that an evidentiary showing of benefit to ratepayers may allow for an exception to this general policy, but no such showing had been made by EPE. Case No. 15-00127-UT, Final Order Partially Adopting Recommended Decision 47, ¶ 100 (6-8-16). Recommended Decision Case No. 15-00261-UT 178 PNM has not shown that its ratepayers benefit from payment of the incentive compensation, see Olson Rebuttal 43, and the PV incentive compensation expense should be excluded from PNM’s cost of service. L. PNM EMPLOYEE INCENTIVE COMPENSATION PROGRAMS PNM seeks rate recovery for the incentive components of two compensation plans in this case. First, PNM seeks approval to recover the costs associated with its Wholesale Power Marketing Plan (“WPM Plan”). PNM’s proposed Test Period expense for the WPM Plan is $767,751. Second, PNM is seeking approval to recover the costs associated with its Business Unit Group Incentive Program for employees in pay grades below director and officer levels. The Group Incentive Program replaced an existing compensation program, beginning in 2015. PNM’s proposed Test Period expense for the Group Incentive Program is $3,943,469. Vavruska-Marcum Direct 21-23. The WPM Department is responsible for off-system sales and the electric power realtime and short-term merchant function that trades in the wholesale market. As of March 31, 2015, all 26 employees in the WPM department were eligible, including the director, managers, traders, pre-schedulers and administrators, under the WPM Plan. The WPM Plan is designed to motivate and reward employees for achieving a variety of reliability and financial performance targets, and the targets must be met for any payout to employees. The WPM Plan reliability performance targets include system reliability, such as 100% compliance with Reliability Based Control, 98% compliance with the Southwest Reserve Sharing Group and 100% compliance with the Disturbance Recovery Standard Recovery. Jurisdictional financial targets include controlling costs and maximizing off-system sales margins for the benefit of customers. The WPM Plan is capped at a specified maximum amount for the award pool. The level of the award pool is dependent upon the achievement of the specified performance targets per the WPM Plan. Individual employee payouts are based on individual employee performance during the WPM Recommended Decision Case No. 15-00261-UT 179 Plan year, the maximum cap per the WPM Plan, and the level of overall award pool funding. Awards are only paid to employees if earned by achieving the performance targets identified in the WPM Plan. Vavruska-Marcum Direct 18-19. The Group Incentive Program is designed to motivate and reward eligible non-union employees for achieving business unit performance metrics that align with the corporate and business unit objectives and strategy. As of March 15, 2015, the Group Incentive Program was applicable to 862 in the pay grades of G14 to G05. The Group Incentive Program is intended to primarily reward eligible employees for achieving business unit performance metrics as a team, in addition to meeting individual performance goals. This program provides benefits to New Mexico customers by encouraging employees to meet business goals that are primarily related to the provision of safe, reliable and cost-effective service to customers. It does not include an earnings per share performance metric. Vavruska-Marcum Direct 22. The Commission has previously authorized rate recovery for the costs of employee incentives not tied to Company earnings. In PNM’s 2007 Rate Case, the Commission ruled that an employee incentive compensation program, in the form of a stock options, that rewarded safety, reliability and customer service, and that was used to retain employees resulted in a benefit customers, was appropriate to include in revenue requirements. Case No. 07-00077-UT, Final Order Partially Adopting Recommended Decision 31-32, ¶¶ 84-85. In PNM’s 2010 Rate Case, the Commission approved a Certification of Stipulation that allowed PNM to recover the costs associated with the WPM Plan as well as the former Merit Plus Program that, like the Group Incentive Program, have provided employee performance-based compensation and neither of these programs tied incentives to any company earnings. Case No. 10-00086-UT, Certification Stipulation 84-86. Recommended Decision Case No. 15-00261-UT 180 The AG opposes any recovery of costs associated with the Group Incentive Program40 and proposes only partial recovery of the costs associated with the WPM Plan. Crane Direct 6769. ABCWUA recommends that PNM be allowed to only recover a portion of the costs of Group Incentive Program. Dittmer Direct 33. AG witness Andrea Crane gives two reasons for her rejection of the costs associated with the Group Incentive Program. First, Ms. Crane contends that PNM failed to support its claim for the Groups Incentive Program. Second, Ms. Crane contends that the costs being sought under the Group Incentive Program are considerably higher than the costs incurred under the previous Spot Bonus Program. Crane Direct at 67. With regard to the WPM incentive program, Ms. Crane recommends that only 71.75% of the requested amount for the program be allowed to be recovered by PNM in rates. Ms. Crane contends that this adjustment is justified based on the historical payouts to employees under this program. Crane Direct 67-68. ABCWUA witness Dittmer recommends that recovery for the costs of the Group Incentive Program be limited to $1,150,000 to be “in line with payouts that were made in all recent prior years.” Dittmer Direct 34. The table below shows actual payouts under the Group Incentive Program for the years 2011-2014 and projected payments for 2015 and 2016: Table 3 Historic Actual and Test Year Projected STIP Payouts Plan Year STIP Payout Actual $1,120,846 2011 $1,191,372 2012 $1,191,282 2013 $1,191,282 2014 $1,192,577 Projected 2015 $1,150,000 2016 $4,874,625 40 ABCWUA Dittmer refers to the Group Incentive Plan as the Short Term Incentive Compensation Plan or STIP. Dittmer Direct 31-34. To avoid confusion, this Recommended Decision uses only the term “Group Incentive Plan.” Recommended Decision Case No. 15-00261-UT 181 FTY Proposed $3,943,469 Dittmer Direct 32. PNM’s Test Period amount is about 308% higher than the actual annual amounts in 2011 through 2014 and the projected 2015 amount. PNM says that it needs to increase payouts under the Group Incentive Program to better attract, motivate, and retain this group of employees. PNM relies on three points to support the need for increased payments: 1. Employee surveys that showed a 78% level of dissatisfaction with the prior Spot Bonus Program 2. A Towers Watson benchmarking study showing that the 2015 level of funding was below market 3. Increased employee attrition. In 2014, thirty-seven percent of voluntary terminations came from the employee job categories now covered under the Group Incentive Program. In 2015, forty-one percent of voluntary terminations came from these eligible employee job categories. Monfiletto Rebuttal 5. PNM says that the Test Period expense level is justified because there is a cost to customers as a result of increased employee attrition. On average, it takes four to five months for a new employee in a professional job position to reach full productivity, according to Bersin by Deloitte Strategic Onboarding Study. That learning curve translates into an average cost of $18,096 per employee, based on an average annual salary of $72,374. With 78 terminations in 2015, the cost of attrition is approximately $1.4 million annually. Monfiletto Rebuttal 5. In establishing the 2016 Group Incentive Program budget, PNM used an internal analysis of existing incentive compensation plans and the Towers Watson benchmarking data to determine reasonable incentive award levels. PNM considered the entire incentive compensation structure for the company so that award levels were aligned by the grade and Recommended Decision Case No. 15-00261-UT 182 salary of these job categories and fit with other incentive plan structures to avoid any compensation compression issues. Although the incentive compensation levels remain below market, PNM was mindful of costs and therefore established funding at an amount between the 25th and 50th percentile of market for these job categories. Monfiletto Rebuttal 6. The AG’s argument in favor of reducing payouts under the WPM incentive program lacks merit and should be rejected. ABCWUA and the AG’s objection to the much-higher than historical level of expense for the Group Incentive Program has some merit — an over 300% increase in this expense is dramatic. On the other hand, PNM’s arguments in favor of increasing the expense level also have merit. A reasonable resolution is to allow PNM to recover roughly one-half of its proposed increase from levels in previous years, or $3,000. Actual payments in 2016 can be examined in PNM’s rate case and may justify a further increase in this expense level. In the meantime, if PNM wants make payouts under the Group Incentive Program higher than $3,000, it is not unreasonable to require shareholders to pay for that share of the expense. M. ROW EXPENSES The following table shows PNM’s Base and Test Period expenses for rights of way for distribution and transmission lines, updated as of April 8, 201641: Rights of Way Distribution Lines Transmission Lines Base Period $119,366 $9,593,661 Test Period $191,995 $10,589,128 In his Rebuttal Testimony filed on February 22, 2016, PNM witness Johnson said that there were nine rights of way over Native American Lands that had expired or would expire during the Test Period for which PNM would need to obtain renewals. PNM forecasted the renewal costs for these rights of way and included these costs in its ECCOS. Johnson Rebuttal 2. PNM relied in part on “bona fide market data about the likely costs that will be incurred with 41 PNM Exh. 38; Tr. (4-15-16) 1310-11 (Monroy). Recommended Decision Case No. 15-00261-UT 183 respect to these right-of-way renewals.” Id. at 5. Mr. Johnson’s Rebuttal Testimony included estimated dates for reaching final agreements. Id. at 5-6. In response to a Bench Request, on April 8, 2016, Mr. Johnson filed updated information on the rights of way. The estimated dates for reaching final agreements on some of the rights of way were extended in Mr. Johnson’s April 8, 2016 testimony. ABCWUA witness Dittmer recommended a $1 million decrease in PNM’s right of way expenses because PNM provided little or no support for its estimates. The $1 million recommended decrease is not cost-based, but to highlight the need for actual information. Dittmer Direct 51-52; Tr. (4-25-16) 2796-97 (Dittmer). Mr. Johnson’s reliance on PNM’s experience and unidentified “bona fide market data about the likely costs that will be incurred with respect to these right-of-way renewals” is insufficient to support PNM’s estimated costs for those rights of way for which final agreements had not been reached by the last day of the April hearings: April 29, 2016. Cf. Case No. 2440, Recommended Decision 49 (11-9-92), adopted in relevant part by Final Order (12-15-92). For those rights of way, PNM should be allowed to recover the Base Period expense if PNM incurred an expense for the right of way in the Base Period; it should be allowed to recover no expense if it did not incur an expense for the right of way in the Base Period. ABCWUA’s recommendation should be rejected because it is not sufficiently cost-based. See Attorney General v. New Mexico Pub. Regulation Comm’n, 2011-NMSC-034, ¶¶ 12, 18 (rates under PUA must be cost-based). XXI. PNM’S EMBEDDED COST OF SERVICE STUDY: FUNCTIONALIZATION; CLASSIFICATION; ALLOCATION A. EMBEDDED COST OF SERVICE STUDY As part of the Amended Stipulation approved in the 2010 PNM Rate Case, PNM agreed to “file a rate design and class cost of service based on embedded cost principles in its next general rate case.” Amended Stipulation, ¶ 34. Recommended Decision Case No. 15-00261-UT 184 Embedded and marginal cost of service studies are alternative methods of establishing both the revenue requirement allocated to each class and the components of the rate design for each class. –‹Ž‹–› ƒ•‡ ‘Ǥ ʹʹ͸ʹǡ ‡…‘‡†‡† ‡…‹•‹‘ ͳͲͷǤ Embedded cost of service studies calculate embedded customer costs while marginal cost of service studies calculate marginal costs. An embedded cost of service study uses booked or historical costs of a utility and assigns cost to the various classes of customers according to the cost each customer class places on the utility. A marginal cost of service study measures an incremental customer’s impact on the system and determines the associated costs. Development of PNM’s ECCOSS occurred in three major steps: (i) functionalization; (ii) classification; and (iii) allocation or assignment. Aguirre Direct 6. Functionalization is the process of categorizing costs by operating function, such as production, transmission, distribution, and customer service. Classification is the process of defining functionalized costs as demand-related, energy-related, or customer-related. Allocation is the process of apportioning costs among customer classes. Chan Rebuttal 13-15. PNM’s functionalization of costs is not challenged and therefore is not discussed. PNM’s classifications and allocations are challenged and discussed below. B. CLASSIFICATION In the classification process, costs for each functional category are divided into classifications based on the components of utility service being provided. The three principal cost classifications for an electric utility are: x Demand-related (costs that vary with the kW demand imposed by a customer) x Energy-related (costs that vary with the amount of energy used by a customer) x Customer-related (costs directly related to the number of customers served) Chan Direct 23; Phillips Direct 7-8. Production or generation costs are classified as either demand-related or energy-related. Whether a generation cost is classified as demand-related or energy-related generally depends Recommended Decision Case No. 15-00261-UT 185 on whether the cost is fixed or variable. Fixed generation costs include cost of capital, depreciation, taxes, and fixed O&M. Variable generation costs are fuel costs, purchased power costs, and some O&M expenses. Fixed generation costs vary with capacity additions, not with energy produced from plant capacity, and are classified as demand-related. Variable generation costs change with the output of energy produced, delivered or purchased and are classified as energy-related. Cost Allocation Manual 35. Transmission costs are classified as either demand-related or energy-related. However, most transmission costs are fixed and are therefore classified as demand-related. Id. at 21. Distribution costs are classified as demand-related, customer-related, and energyrelated. However, they are generally classified as demand-related and customer-related because they are affected primarily by demand and the number of customers. Id. at 21. However, the above cost classifications are only “typical.” “[T]here is no general agreement as to what items or portions of total costs should be included among the demandrelated costs, perhaps because cost functions are far too complex to be reflected by the arbitrary, three-way classification of customer, energy, and demand.” Bonbright 494. NMIEC recommends that PNM classify the following costs as fixed rather than variable, and classify them as demand-related costs: x x x x 1. Generation O&M expenses other than fuel and purchased power Fuel transportation Demand-related purchased power agreements Coal mine decommissioning costs. GENERATION O&M EXPENSES OTHER THAN FUEL AND PURCHASED POWER NMIEC challenges PNM’s classification of certain generation O&M expenses, other than fuel and purchased power, as energy-related. NMIEC argues that these expenses, which it says are not labor-related, are more fixed than variable in nature and should be classified as demandrelated. Mr. Phillips said that “the vast majority of these costs do not vary in any appreciable Recommended Decision Case No. 15-00261-UT 186 way with the number of kilowatt hours generated, but occur primarily as a function of the existence of the plants, the hours of operation and the passage of time.” Phillips Direct 21. Because PNM classified the challenged expenses as energy-related, they were allocated using the average demand allocation factor, which approximates the energy allocator, as it reflects total energy divided by 8760 (the number of hours in a year). If the challenged expenses were classified as demand-related, they would be allocated using a net production demand allocation factor, which is equivalent to the 3S1WCP42 allocator. Chan Rebuttal 29. Ms. Chan responded that one of the accounts for which Mr. Phillips proposes to change the allocator is Account 510, which the FERC System of Accounts defines as including the cost of labor incurred in the general supervision and direction of maintenance of steam generation facilities. Chan Rebuttal 29-30. In her response, Ms. Chan justified PNM’s classification of the challenged costs — maintenance for base load plant — as energy-related because this maintenance is directly related to the daily, reliable supply of energy that PNM provides its customers. Id. at 30. 2. FIXED FUEL TRANSPORTATION AND DEMAND PPA COSTS PNM classifies costs associated with fixed gas transportation and demand-related costs associated with the Valencia Purchased Power Agreement (PPA) as energy-related. NMIEC argues that they should be classified as demand-related because the gas transportation costs are fixed by contract and do not fluctuate significantly over time, and the Valencia demand payments are based on contract rates and capacity. Phillips Direct 22. In her response, Ms. Chan justified PNM’s classification of these costs as energy-related. While PNM incurs the costs under fixed contracts, the broader purpose of the contracts is to deliver fuel to PNM’s plants to provide energy on a constant daily basis, not exclusively to meet 42 See § for discussion of the 3S1WCP allocator. Recommended Decision Case No. 15-00261-UT 187 PNM’s peak demands. Chan Rebuttal 31. It is more appropriate to “look through” the PPAs and classify the underlying costs accordingly. 3. COAL MINE DECOMMISSIONING COSTS PNM classifies its annual contribution to the coal mine decommissioning fund as energyrelated. NMIEC argues that this cost should be classified as demand-related because contributions to the fund are fixed and even if no more coal is mined, PNM would still be required to contribute its ownership share to the fund. Phillips Direct 22-23. In her response, Ms. Chan said that PNM’s contributions to the decommissioning fund related to the Four Corners Plant in fact are not fixed. She agreed that PNM’s contributions to the fund related to the San Juan Plant are fixed, but she justified PNM’s classification of these costs as energy-related because the amount of coal taken from the mine depends on how much coal is needed to produce electricity. Chan Rebuttal 33. 4. DISTRIBUTION COSTS NMIEC objects to PNM’s classification of distribution costs in Accounts 364 (Poles, Towers and Fixtures), 365 (Overhead Conductors and Devices), 366 (Underground Conduit) and 367 (Underground Conductors and Devices) as entirely demand-related and recommends reclassifying 30% of these costs as customer-related to reflect that certain distribution system investments are necessary just to connect a customer to the system and are not affected by demand or energy use. Mr. Phillips argues that this recommendation is consistent with the minimum distribution system (MDS) method, which recognizes that a utility incurs a minimum cost when it extends its primary and secondary distribution systems to connect more customers. The MDS system includes every component of the distribution system required to provide service, i.e., meters, services, secondary and primary wires and substations. However, the MDS method does not include costs specifically incurred to meet peak demand. Phillips Direct 24-25. Recommended Decision Case No. 15-00261-UT 188 Mr. Phillips supported his recommendation through an example involving two customer classes — A and B — which each place the same total demand on the system: 120 kW. However, Class A serves 12 customers, each with a 10 kW load, while Class B serves only one customer with a 120 kW load. As the Figure below shows, a more extensive distribution system is required to serve the Class with 12 customers (A) than the Class with one customer (B): Figure NLP-3 &ODVVLILFDWLRQ RI 'LVWULEXWLRQ ,QYHVWPHQW 7RWDO 'HPDQG N: 7RWDO 'HPDQG N: &ODVV $ &ODVV % If 30% of the costs of these distribution systems are classified as customer-related rather than demand-related, the costs assigned to the one customer in Class B will decrease. Mr. Phillips quoted from the Cost Allocation Manual, which supports his recommendation: Distribution plant Accounts 364 through 370 involve demand and customer costs. The customer component of distribution facilities is that portion of costs which varies with the number of customers. Thus, the number of poles, conductors, transformers, services, and meters are directly related to the number of customers on the utility’s system. As shown in Table 6-1, each primary plant account can be separately classified into a demand and customer component. Phillips Direct 29 (quoting Cost Allocation Manual 90). Recommended Decision Case No. 15-00261-UT 189 The Cost Allocation Manual says that two methods are used to determine the demand and customer components of distribution facilities: the minimum-size-of-facilities method and the minimum-intercept cost of facilities. Cost Allocation Manual 90. Mr. Phillips identified multiple state commissions that have classified a portion of the costs in Accounts 364 through 367 as customer-related. His research shows that, among utilities that classify a portion of the costs in these Accounts as customer-related, they classify 30% to 50% as customer-related. Mr. Phillips recommends classifying 30% of the costs in Accounts 364 through 367 as customer-related. Phillips Direct 30-32. PNM does not have the data necessary to classify distribution facilities as demand-related or customer-related under either NARUC-identified method. Therefore, Mr. Phillips used 30% as a proxy. Tr. (4-27-16) 3163 (Phillips). Ms. Chan agreed that Mr. Phillips’ recommendation is consistent with the MDS method. However, she opposed his recommendation for two reasons. First, doing so would likely significantly increase the customer charge for some classes, particularly the residential class. PNM’s $13.14 proposed residential customer charge does not recover the cost of secondary and primary wires, pole, or substations, which would be recovered through the customer charge under the MDS method. Mr. Phillips did not calculate the residential customer charge resulting from his recommendation, but his proposed cost of service study reclassifies $123 million of rate base from demand-related to customer-related. Chan Rebuttal 35. Second, PNM does not maintain the data necessary to use the MDS method. Ms. Chan said that Mr. Phillips’ classification of 30% of costs in Accounts 364 through 367 is based on assumptions that can’t be verified. Id. at 37. NMIEC’s recommendation should be rejected for the two reasons given by Ms. Chan. However, within two months of issuance of a final order in this case, PNM should begin collecting and maintaining data necessary to determine the demand and customer components Recommended Decision Case No. 15-00261-UT 190 of distribution facilities. In its next base rate case filing, PNM should identify which of the two methods identified by NARUC it proposes to use in its following rate case filing to determine the demand and customer components of distribution facilities and shall describe the procedure it has adopted to collect and maintain the data necessary to use this method. C. ALLOCATION 1. BACKGROUND Allocation assigns costs to customer classes based on criteria that reflect each class’ responsibility for the cost. Chan Rebuttal 15. The functionalized and classified costs are allocated among the classes as follows: x Demand-related costs: allocated on the basis of demands (kW) imposed on the system during specific peak hours x Energy-related costs: allocated on the basis of energy (kWh) which the system must supply to customers x Customer-related costs: allocated on the basis of the number of customers Cost Allocation Manual 22; Phillips Direct 8. Many methods are used to allocate costs, and different cost categories require different allocation methods. The three basic allocation factors are capacity, energy and customer. No single costing method is superior to any other and the choice of method depends on the unique circumstances of each utility. Individual cost methods are complex and have inspired numerous debates on application, assumptions, and data. Further, the role of cost in ratemaking is itself not without controversy. Cost Allocation Manual 22. Bonbright frankly says that treatment of demand costs “has made a nightmare of utility cost analysis.” Bonbright 494. There are a variety of formulas of apportionment. “Most of them have no claim whatever to validity from the standpoint of cost determination, and only a dubious claim to acceptance as compromise measures of reasonable rates.” Id. at 495. The United States Supreme Court has even said, “Allocation of costs is not a matter for the slide rule. Recommended Decision Case No. 15-00261-UT 191 It involves judgement on a myriad of facts. It has no claim to an exact science.” Colorado Interstate Gas Co. v. Federal Power Comm’n, 324 U.S. 581, 589 (1945). In many cases, the method chosen for determining the cost applicable to each class of customer “has been based on a subjective perception of fairness and equity so that the resultant costs could be translated directly into rates. However, in some cases, the selected method may be dictated by a desired end result.” Hahne & Aliff, § 10.04[1] at 10-16. As Ms. Chan aptly observed, “Much of the intervenor testimony regarding allocation methodologies in this case has the result of shifting cost responsibility from one particular rate class to another rate class or several rate classes.” Chan Rebuttal 15. PNM correctly cautions the PRC to “fully consider the balance that is required, in that one change to an allocation methodology could affect other customers in significant ways.” Id. at 16. 2. GENERATION ALLOCATION METHOD PNM currently uses the Average and Excess Demand Formula to allocate generation costs among customer classes. Chan Direct 18. However, as part of the Amended Stipulation approved in the 2010 PNM Rate Case, PNM agreed “not to file an average-and-excess demand allocation in its next general rate case filing.” Amended Stipulation, ¶ 39. PNM now proposes to use a 3-Summer/1-Winter Coincident Peak (3S1WCP) demand method to allocate generation demand costs. Chan Direct 29. The 3S1WCP Method is a type of coincident peak responsibility (CP) method of allocating demand costs. The CP Method divides system capacity costs among classes of service in proportion to the peak demands imposed by classes at the time of the system peak. A summer and winter peak method, as proposed by PNM, is intended to reflect the effect of two distinct seasonal peaks on customer cost assignment. “If the summer and winter peaks are close in value, and if both significantly affect the utility’s generation expansion planning, this approach may be appropriate.” Cost Allocation Manual 45. Recommended Decision Case No. 15-00261-UT 192 Under the CP Method, capital costs are imputed to those services that are rendered at the time of system peak and in proportion to the kW demand imposed at this time, measured over a short period such as 15 minutes. Under this method, service rendered completely offpeak should theoretically be assigned no responsibility whatever for capacity costs. Bonbright 495. Under the CP Method and from a cost causation perspective, “there would be general agreement that, with certain qualifications the cost to the company of rendering any type of service which can be counted upon positively to stay off the system (or subsystem) peak does not include any capacity cost.” Id. at 498. Whether rates for such a service should nevertheless recover some capacity costs “because of a widely held view that even off-peak users and interruptible power users should make some fair contribution to the costs of a plant which confers upon them a benefit, is another question[.]” Id. at 499 (emphasis in original). Despite the infirmities of the CP Method, as well as any other method, economists in general tend to favor it over the alternatives. Id. at 496. The different subtypes of the CP Method differ in how they define the system peak. For example, the 1-CP Method defines the system peak as the highest single hour’s system load during the entire year. Each Class’ CP is that Class’ load during that hour the system peak occurs. The 4-CP Method identifies the highest single hour’s system load during each of the individual 12 months and then defines the system peak as the average of the four highest of these 12 system loads. Each Class’ CP is that Class’ average load over those four hours. The 12CP Method identifies the highest single hour’s system load during each of the individual 12 months and then defines the system peak as the average of all 12 system loads. Each Class’ CP is that Class’ average load over those 12 hours. Chan Direct 30; Michael E. Small, A FERC Electric Rate Primer, 5 Energy L.J. 107, 135 (1984). Recommended Decision Case No. 15-00261-UT 193 If a utility’s system demand is relatively flat, then that supports use of a 12-CP Method. If a utility experiences a pronounced peak during one or four consecutive months, then that supports use of a 1-CP or 4-CP Method. Id. The CP Method proposed by PNM —the 3S1WCP Method — defines the system peak as the average of the highest single hour’s system load during three summer months (June, July, and August) and one non-summer month (December). Each Class’ CP is that Class’ average load over those four hours. Chan Direct 29. PNM argues that the 3S1WCP Method is the most appropriate method to allocate generation demand costs because PNM has both a summer and a winter peak. Its winter CP demands are about 82% of its summer CP demands. Chan Direct 31. The following chart compares PNM’s summer and winter peaks: 310 )LJXUH 6& 5HEXWWDO Ϯ͕ϬϬϬ ^ƵŵŵĞƌ WĞĂŬ ϮϬϭϱ ĐƚƵĂůͬ&ŽƌĞĐĂƐƚ ϭ͕ϵϬϬ DŽŶƚŚůLJ ǀĞƌĂŐĞ ΖϬϳ ͲΖϭϳ& ϭ͕ϴϬϬ ϭ͕ϳϬϬ tŝŶƚĞƌ WĞĂŬ Dt ϭ͕ϲϬϬ ϭ͕ϱϬϬ ϭ͕ϰϬϬ ϭ͕ϯϬϬ ϭ͕ϮϬϬ ϭ͕ϭϬϬ DĂƌ Ɖƌ Recommended Decision Case No. 15-00261-UT DĂLJ :ƵŶ :Ƶů ƵŐ 194 ^ĞƉ KĐƚ EŽǀ ĞĐ :ĂŶ &Ğď PNM says its selection of the 3S1WCP Method aligns with the following description of the summer and winter peak method in the Cost Allocation Manual: The objective of the summer and winter peak method is to reflect the effect of two distinct seasonal peaks on customer cost assignment. If the summer and winter peaks are close in value, if both significantly affect the utility generation expansion planning, this approach may be appropriate. Chan Direct 32. The Manual further says that “[t]he number of summer and winter peak hours may be determined judgmentally or by applying specified criteria. One method is simply to average the class contributions to the summer peak hour demand and the winter peak hour demand.” Chan Direct 32-33. Staff supports PNM’s use of the 3S1WCP Method because it best reflects PNM”s actual peak demand characteristics. Reynolds Direct 7. NMIEC argues that PNM’s winter peak is not significant enough to be recognized in the allocator and recommends using a 3 Summer CP Method, using only the June, July, and August system peaks. NMIEC witness Phillips said that “the winter peak is statistically indistinguishable in magnitude from the peak demand occurring in the shoulder months and significantly different from the summer peaks.” He pointed out that PNM’s September peak is actually higher than its December peak and that the peak in PNM’s winter and shoulder months ranges from 60% to 85% of the summer peak. Phillips Direct 17-18. Through Stella Chan’s Rebuttal Testimony, PNM demonstrated that using a 3S1WCP Method to allocate generation demand costs is superior to a 3 Summer CP Method. She explained that if PNM used a 3 Summer CP Method, PNM’s Streetlighting and Private-Area Lighting customers would be allocated zero generation costs because PNM’s CPs tend to occur before the sun goes down in the summer. Because streetlights do not turn on until after the sun goes down, PNM’s historical data shows little or no use by the Streetlighting class during the peak periods in June, July, August, and September. Chan Rebuttal 20-22. Recommended Decision Case No. 15-00261-UT 195 If the Streetlighting Class does not contribute to peak demand, exempting it from paying demand-related costs is only justifiable if the demand-related cost being recovered is incremental capacity cost — the cost per kW of enhancing capacity rather than the average cost per kW of total capacity. “To the extent to which this embedded cost either exceeds or falls short of incremental cost, it is unallocable on any principle of cost analysis.” Bonbright observes, “Unfortunately, this fact is ignored by fully distributed cost analysis of the public utility type.” Bonbright 504. In fact, one of three “especially serious” deficiencies of a public utility cost analysis identified by Bonbright is: [T]he capacity costs or demand-related costs are usually derived from book values of plant and equipment that reflect sunk costs in dollars of original investment, not costs that can be said to vary, except in a very indirect way, with present and future increases in plant capacity. Bonbright 511; see also Charles F. Phillips, Jr., The Regulation of Pub. Utils. 461 (1993) (an often-voiced objection to the peak responsibility allocation method is that “utility plant is required for the service of both on-peak and off-peak users and that both, therefore, should make some contribution toward its capital cost.”). In Colorado Interstate Gas Co. v. Federal Power Commission, the United States Supreme Court addressed a similar situation. In that case, the Colorado Interstate Gas Company objected to the Federal Power Commission’s assignment of transmission costs as 50% demand-related and 50% energy-related. The Company argued all transmission costs should be classified as demand-related because “volumetric costs have no relation to the property required for meeting the maximum demands of the wholesale business[.]” In rejecting this argument, the Court said: It is not apparent why direct industrial sales should carry a lighter share of the costs merely because their use of the pipeline may be less on the system peak day. As the Commission points out, if the method advanced by Colorado Interstate were used, the amount paid by the industrial customer for transportation of the gas through the pipeline would be measured not by the customer’s use throughout the year, which might be substantial, but by its use on the system peak day which might Recommended Decision Case No. 15-00261-UT 196 be slight. In that event the industrial customer would obtain to an extent free transportation of gas. Colorado Interstate Gas Co. v. Federal Power Comm’n, 324 U.S. 581, 592. There is no evidence PNM’s demand-related costs being recovered are only PNM’s incremental capacity costs or only the costs of PNM’s peak load generators. “Clearly, PNM’s generation plants are utilized and costs are incurred to provide electricity to Streetlighting and private area lighting customers throughout the year.” Chan Rebuttal 20-22. Therefore, using the 3S1WCP Method to allocate generation demand costs is reasonable and should be approved. 3. TRANSMISSION ALLOCATION METHOD PNM currently uses, and proposes to continue to use, the 12CP Method to allocate transmission demand costs. PNM explains, “Given that PNM’s transmission system is used at a constant level throughout the year to ensure reliability, the 12 CP demand allocator is appropriately used for transmission costs[.]” Chan Direct 33-34. NMIEC argues that using the 12CP Method to allocate transmission demand costs is inappropriate because the transmission system is built to meet the annual system peak demand, which occurs, according to NMIEC, in the summer, and is not equal to the average of the 12 monthly peak demands. NMIEC argues that a 3 Summer CP Method should be used to allocate transmission demand costs. Phillips Direct 20. The City/County argue that PNM should use the same method to allocate transmission demand costs that it uses to allocate generation demand costs: the 3S1WCP Method.43 Dr. Ankum explained: Generation and transmission are subject to the same variations in peak demand. Except for a few limited renewable energy sources, once energy is generated it needs to be transmitted: there is no other place for it to go. Thus, if 43 A 12CP allocator allocates almost three times as many transmission-demand costs to Streetlighting as 3S1WCP. Ankum Direct 36-37. Recommended Decision Case No. 15-00261-UT 197 generation facilities experience peak demand, so do transmission facilities. Likewise, if generation facilities are operating at off-peak levels, so do transmission facilities. That is, the load on generation and transmission facilities operate in tandem and both are subject to “significant variations in peak demand” as they are both subject to seasonal variations in retail demand. Ankum Direct 35. Dr. Ankum submitted the following figure showing that the monthly coincident peaks coincide closely for generation and transmission: Figure 1: Coincident Peak Demand over 24 Months (BY and FTY) Ϯ͕ϬϬϬ͕ϬϬϬ ϭ͕ϴϬϬ͕ϬϬϬ ϭ͕ϲϬϬ͕ϬϬϬ ϭ͕ϰϬϬ͕ϬϬϬ ϭ͕ϮϬϬ͕ϬϬϬ dƌĂŶƐŵŝƐƐŝŽŶ ϭ͕ϬϬϬ͕ϬϬϬ 'ĞŶĞƌĂƚŝŽŶ ϴϬϬ͕ϬϬϬ ϲϬϬ͕ϬϬϬ ϰϬϬ͕ϬϬϬ ϮϬϬ͕ϬϬϬ Ϭ ϭ Ϯ ϯ ϰ ϱ ϲ ϳ ϴ ϵ ϭϬ ϭϭ ϭϮ ϭϯ ϭϰ ϭϱ ϭϲ ϭϳ ϭϴ ϭϵ ϮϬ Ϯϭ ϮϮ Ϯϯ Ϯϰ dŝŵĞ ŽĨ ĂLJ Ankum Direct 36. PNM responded that “building generation to serve the annual system peak does not translate one-for-one to the transmission system and vice versa.” While new plant might be added to meet new peak demands, the transmission system might already have enough capacity so that new transmission is not needed. Ms. Chan said that while PNM’s transmission system is designed to meet peak demands, it also is designed to maintain a constant level of reliability throughout the year, not just at peak. Chan Rebuttal 25-26. PNM’s attempt to distinguish the transmission system from the generation system for purposes of allocating demand costs falls flat. Ms. Chan admitted that PNM’s transmission system is designed to meet peak demands. PNM could hardly say that its generation system, Recommended Decision Case No. 15-00261-UT 198 unlike its transmission system, is not designed to also maintain a constant level of reliability throughout the year, not just at peak. PNM has shown by a preponderance of the evidence that the 3S1WCP Method should be used to allocate generation demand costs. The rationale for using the 3S1WCP Method to allocate generation demand costs extends to allocating transmission demand costs. Attachment D to this Recommended Decision compares, without banding, class revenue allocations using the 12CP method and the 3S1WCP method to allocate transmission demand costs, if allocations were based strictly on the cost causation results of PNM’s ECCOSS. The rate impact from using the 3S1WCP method is not dramatic. In any event, the rate impact on particular customer classes should not inhibit an appropriate allocation that is reflective of cost causation. Allocation can proceed consistent with cost causation and rate design is addressed separately. Tr. (4-27-16) 3164 (Phillips). The 3S1WCP Method should be used to allocate transmission demand costs. 4. FUEL COSTS In three lines of his prefiled testimony, NMIEC witness Phillips recommended allocating fuel costs to reflect summer and non-summer on- and off-peak periods consistent with PNM’s TOU rate structure. Phillips Direct 15. Nowhere in his testimony did he discuss this recommendation. This recommendation should be denied. See State v. King, 2007-NMCA-130, ¶ 17, 142 N.M. 699 (court may refuse to consider arguments unsupported by authority or analysis). XXII. RATE DESIGN Once all costs have been functionalized, classified, and allocated, the next step is to determine the appropriate level of revenue to collect from each rate class. If PNM recovered its claimed non-fuel revenue deficiency from each class based on each Class’ cost responsibility as Recommended Decision Case No. 15-00261-UT 199 shown by PNM’s ECCOSS, some of its classes would receive a revenue increase and some a revenue decrease, as shown in the tables below: Class 1A/1B Residential 2A/2B Small Power 3B/3C General Power 4B Large Power 5B Large Service 10A/10B Irrigation 11B Water/ Sewage 24.11% 7.38% 2.29% 9.52% (6.17%) 50.09% 3.31% % NonFuel Revenue Increase Or Decrease Class % Non-Fuel Revenue Increase or Decrease 15B 30 33B Univ. Manuf. Station Power 7.78% 18.45% (3.67%) 35B Large Power 12.6% 6 20 Private Street Lighting Lighting (23.05%) 10.01% Exh. SC-9, pp. 2-3, row 13, to Chan Direct. The results of PNM’s ECCOS study show that PNM’s Residential Class is responsible for 75%, or $91.78 million of PNM’s alleged non-fuel revenue deficiency. Chan Direct 38. Eliminating the entire subsidy for the Residential Class — bringing the Residential Class to “unity44” — would, according to PNM, require a 24.11% non-fuel revenue increase to the Residential Class. Tr. (4-21-16) 2141 (Chan). Ms. Chan said that cost allocation to the Residential Class has increased because the Residential Class’ contribution to peak has increased and therefore its allocation of generation and transmission costs has increased.” Tr. Id. at 2145. PNM provided the following assessment of its current rate structure with respect to cost allocation and rate design: First, PNM’s rate design is entirely outdated and does not accurately reflect the costs the Company incurs to serve its customers. Specifically, current rates are “Unity” or a “relative rate of return” of 1.0, means that a rate class pays the true cost to serve that rate class. Rate classes with a relative rate of return greater than 1.0 subsidize rate classes that have a relative rate of return less than 1.0. Chan Rebuttal 7. The relative rate of return is calculated by dividing net operating income by rate base. Tr. (4-21-16) 2147-48 (Chan). 44 Recommended Decision Case No. 15-00261-UT 200 the product of continued reliance on across-the-board changes applied to marginal cost study results in final rates since 2007. As a result of these across-the-board changes, PNM’s current rates do not accurately reflect the cost to serve its customers and have not for some time. Second, given past stakeholder resistance to efforts to increase customers’ fixed charges, the current rate design places the recovery of too much of the Company’s fixed costs in the volumetric charges. Chan Direct 9-10. For each rate class except Large Power, PNM’s proposed rates move toward unity, meaning that all rate classes except Large Power are moving toward 1.0 when the Base Period is compared to the Test Period. A relative return of 1.0 means that a rate class is responsible for all costs that PNM incurs to serve that class. Chan Rebuttal 8. Ms. Chan said that “for each rate class, the goal is to move people closer to 1.0.” Tr. (4-21-16) 2144. A. BANDING 1. PNM’S PROPOSAL PNM does not propose to increase rates for each rate class based on the allocated revenue requirement amounts resulting from its ECCOSS. This is because the resulting rate increases to the Residential and Irrigation classes would be too dramatic. To mitigate the rate impact to the Residential and Irrigation classes, PNM proposes to apply “banding” so that no class receives a greater revenue increase than 110% of the system average and no class receives a lower revenue increase than 65% of the system average. PNM witness Chan said in her Direct Testimony, “From an overall perspective, the proposed non-fuel revenue requirement increase for the system is 14.19%.” Chan Direct 38. However, PNM does not propose to apply banding to Class 11B — Water & Sewage. PNM proposes to increase revenues collected from Class 11B by only 3.3%. Chan Direct 44. This proposal to exclude Rate 11B from banding is discussed in Section XXII(A)(3). Also, PNM proposes to adjust the revenues allocated to each class after banding to recover the discounts received by PNM customers who participate in PNM’s Incremental Recommended Decision Case No. 15-00261-UT 201 Interruptible Power Rate. After this adjustment, Rate 35B would actually receive a revenue decrease. Chan Direct 45. This proposal is discussed in Section XXII(A)(2). NMIEC witness Phillips correctly pointed out that PNM miscalculated its requested percentage non-fuel revenue increase because it compared existing total revenues (fuel and non-fuel) to proposed non-fuel revenues only, which is an apples-to-oranges comparison and understates the percentage increase in non-fuel revenues. Phillips Direct 12-13, 35. An apples-to-apples comparison — existing non-fuel revenues only to proposed non-fuel revenues only — results in a higher revenue allocation to the Residential Class because the Residential Class, compared to other Classes, has a lower load factor, and a lower percentage of a residential customer’s bill arises from fuel charges. Tr. (4-27-16) 3158 (Phillips). PNM defends its comparison of current total revenues to requested non-fuel revenues only. It says that comparing current non-fuel revenues only to proposed non-fuel revenues only does not more accurately represent PNM’s proposed non-fuel revenue requirement increase “when determining customer bill impacts.” Chan 2-29-16 Supp. 4-5. Ms. Chan acknowledged that comparing current non-fuel revenues only to proposed non-fuel revenues only, results in a higher percentage non-fuel revenue increase to the Residential Class — 20.85%. And, she says that PNM would have to adjust its banding proposal if banding is based on the increase in current non-fuel revenues only to proposed non-fuel revenues only. Id. 7-9; Tr. (4-21-16) 226465. Any proposed bands should be determined based on a comparison of the increase from current non-fuel revenues to proposed non-fuel revenues. In the 2012 SPS Rate Case, the PRC rejected the argument that banding should be based on the amount of the total increase in fuel and non-fuel revenues. The Hearing Examiner recommended adoption of Staff’s proposal of a gradualism limit of 1.25 times (125%) the overall non-fuel revenue increase. The Hearing Examiner did not follow the AG’s recommendation to limit the rate increase to any one class to Recommended Decision Case No. 15-00261-UT 202 150% of the overall increase in base rates, including fuel and purchased power. The AG argued that placing limits on the rate increase based only on non-fuel revenues skewed the overall impact. SPS disagreed, arguing that the comparison was “not very meaningful because the fuel and purchased power costs can vary so much over time.” Recommended Decision 225-30. In its Final Order, the PRC rejected the AG’s exception and recommendation of a gradualism adjustment of 150% of the average overall base rate increase, including both fuel and non-fuel revenues. The PRC explicitly found that fuel revenues should not be included in determining a revenue increase ceiling. Final Order 18, ¶ 38. 2. BANDING AND ALLOCATION OF DISCOUNTS FROM INTERRUPTIBLE RATE PNM proposes to apply banding so that no class receives a greater revenue increase than 110% of the system average and no class receives a lower revenue increase than 65% of the system average. Based on PNM’s calculation of its proposed revenue increase by comparing current total revenues to requested non-fuel revenues only, Ms. Chan said: The upper band means no rate schedule will see an increase higher than 15.6%. The lower band implies no rate schedule will see an increase less than 9.22%. Chan Direct 44 (emphasis added). Ms. Chan used the word “implies” because revenues collected from customers moving onto PNM’s new proposed Rate 35B would actually decrease by (2.14%). This is because PNM proposes to recover from all customer classes the discounts received by customers who participate in Rider No. 8, Incremental Interruptible Power Rate (IIPR). The IIPR is available to qualifying customers in Classes 3C, 4B, and 35B (if approved) who can interrupt their incremental On-Peak billed demand requirements during the on-peak period. A participating customer’s demand rate is discounted during the interruption period. PNM seeks to include in its revenue requirement the amount of the discounts received by participating customers. Tr. (4-21-16) 2260 (Chan). For the Test Period, PNM estimates $1,326,140 total discounts to customers participating in the IIPR. The projected discounts to the participating Classes are: Recommended Decision Case No. 15-00261-UT 203 Class Projected Discount ($109,224) ($124,092) ($1,092,804) 3B/3C 4B 35B Exh. SC-9, pp.2-3, row 27, to Chan Direct; Tr. (4-21-16) 2260-61 (Chan). In its Cost of Service Study, PNM allocates the total $1,326,140 in projected discounts to all customer classes, to be recovered from all classes through rates. The amounts allocated to each Class, using a 12CP allocator, are: Class Allocation $616,396 $162,720 $275,649 $152,365 $11,574 $3,415 $13,297 $9,222 $54,859 $361 $16,734 $2,280 $7,269 1A/1B 2A/2B 3B/3C 4B 5B 10A/10B 11B 15 30 33B 35B 6 20 Exh. SC-9, pp.2-3, row 28, to Chan Direct; Tr. (4-21-16) 2260-61 (Chan). The following table shows the percentage revenue increase by class based on PNM’s calculation of its proposed revenue increase by comparing current total revenues to requested non-fuel revenues only. The second-to-last row shows the percentage revenue increase by class before allocation of the discounts. The last row shows the percentage revenue increase by class after allocation of the discounts. The table shows that the revenue increase to Rate 35B would be 9.22% before allocation of the discounts but negative (2.14%) after the allocation. Two of three customers who would receive service under Rate 35B participate in the IIRP. Tr. (4-2216) 2535 (Aguirre). Recommended Decision Case No. 15-00261-UT 204 Figure SC-5 Currently, PNM does not allocate its lost revenues from IIRP participation for recovery from all of its rate classes. Rather, it allocates the lost revenues for recovery only from its Classes that participate in the IIRP. Tr. (4-22-16) 2534 (Aguirre). This allocation method was approved by the PRC in Case No. 2761, in which the PRC approved continuing PNM’s Experimental IIPR on a non-experimental basis. The Hearing Examiner’s Certification of Stipulation and Recommended Decision says, “Ms. Cost testified that the credits for the EDR riders and the IIPR rider were made part of the revenue responsibility of the rate class in which those customers belonged, and she quantified the credit amounts by class.” Certification of Stipulation & Recommended Decision 22. Ms. Cost explained in her testimony: [T]he credits calculated for customers under the EDR and the EIIPR rider were made part of the revenue responsibility of the rate class in which these customers belong. If this calculation had not been made, the revenue requirements would Recommended Decision Case No. 15-00261-UT 205 have increased in the 1996 test year by the amount of the credits in order for PNM to be kept whole. 0DUJDUHW &RVW 'LUHFW 7HVWLPRQ\ LQ 6XSSRUW RI 6WLSXODWLRQ PNM did not make clear in its prefiled testimony in this case that it seeks to change its method of allocating and recovering the IIPR discounts. The only discussion of this issue is on page 18 of Mr. Aguirre’s Direct Testimony, as follows: Q. WHAT METHOD IS PNM PROPOSING TO USE TO ALLOCATE THE RIDER 8 IIPR DISCOUNTS TO CUSTOMER CLASSES? A. Under the terms of the Rider 8 IIPR tariff, load interruptions can occur on any given month throughout the year. This tariff provides all retail customers with reliability benefits in the event that an emergency interruption is called upon Rider 8 IIPR customers. Therefore, PNM is utilizing a 12CP allocator for the assignment of the Rider 8 IIPR discounts to customer classes in this case. In response to this testimony alleging that the IIPR “provides all retail customers with reliability benefits in the event that an emergency interruption is called upon Rider 8 IIPR customers,” the Hearing Examiner issued a Bench Request to PNM for the following information: 1. For each calendar year 2010 through 2015, list each interruption made under PNM’s Rate Rider No. 8 and the length of each interruption. 2. For the base period in this case, state the number of interruptions made under PNM’s Rate Rider No. 8 and the length of each interruption. 3. For the Test Period in this case, state the number of interruptions made under PNM’s Rate Rider No. 8 and the length of each interruption used by PNM. PNM provided the following tables in its response to the Bench Request: Recommended Decision Case No. 15-00261-UT 206 7DEOH -&$ 0DUFK 6XSSOHPHQWDO ZĞƐƉŽŶƐĞ ϭ͗ &Žƌ ĞĂĐŚ ĐĂůĞŶĚĂƌ LJĞĂƌ ϮϬϭϬ ƚŚƌŽƵŐŚ ϮϬϭϱ͕ ůŝƐƚ ĞĂĐŚ ŝŶƚĞƌƌƵƉƚŝŽŶ ŵĂĚĞ ƵŶĚĞƌ WEDΖƐ ZĂƚĞ ZŝĚĞƌ EŽ͘ ϴ ĂŶĚ ƚŚĞ ůĞŶŐƚŚ ŽĨ ĞĂĐŚ ŝŶƚĞƌƌƵƉƚŝŽŶ͘ >ŝŶĞ η zĞĂƌ ǀĞŶƚ ĂLJ ĂƚĞ ĂůůĞĚ ^ƚĂƌƚͲ ϯϬ ŵŝŶ ĂĨƚĞƌ ĞǀĞŶƚ ŶĚ ;,,͗DDͿ ƵƌĂƚŝŽŶ ;,,͗DDͿ ;,,͗DDͿΎ ŝƐ ĐĂůůĞĚ ;,,͗DDͿ ϭ ϮϬϭϬ tĞĚ͕ ϬϲͬϭϲͬϮϬϭϬ ϭϭ͗ϬϬ ϭϭ͗ϯϬ ϭϵ͗ϬϬ ϳ͗ϯϬ Ϯ ϮϬϭϬ dƵĞ͕ ϭϭͬϭϲͬϮϬϭϬ ϭϭ͗ϬϬ ϭϭ͗ϯϬ ϭϵ͗ϬϬ ϳ͗ϯϬ ϯ ϮϬϭϭ dƵĞ͕ ϬϴͬϮϯͬϮϬϭϭ ϭϰ͗ϯϬ ϭϱ͗ϬϬ Ϯϭ͗ϬϬ ϲ͗ϬϬ ϰ ϮϬϭϮ zĞĂƌ ϮϬϭϮ EŽŶĞ Eͬ Eͬ Eͬ ϱ ϮϬϭϯ DŽŶ͕ ϬϴͬϮϲͬϮϬϭϯ ϭϮ͗ϬϬ ϭϮ͗ϯϬ ϮϬ͗ϬϬ ϳ͗ϯϬ ϲ ϮϬϭϰ zĞĂƌ ϮϬϭϰ EŽŶĞ Eͬ Eͬ Eͬ ϳ ϮϬϭϱ zĞĂƌ ϮϬϭϱ EŽŶĞ Eͬ Eͬ Eͬ ZĞƐƉŽŶƐĞ Ϯ͗ &Žƌ ƚŚĞ ďĂƐĞ ƉĞƌŝŽĚ ŝŶ ƚŚŝƐ ĐĂƐĞ ; Ɖƌ͘ ϮϬϭϰ Ͳ DĂƌ͘ ϮϬϭϱͿ͕ ƐƚĂƚĞ ƚŚĞ ŶƵŵďĞƌ ŽĨ ŝŶƚĞƌƌƵƉƚŝŽŶƐ ŵĂĚĞ ƵŶĚĞƌ WEDΖƐ ZĂƚĞ ZŝĚĞƌ EŽ͘ ϴ ĂŶĚ ƚŚĞ ůĞŶŐƚŚ ŽĨ ĞĂĐŚ ŝŶƚĞƌƌƵƉƚŝŽŶ͘ >ŝŶĞ η ǀĞŶƚ ĂLJ ĂƚĞ ĂůůĞĚ ^ƚĂƌƚͲ ϯϬ ŵŝŶ ĂĨƚĞƌ ĞǀĞŶƚ ŶĚ ;,,͗DDͿ ƵƌĂƚŝŽŶ ;,,͗DDͿ ;,,͗DDͿΎ ŝƐ ĐĂůůĞĚ ;,,͗DDͿ ϴ Ɖƌŝů ϮϬϭϰ Ͳ DĂƌĐŚ ϮϬϭϱ EŽŶĞ Eͬ Eͬ Eͬ ZĞƐƉŽŶƐĞ ϯ͗ &Žƌ ƚŚĞ ƚĞƐƚ ƉĞƌŝŽĚ ŝŶ ƚŚŝƐ ĐĂƐĞ ;KĐƚ͘ ϮϬϭϱͲ^ĞƉ͘ ϮϬϭϲͿ͕ ƐƚĂƚĞ ƚŚĞ ŶƵŵďĞƌ ŽĨ ŝŶƚĞƌƌƵƉƚŝŽŶƐ ƵŶĚĞƌ WEDΖƐ ZĂƚĞ ZŝĚĞƌ EŽ͘ ϴ ĂŶĚ ƚŚĞ ůĞŶŐƚŚ ŽĨ ĞĂĐŚ ŝŶƚĞƌƌƵƉƚŝŽŶ ƵƐĞĚ ďLJ WED͘ >ŝŶĞ η ǀĞŶƚ ĂLJ ĂƚĞ ĂůůĞĚ ^ƚĂƌƚͲ ϯϬ ŵŝŶ ĂĨƚĞƌ ĞǀĞŶƚ ŶĚ ;,,͗DDͿ ƵƌĂƚŝŽŶ ;,,͗DDͿ ;,,͗DDͿΎ ŝƐ ĐĂůůĞĚ ;,,͗DDͿ ϵ KĐƚŽďĞƌ ϮϬϭϱͲ^ĞƉƚĞŵďĞƌ ϮϬϭϲΎΎ EŽŶĞ Eͬ Eͬ Ύ,ŽƵƌƐ ĂƌĞ ŶŽƚ ĂĚũƵƐƚĞĚ ďLJ ĚĂLJůŝŐŚƚ ƐĂǀŝŶŐ ƚŝŵĞ͘ ΎΎWED ĐĂŶŶŽƚ ƉƌĞĚŝĐƚ ǁŚĞƚŚĞƌ ĂŶ //WZ ŝŶƚĞƌƌƵƉƚŝŽŶ ǁŝůů ŽĐƵƌƌ ĨƌŽŵ ƚŚĞ ƉƌĞƐĞŶƚ ĚĂLJ ƚŚƌŽƵŐŚ ^ĞƉƚĞŵďĞƌ ϮϬϭϲ Eͬ These tables show that PNM made no interruptions under the IIPR in 2014 or 2015 or the Base Period and projects no interruptions in the Test Period. Currently, eight customers subscribe to the IIPR: five Class 3C cust0mers; one Class 4B customer; and two customers who will receive service under 35B if that rate is approved. Tr. (422-16) 2535 (Aguirre). When asked, given that PNM made no interruptions in 2014 and 2015 and in the Base Period and projects no interruptions in the Test Period, how the IIPR “provides all retail customers with reliability benefits in the event that an emergency interruption is called upon Rider 8 IIPR customers?” Mr. Aguirre answered: [P]robably somebody more familiar with how the system is operated will answer whether or not this capacity is providing benefits. I’m not sure that I can answer that question because I don’t know how to quantify those benefits for reliability purposes. Recommended Decision Case No. 15-00261-UT 207 When next asked, “But you do say in your testimony that it provides reliability benefits. So are you saying you’re not really sure if that’s true?” Mr. Aguirre answered: No, I think there are reliability benefits. What I don’t know is if the value or the discount we offer today are fair or pay for the benefits that the customers are providing through those interruptions or potential interruptions. Tr. (4-22-16) 2537-38. Mr. Aguirre said that in 2015, PNM made 15 interruptions under its Residential Power Saver and Peak Saver load-management programs. He did not know why PNM chose to interrupt residential customers under that program, but not customers who participate in the IIPR. Tr. (4-22-16) 2540. PNM’s proposed allocation of the IIPR discounts to all customer classes for recovery should be denied because PNM did not request approval to change its allocation method, and because, even if the PRC considered changing the allocation method, PNM did not justify that it should be changed. PNM provided no evidence in support of Mr. Aguirre’s assertion that the IIPR “provides all retail customers with reliability benefits in the event that an emergency interruption is called upon Rider 8 IIPR customers.” PNM should continue to allocate the IIPR discounts only to participating rate classes for recovery. Given that PNM expects only three customers to receive service under new Rate 35B, the total discounts should be collectively recovered from the three eligible classes. Tr. (4-21-16) 2271-72 (Chan); Tr. (4-22-16) 2534-35 (Aguirre). Given the evidence that PNM made no interruptions in 2014 and 2015 or in the Base Period under the IIPR and projects no interruptions in the Test Period, parties and Staff are put on notice that the PRC will consider in PNM’s next rate case whether the IIPR should be discontinued. If PNM proposes to continue the IIPR in its next rate, it shall file direct testimony justifying why it should continue. Recommended Decision Case No. 15-00261-UT 208 3. RATE 11B — WATER & SEWAGE PNM originally proposed to exclude Rate 11B from banding. In its posthearing briefs, PNM says that the PRC should reconsider excluding Rate 11B from banding if the PRC does not approve PNM’s proposed change in the Time of Use Period. ABCWUA disagrees. This issue relates to the Amended Stipulation in the 2010 PNM Rate Case, in which PNM agreed to: [R]educe any monthly CP demand for Rate Schedule 11B where the monthly CP date and time occur during a current PNM TOU off-peak hour. The amount of the reduction will recognize Rate Schedule 11B customers’ operational load shifting capabilities, and will be determined jointly, in good faith, by PNM and the Rate Schedule 11B customers. PNM and the Rate Schedule 11B customers will determine, in good faith, whether reductions should be made to Rate Schedule 11B CP demands occurring within a current PNM TOU on-peak hour to adjust demands to appropriately recognize Rate Schedule 11B’s operations and load shifting capabilities. Amended Stipulation, ¶ 39 (8-11-11). The goal of Paragraph 39 of the Amended Stipulation was to ensure that Rate 11B customers are not unduly penalized by PNM’s Proposed TOU Period adjustment. Aguirre Direct 66. PNM has 157 customers who receive service under Rate 11B. ABCWUA is PNM’s largest customer in Class 11B. Tr. (4-22-16) 2545. PNM and ABCWUA met and reached an agreement. However, what ABCWUA and PNM agreed to is not clear, and PNM and ABCWUA disagree on whether the agreement applies if the PRC does not shifting the TOU period. One of the questions that Mr. Aguirre answers in his Direct Testimony is, “What is the mutually-agreed solution to satisfy the requirements of Paragraph 39 of the Amended Stipulation?” Mr. Aguirre answered: As PNM Witness Chan notes, it was agreed that the simplest and most direct solution was to shift the Base Period data by two hours such that all hourly Rate 11B — Water and Sewage load data for the Base Period simulated the customers’ load shifting capabilities as a result of the Proposed TOU Period shift. Specifically, the proposed resolution moves the CP demand for the Base Period for the Rate 11B — Water and Sewage class from 8 AM to 8 PM, Monday through Recommended Decision Case No. 15-00261-UT 209 Friday, (Current TOU Period) to 10 AM to 10 PM, Monday through Friday (Proposed TOU Period). In addition, if the system peak for a particular month in the Base Period occurs during a weekend day, the proposal moves the Rate 11B — Water and Sewage CP to the nearest on-peak hour. Aguirre Direct 65-66. The result of the agreement is a 12% decrease to the Class 11B CP demands for the Base Period. Mr. Aguirre said that reducing the CP demands would reduce Class 11B customers’ allocation of generation and transmission plant revenue responsibility, so other classes would be allocated the costs of this reduction. Mr. Aguirre said that while any revenue shift to other rate classes “as a result of a benefit to one class warrants scrutiny,” PNM believes that the change is consistent with the Amended Stipulation “and is appropriate given the responsiveness to TOU pricing that this class has demonstrated over the years.” Aguirre Direct 66-67. Exhibit 12 to Ms. Chan’s Direct Testimony contains a letter from Ms. Chan to ABCWUA. The letter says in part: What follows is a description of development of the jointly supported methodology and the results of that methodology. This new methodology is referred to as the “Shifting All Hours Case”. 1. As background, for this rate case PNM will propose a change to its TOU peak period, which shifts the peak period by two hours from 8 AM to 8 PM to a proposed 10 AM to 10 PM Monday through Friday. 2. To adjust CP demand, all of the hourly Rate 11B load information for the Base Year was shifted so that the class now appears to operate on the proposed TOU peak period of 10 AM to 10 PM Monday through Friday. 3. Using the shifted hourly loads, CP loads were then pulled for the Base Year’s date and time of each monthly system CP. 4. If a CP for a month occurred during a weekend, that CP load was adjusted down to the value of the nearest proposed on-peak hour. Exh. 12, pp. 6-7 to Chan Direct (footnotes omitted). No Intervenor or Staff filed response testimony on the agreement between PNM and ABCWUA. ABCWUA witness Herz’s Direct Testimony addresses PNM’s proposed shift in its TOU period. In that Testimony, he identified shortcomings with PNM’s analysis supporting the Recommended Decision Case No. 15-00261-UT 210 Proposed TOU Period and said that while ABCWUA could adapt to the shift, he recommended deferring a shift pending a more complete analysis. His Direct Testimony does not discuss the agreement between PNM and ABCWUA. Herz Direct 10-14. The Hearing Examiner questioned Mr. Herz at the hearing to obtain clarification of PNM and ABCWUA’s agreement. The Hearing Examiner told Mr. Herz that his Direct Testimony indicated that ABCWUA does not support the shift in the TOU period, but Mr. Aguirre’s Direct Testimony indicated that ABCWUA does support the shift under the parties’ agreement. Mr. Herz said that Mr. Aguirre’s Direct Testimony addresses two matters. He said that one matter, which addresses the Amended Stipulation is: [T]hat if there is a time shift for the on-peak period, that the company would work with the Water Authority to go about the process adjusting the demands, the coincident peak demands that may have occurred outside that period. And that was done. That process was done and completed. Tr. (4-25-16) 2807. He continued, saying that “the second part of what’s addressed here [in Mr. Aguirre’s Testimony” relates to PNM’s Proposed TOU Period. Mr. Herz said, “But it’s my understanding that no agreement was reached between the company and the Water Authority in support of the two-hour shift.” Id. at 2807-08. Mr. Herz said that ABCWUA does not necessarily support the change in the TOU period. Id. at 2815. In a continued attempt to obtain clarification, the Hearing Examiner asked Mr. Herz: Q. So are you saying the methodology that the Water Authority and PNM agreed to, that the Water Authority supports that methodology regardless of whether the commission changes the TOU period? Mr. Herz answered: A. Yes, but if I can just add a clarification to that. If the commission does not approve the proposal, well, that means then the CP demands that are in the filing and used in the class cost of service study for assigning cost responsibility would go back to what they were. But by the methodology that the Water Authority and the company agreed to, it would not affect the class cost of service results. So the class cost of service results here are appropriate and would work regardless of whether the commission adapts or adopts or denies the proposal to shift the TOU peak period two hours. Recommended Decision Case No. 15-00261-UT 211 Id. at 2811-12. The Hearing Examiner then asked Mr. Herz to explain the method agreed to by ABCWUA and PNM. Mr. Herz answered: Basically what we did is utilize the actual load pattern for the Water Authority’s 11B account and just shifted that load pattern by two hours to correspond with the time-of-use shifting by two hours. . . . Now, what that impacted, though, was that for purposes of assigning cost responsibility, that moved the demands at the time of the company’s peak. Id. at 2812. Upon recross examination, Mr. Herz agreed that PNM proposed to exclude Rate 11B from banding in order to effectuate some of the goals of the stipulation. However, he said that ABCWUA should be excluded from banding even if the CP demand shift is not needed. When asked: “Does your direct testimony address that ABCWUA should be left out of the banding process as it was if the TOU shift did not occur?” he responded, “I guess I didn’t see where that would be necessary.” Id. at 2816-18. In its Initial Posthearing Brief, PNM says that the CP adjustment and exclusion of Rate 11B from banding “[were] specifically and explicitly stated to be in accordance with Paragraph 39 of the Amended Stipulation, which presumed that the TOU pricing period would be changing.” PNM says that if the PRC does not approve shifting the TOU Period, “the Commission should re-visit PNM’s proposal to leave Rate 11B out of the banding process.” PNM’s Initial Brief in Chief 326-27. In its Response Brief, ABCWUA refers to this portion of PNM’s Brief as “PNM’s BrandNew Rate 11B Increase.” It asserts, “Not only is PNM’s suggestion to ‘re-visit [its] proposal to leave Rate 11B out of the banding process’ bad manners, it also denies the due process built into the purpose for pre-filing testimony and a hearing on the pre-filed evidence.” ABCWUA suggests that PNM’s position cannot be considered because “each and every party to this Recommended Decision Case No. 15-00261-UT 212 proceeding (including PNM) that filed testimony on this matter did not include Rate 11B in the banding process.” ABCWUA’s Response Brief 28-31. ABCWUA concludes its response to this issue by saying, “[T]he Water Authority recommends the Commission defer any shift until a more complete analysis of TOU scenarios and periods has been conducted.” Id. at 31. ABCWUA’s attack on PNM is, by extension, an attack on the Hearing Examiner because it is the Hearing Examiner who tried to get clarification at the hearing on whether PNM and ABCWUA’s agreement is contingent on shifting the TOU period. ABCWUA and PNM obviously disagree on this point. Given this lack of agreement and the Hearing Examiner’s recommendation to not change the TOU period (which ABCWUA supports), the “CP adjustment45” and exclusion of Rate 11B from banding should be rejected. Even if the Hearing Examiner were recommending changing the TOU period, she would not recommend approving the CP adjustment or excluding Rate 11B from banding. It is still not clear what the “CP adjustment is,” but, according to Ms. Aguirre, it would result in significant shifting of revenue responsibility from Rate 11B to PNM’s other classes. However, Mr. Aguirre did not state the amount of revenue that would be shifted. The PRC cannot approve an apparently significant shift in revenue responsibility among classes without knowing the amount, understanding what adjustment is proposed, and why that adjustment should be approved. This evidence must be submitted in any future case in which these changes are proposed. 4. ANALYSIS AND HEARING EXAMINER’S RECOMMENDATION ON BANDING The PRC has applied banding without calling it “banding.” In the 2007 SPS Rate Case, the PRC adopted SPS’s proposal to limit the percentage increases to the Residential Class and Lighting Class to 150% of the overall percentage increase. Adherence to the results of SPS’s cost of service study would have resulted in rates for the Lighting Class increasing 28.66%, rates for 45 PNM’s Initial Posthearing Brief 326. Recommended Decision Case No. 15-00261-UT 213 the Residential Class increasing 11.39%, and rates for the Small General Service Class decreasing 6.75%. SPS’s gradualism adjustment produced a 7.30% rate increase for the Residential and Lighting Classes, which “move[d] the Residential and Lighting classes in the direction of meeting their costs of service in a fair and reasonable manner.” Recommended Decision 177-83. In the 2012 SPS Rate Case, followed its determination from the 2007 SPS Rate Case that “SPS’s proposal to limit rate increases for any one class to 150% of the system overall non-fuel base rate increase, was fair and reasonable.” The PRC found that this limit was “a reasonable gradualism approach” that appropriately balanced the competing interests of charging a rate class its cost of service while tempering the rate shock that a rate class may otherwise experience.” Final Order 18-20, ¶¶ 38-41. In the 2012 SPS Rate Case, the PRC said: “[I]n designing rates, generally accepted principles of rate design are utilized. These principles include continuity, bill impact, ease of understanding, and gradualism.” Final Order 17, ¶ 35. The PRC then explained: Simply stated, gradualism is an adjustment utilized to place a limit on the relative percentage increase each rate class receives compared to the percentage increase for each rate class resulting from the class cost of service study. The Commission has previously determined that gradualism should achieve a balance between the goal of eliminating or reducing cross-subsidies among rate classes and shielding customers from rate shock. Id. at 17-18, ¶ 36. NMIEC supports PNM’s proposed lower band of 65% but recommends an upper band of 150%. Tr. (4-27-16) 3159 (Phillips). Mr. Phillips said that PNM’s band is “excessively narrow” and would only result in “glacial” movement toward cost-based rates. He said that if the PRC reduces PNM’s proposed revenue requirement, the bands could be widened without excessively burdening the Residential and Irrigation Classes. Phillips Direct 35-37. PNM says that its proposed rate design results in about $37 million in interclass subsidies, with about $32.4 million flowing to the Residential Class. Chan Direct 47. Other classes receiving subsidies would include: Irrigation, Manufacturing, Large Power, and Recommended Decision Case No. 15-00261-UT 214 Streetlighting. Exh. SC-9, p.4, row 33, to Chan Direct; Tr. (4-21-16) 2139-40 (Chan). NMIEC says that its proposed rate design reduces interclass subsidies to about $29.6 million, with about 94% flowing to the Residential Class. Phillips Direct 39. Under PNM’s proposed rates, Rate 30B would receive a 9.22% revenue increase. Exh. SC-9, p.3, to Chan Direct. Setting rates at a level to recover all of Rate 30B’s cost of service would require a 18.45% revenue increase. Chriss Direct 16. At PNM’s proposed revenue requirement, Walmart recommends setting the revenue requirement increase for Rate 30B at a level equal to PNM’s proposed 15.6% banding cap. Walmart argues that limiting the revenue increase for Rate 30B to 9.22% is inequitable in light of the 14.2% revenue increases to Rates 2, 3, and 4 under PNM’s proposed rates. Chriss Direct 5, 16. Walmart Witness Chriss proposed that if the Commission reduces PNM’s proposed revenue requirement, the following steps should be taken: (1) bring Rate 30B up to the 15.6% banding cap; (2) increase the revenue requirement allocation for Rate 20 to its cost of service; (3) allocate 50% of the reduction of the Company’s proposed revenue requirement increase to the customer classes that bear the subsidy burden; and (4) allocate the remaining 50% of the reduction to the Company’s proposed revenue requirement increase to all customer classes on an equal percentage basis. Chriss Direct 5. Additionally, Mr. Chriss said that no customer class should receive a decrease, and no customer class that currently bears the subsidy burden should be moved to a subsidized position. Id. at 6. The New Mexico Supreme Court has explicitly discouraged the use of cost of service as a sole criterion in designing rates. In re PNM Gas Servs., 2000-NMSC-012, ¶ 100, 129 N.M. 1. And, the Court has reversed the PRC when it departed from the principle of gradualism. Particularly, the Court said that by increasing a utility’s residential access fee by over 60%, the PRC “improperly overlooked the potential that a dramatic shift in rates for residential Recommended Decision Case No. 15-00261-UT 215 consumers would cause rate shock, and the Commission thereby violated the fundamental ratedesign principle of stability in rates.” Id., ¶ 102. At the April hearing, former PRC Commissioner Doug Howe was asked: Do you think that a goal of state public utility commissions should be to eventually eliminate the subsidy to residential customers if gradualism is applied? He answered: I think as a general proposition, and I’ve stated before that the cost — those who incur the costs should bear the costs. I believe that as a general principle and that it should be one of the guiding rules of this or any commission. The caveat is gradualism. And then on top of that, I would add that there are a lot of very valid social, what is called social engineering reasons why you might — why a commission might want to deviate from that. Tr. (4-25-16) 2680-81. Under PNM’s ECCOSS, revenue allocation to the Residential Class is the driver behind the amount of revenue allocated to other classes because the Residential Class is far below its share of revenue recovery responsibility. The amount of revenue increase to allocate to the Residential Class is a judgment call based on the authorized revenue requirement. See Mountain States Tel. & Tel. Co. v. New Mexico State Corp. Comm’n, 1977-NMSC-032, ¶ 65, 90 N.M. 325 (“Determining the level of subsidies, if any, is a Commission function.”). PNM’s proposal to apply upper and lower bands to the amount of revenue to be allocated among customer classes should be adopted. It is a practical way to ensure that no class or classes suffer rate shock from a rate increase. It is a recognition of “fairness” as a restraint “against the unqualified acceptance of general principles of ratemaking based on considerations of maximum economic or social efficiency.” Bonbright 192. In determining what percentage of its requested revenue requirement to allocate among classes, PNM allocated to the Residential and Irrigation Classes — which its ECCOSS results show are being subsidized by other classes — a greater percentage of the revenue requirement than the system average with the goal of moving to cost-based rates. PNM allocated additional Recommended Decision Case No. 15-00261-UT 216 revenue requirements to other classes to effectuate the mitigation for the Residential and Irrigation Classes, but allocated to them a lesser percentage of the revenue requirement than the system average. Chan Direct 46. PNM’s reasoning is reasonable and should be applied to the Hearing Examiner’s proposed banding. NMIEC witness Phillips explained that if PNM’s requested revenue requirement is reduced, the bands can be widened without excessively burdening the Residential and Irrigation Classes. Phillips Direct 36. PNM said that it would no longer support its proposed bands if the bands are based on a comparison of current non-fuel revenues only to proposed non-fuel revenues only. Chan 2-25-16 Supp. 9. Under the Hearing Examiner’s recommended revenue requirement, the system average revenue increase is 6.43%. Applying a 130% upper band and a 65% lower band achieves a just and reasonable result: x The Residential and Irrigation Classes each receive an 8.36% non-fuel revenue increase; x The Small Power, General Power, and Large Power Classes each receive a 4.84% nonfuel revenue increase; and x The remaining classes receive a 4.18% non-fuel revenue increase. See Attachment E. Attachment F to this Recommended Decision shows movement toward and away from unity for each class under PNM’s current rates, PNM’s proposed rates, NMIEC’s proposed rates, and the Hearing Examiner’s proposed rates. B. PNM’S PROPOSED CHANGES TO ITS FIXED CUSTOMER CHARGES Except for the Irrigation Classes (Rates 10A/10B), PNM proposes to adjust its customer charges to recover all of each Class’ customer-related costs through each Class’ customer charge. Customer charges would recover costs for customer service, meters, billing, meter reading, bill processing and other customer-related activities. Chan Direct 7. PNM proposes to adjust the Recommended Decision Case No. 15-00261-UT 217 customer charges (and meter charges for the Irrigation Class) for the retail classes with a twopart tariff as follows (only non-TOU rates shown): Rate Class Current Customer Charge PNM Proposed Customer Charge Current Meter Charge PNM Proposed Meter Charge % Increase in Meter Charge $13.14 % Increase in Customer Charge 162.8% 1A – Residential 2A – Small Power 10A – Irrigation 11B – Water & Sewage $5.00 N/A N/A N/A $8.46 $17.87 111.2% N/A N/A N/A $8.19 $30.03 266.6% $2.81 $17.32 516% $491.60 $327.75 (33.3%) N/A N/A N/A For these rate classes — which are not subject to a demand charge — all of the demandrelated costs would be recovered through the volumetric charges. Aguirre Direct 14. PNM says, “To avoid a significant increase to this rate Class’ customer charge,” it proposes a customer charge for Rate 10A – Irrigation and 10B – Irrigation Time-of-Use that covers only 50% of the customer-related costs for these rate classes. Chan Direct 51, n.25 (emphasis added). For several rate classes, PNM proposes to reduce the customer charge to recover only customer-related costs. Previously, the customer charges for these classes recovered both customer-related costs and minimum demand. PNM proposes to recover minimum demand through its seasonal demand rate. Chan Direct 52; Aguirre Direct 30. PNM’s proposes to set customer charges to recover the cost it incurs to connect and provide basic customer services to each class for several reasons. First, from a cost causation, cost recovery and rate design perspective, PNM says it is appropriate to recover these customerrelated costs through a fixed monthly charge. Costs for meters, billing, meter reading, bill processing, customer service and other customer-related activities are constant for every customer in a given rate class. The cost for these services does not change with sales and the Recommended Decision Case No. 15-00261-UT 218 delivery of electricity. Regardless of the amount of electricity a customer uses, PNM has to install a meter, read that meter monthly, set up an account in the billing system, process the bill monthly and have customer service available to assist customers when the need arises. Chan Direct 52-53.46 Second, PNM says that increasing the Residential Class’ customer charge is an important first step to address the growing subsidy for this class. Currently, the Residential Class customer charge recovers less than 10% of the Residential Class’ fixed costs. PNM says, “The proposed increased customer charge is a small, reasonable step toward increased fixed cost recovery from this rate class and may alleviate further growth in the subsidy.” Id. at 53 (emphasis added). Increasing the residential customer charge does not mean that PNM would recover all of its fixed costs associated with serving residential customers through the customer charge. In addition to customer-related costs, PNM incurs other fixed costs to serve residential customers, including primary and secondary distribution, transmission, substation, and generationdemand costs. If PNM included these additional fixed costs in the Residential customer charge, it would have to collect an additional $48.48 from residential customers, which would result in an estimated $61.62 total customer charge. Id. Staff and some Intervenors oppose PNM’s proposed Residential customer charge. Staff also opposes PNM’s proposed Small Power customer charge. PNM’s proposed customer charges for the Irrigation and Water and Sewage Classes are unopposed. C. PROPOSED RESIDENTIAL CLASS RATES 1. POSITIONS OF PARTIES AND STAFF See Chan Direct 53:Table SC-2 (showing the following breakdown of the $13.14 residential customerspecific costs that PNM incurs per month and per customer under the proposed revenue requirement: Customer Service – $1.95; Customer Meter – $2.73; Customer Meter Reading – $1.83; Customer Billing and Processing – $3.38; and Other Customer-Related Activities – $3.25). 46 Recommended Decision Case No. 15-00261-UT 219 The table below shows recommended residential customer charges among those parties and Staff who recommend a residential customer charge: Party Witness PNM AG CCAE NEE Chan Gegax Howat Van Winkle LeybaTercero Staff 2. Current Residential Customer Charge $5.00 $5.00 $5.00 $5.00 Proposed Residential Customer Charge $13.1447 $7.0048 $5.0049 $5.0050 Percentage Increase in Residential Customer Charge 162.8% 40% 0% 0% $5.00 $7.0051 40% COMMISSION PRECEDENT ON RESIDENTIAL RATE DESIGN Discussion of PRC precedent on residential rate design is critical because it demonstrates the PRC’s consistent and repeated rejection of sizable increases in the residential customer charge and application of gradualism constraints. Most recently, in the 2015 EPE Rate Case, the Hearing Examiner recommended a $639,747 increase in EPE’s non-fuel revenues. Recommended Decision 259, ¶ 10. She recommended that the $639,747 be collected by increasing the non-fuel revenue requirements collected from the following classes by the following amounts: Customer Class % Revenue Increase Street Lighting Outdoor Recreational Lighting Seasonal Agricultural Irrigation Service Residential 1.1% 1.1% 4.5% 1.6% 9.0% Final Order 51, ¶ 110. Chan Direct 51. Gegax Direct 16. 49 Howat Direct 12. 50 Tr. (4-28-16) 3328. 51 Leyba-Tercero Direct 7. 47 48 Recommended Decision Case No. 15-00261-UT 220 The Hearing Examiner also recommended increasing the Residential Class relative rate of return close to unity: from 0.6 to 0.989. This increase, combined with increasing the Residential Class’ non-fuel revenue requirement by 9.0%, would result in a 14.05% non-fuel revenue increase, or $7,710,825, to the Residential Class and a $7.62 per month increase to an average-use residential customer’s bill. She recommended decreasing the rates for eight rate classes. Id. at 51-52, ¶ 111-113. The Hearing Examiner also recommended increasing the residential customer charge from $7.00 to $9.00. Id. at 53, ¶ 115. This would have decreased that Class’ variable charges. Id. at 58, ¶ 129. EPE had proposed increasing the residential customer charge from $7.00 to $10.00 to move closer to a cost-based residential charge, but apply gradualism. Moving to a 100% cost-based residential customer charge would have increased the customer charge to $14.79. Staff recommended an $8.50 customer charge, one Intervenor recommended no increase in the customer charge, and two Intervenors recommended a $9.00 customer charge. The Hearing Examiner found that the zone of reasonableness for the customer charge was from no increase to $10.00. She recommended a $9.00 customer charge as balancing the competing goals of avoiding rate shock and moving toward cost-based rates. Id., Recommended Decision 213-15. In its Final Order, the PRC found that a 14% non-fuel revenue increase to the Residential Class “is too great” in light of (1) The PRC’s authorization of a $1.1 million revenue increase, less than 1% of the authorized revenue requirement; and (2) The majority of EPE’s other rate classes would see a decrease in their base rates. The PRC explained that while it “generally agrees that guiding principles of rate setting include moving the rates of each customer class closer to a relative rate of return on 1.00, that principle must be measured against the equally important principles of gradualism and the avoidance of rate shock.” To balance those interests, the PRC allocated the entire $1.1 million Recommended Decision Case No. 15-00261-UT 221 non-fuel revenue increase to the Residential Class and made no changes to rates for the other customer classes. This resulted in a 1.28% non-fuel revenue increase to the Residential Class. Id., Final Order 57, ¶¶ 127-28. The PRC then found that the Hearing Examiner’s recommendation of a $9.00 residential customer charge and lower variable charges “hurts low income and average volume users.” The Hearing Examiner’s recommendations would result in a 2.3% bill increase for a residential customer consuming 500 kWh per month, but only a 1.3% increase for a residential customer consuming 750 kWh per month. The PRC also said that the Hearing Examiner’s recommended rate design “discourages conservation, which can ultimately, and unnecessarily, lead to the need for additional generation and higher rates.” The PRC ordered no increase to the $7.00 residential customer charge and recovery of the remaining revenues allocated to the Residential Class through variable charges. Id., Final Order 58-59, ¶¶ 129-30. In the 2007 SPS Rate Case, SPS initially proposed increasing its residential customer charge from $4.75 to $7.00, a 47% increase. However, in response to Staff testimony proposing a more gradual increase to only $5.71, SPS lowered its proposed increase to $6.00. Staff accepted this compromise. Final Order 34, ¶ 95. In its Final Order, the PRC lowered the residential customer charge to $5.00, explaining: Staff’s concern regarding the disproportionate impact that increasing the service availability charge has on consumers that use lower amounts of energy is well placed and is one shared by the Commission. The Commission believes that a low service availability or similar fixed monthly charge imposed on residential customers is preferable because it imposes less of an economic burden on low energy-consuming customers. However, the Commission does not agree with Staff that reducing the service availability charge to $6.00 per month adequately addresses the Commission’s concern. Based upon the facts and circumstances of this case, the Commission finds that the residential service availability charge should be $5.00 per month, which represents a percentage increase that is substantially similar to the 5.01% average increase for all residential customers[.] Id. at 34-35, ¶¶ 95-96. Recommended Decision Case No. 15-00261-UT 222 In the 2008 PNM Electric Rate Case, the PRC approved a provision in the stipulation that increased PNM’s residential customer charge from $3.10 to $4.00. The Stipulating Parties described this increase as “a small movement toward placing more of PNM’s fixed costs in the non-variable customer charge.” 2008 PNM Electric Rate Case, Final Order 31. In Case No. 10-00379-UT, the PRC rejected Kit Carson Electric Cooperative, Inc.’s proposed more than double increase in its residential customer charge — from $10 to $20.50. Instead, the PRC approved a $14.50 customer charge. In doing so, the PRC said, “[A]ny benefit resulting from the increased stability in earnings and profit resulting from a high fixed monthly charge is counterbalanced, if not outweighed, by the reduction in Kit Carson’s incentive to keep its costs at reasonable levels and reduce the rates to its customers.” Case No. 10-00379-UT, Final Order Partially Adopting Recommended Decision, ¶ 28 (9-20-11). The PRC identified as an important detriment of the Cooperative’s proposed rate design that it would have resulted “in [the Cooperative’s] higher usage, and more affluent customers receiving a rate decrease at the expense of its lower-usage and lower income customers.” Id. at ¶ 31. The PRC said that “the fact that a customer charge may result in certain customers being subsidized by other customers within the same class is not . . . a reason to not approve a rate design.” Id. at ¶ 26. For residential volumetric charges, the PRC prefers inclining block energy rates to promote conservation. For example, in an Indian Hills Waterworks’ rate case, Staff testified that there was insufficient customer data to allow it to propose inclining block rates. The Recommended Decision recommended Staff’s proposal for a single commodity rate. The PRC remanded the case to take additional evidence and consider whether an inclining block rate structure should be adopted. Case No. 11-00089-UT, Partial Remand Order 3, ¶ A (6-7-12). Following remand, the PRC adopted an inclining block rate structure for commodity rates. Final Order (10-4-12). The PRC said, “Inclining block rates are desirable because they encourage conservation.” Supplemental Recommended Decision on Remand 7 (9-14-12). Recommended Decision Case No. 15-00261-UT 223 In the 2010 PNM Rate Case, the PRC rejected the Stipulating Parties’ proposed residential rate design and adopted a rate design that resulted in a lower average annual rate increase for use between 350 kWh and 1,000 kWh per month, which was the range in which a significant number of low-income customers consumed and in which the average residential customer consumed. While the Stipulation’s rate design would have resulted in a lower rate increase for ratepayers consuming up to 200 kWh, the evidence did not support a finding that those ratepayers were mostly low-income customers. Rather, the evidence showed that ratepayers consuming up to 200 kWh often took service at vacant premises, such as empty rentals, homes for sale, and vacation homes. The adopted rate design resulted in a higher average annual rate increase for use above 1,000 kWh, which, the PRC said, “sends a desirable price signal and appropriately recovers more costs from the higher use customers who drive the need for more revenue[.]” Certification of Stipulation 107-09. In its Final Order, the PRC said that the Stipulation’s rate design “provide[d] an unreasonable subsidy to customers consuming 200 kW or less per month and imposes unnecessary increases on the majority of PNM customers, including low income customers.” Final Order 28, ¶ 57. 3. ANALYSIS AND RECOMMENDATION The theoretical basis for PNM’s desire to collect all customer-related costs through the customer charge is well-recognized.52 As more customer-related costs are recovered through the customer charge, a more accurate price signal is sent to customers of the cost to have service available regardless of how much energy is used. Ortiz Direct 40-41. However, in light of PRC precedent, PNM’s proposal to increase the residential customer charge by 162.8% is astonishing. See, e.g., American Water Works Association, Principles of Water Rates, Fees, and Charges 113 (5th ed. 2000) (stating that a cost-of-service approach to setting rates allocates costs to each customer class based on the theory of cost causation and a duel set of fees — fixed and variable — “is an extension of this cost causation theory”). 52 Recommended Decision Case No. 15-00261-UT 224 CCAE and NEE’s proposal to not increase the residential customer charge at all is not reasonable and not supported by the evidence. CCAE and NEE argue that increasing the customer charge would disproportionately harm low-income and elderly ratepayers. CCAE also argues, as does NEE, that increasing the customer charge will discourage energy efficiency. Howat Direct 11-12; Tr. (4-28-16) 3328 (Van Winkle). However, they provided no evidence either particular to PNM or statewide or nationwide that shows a link between the level of the customer charge and participation in energy efficiency programs. See Howat Direct 11-12. Staff and the AG reasonably recognize the desirability of collecting more customerrelated costs through the customer charge, but they also recognize the need for gradualism. Their proposed $7.00 Residential customer charge reflects these competing goals. LeybaTercero Direct 7; Gegax Direct 12-16. PNM’s response that a $2.00 customer charge increase “does not go far enough to count as even a gradual increase,” Chan Rebuttal 46, flatly ignores PRC precedent. The AG, who represents residential and small business customers before the PRC,53 faces a particularly difficult task in recommending the amount of a residential customer charge because, as AG witness Gegax frankly explained, the interests of residential customers differ depending on how much energy they use: For a given residential revenue requirement, a higher customer charge will mean lower average variable charges and vice versa. This trade-off between the customer charge and the variable charge in the rate structure inevitably pits households that have — to any degree — above-average usage against households that have — to any degree — below-average usage. Gegax Direct 16 (emphasis in original). Therefore, households using more energy than the average monthly amount used by a PNM residential customer — 580 kWh — prefer a higher customer charge and lower variable charges, and households using less energy than the average monthly amount prefer a lower customer charge and higher variable charges. “The AG 53 NMSA 1978, § 8-5-17(A). Recommended Decision Case No. 15-00261-UT 225 represents all residential customers and must, therefore, balance the interests of both these groups when it comes to the customer-variable charge trade-off.” Id. at 16. Mr. Aguirre said that PNM’s proposed inclining block rates do not increase by a constant percentage amount. Rather, in designing the rates, PNM considered various factors including the separate impacts on PNM North and South customers, and the amount of the increase between seasons. If the PRC decreases PNM’s requested revenue increase, but wants to follow PNM’s inclining block rate design, it can revise the inclining block rates by decreasing each block rate by the same percentage. Tr. (4-22-16) 2549-50 (Aguirre). AG witness Gegax designed his proposed Residential Class rates to (1) mitigate the bill impact on extremely low use customers, which he defines as using 300 kWh or less monthly; and (2) ensure that average-use customers receive the smallest bill increase. He estimated that his proposed $7.00 residential customer charge would collect about $38.5 million from the Residential Class. He calculated the proposed remaining amount to be recovered from the Residential Class — about $276 million — to be collected from each energy rate block by starting with the percentage shares that PNM proposes to be collected from each block and then adjusting the results to mitigate the bill impact on extremely low use customers and ensure that average-use customers receive the smallest bill increase. Id. at 17-19. PNM witness Aguirre responded that Mr. Gegax’s proposed rate design for the Residential Class’ inclining block rates would have a disproportionate rate impact on PNM South residential customers if the Consolidation Adjustment Rider (CAR) is eliminated, as proposed by PNM, for all rate classes except Rate 20. PNM South residential customers who use more than 450 kWh per month will automatically receive a rate increase upon elimination of the CAR, and would likely receive an even greater rate increase upon elimination of the CAR under the AG’s proposed volumetric rates. Mr. Aguirre said that a PNM South residential customer using 2,000 kWh per month would face a 59% bill increase in the summer months. PNM serves Recommended Decision Case No. 15-00261-UT 226 about 45,000 residential customers in its PNM South Service Territory, and about one-half of those customers have experienced use greater than 450 kWh/month. Aguirre Rebuttal 20-23; Exh. JCA-2 to Aguirre Rebuttal. The AG replied that PNM’s representation of the bill impact on PNM South residential customers is based on PNM “scal[ing] up the Attorney General’s proposed residential rates to meet PNM’s proposed revenue requirement.” The AG says that its proposed rate design is unique to its proposed revenue requirement. Under the AG’s proposed revenue requirement, the bill impact from the AG’s rate design is “on par” with the bill impact from PNM’s rate design. AG’s Posthearing Response Brief 7-8. Staff and the AG’s recommendation of a $7.00 residential customer charge is just and reasonable. It appropriately balances the interests of residential customers using 580 kWh or less per month with the conflicting interests of residential customers using more than 580 kWh per month. Designing the residential variable charges to achieve fairness to both PNM North and PNM South residential customers is difficult because the Hearing Examiner is recommending approval of PNM’s unopposed proposal to eliminate the Consolidation Adjustment Rider (CAR) for all PNM South classes except Streetlighting. Currently, for PNM South Residential customers, the CAR is a $0.0138612 per kWh charge for use up to 450 kWh. For the next 450 kWh of use, the CAR is a credit charge of ($0.0274738) per kWh, and for all other use, the CAR is a credit charge of ($0.0454779) per kWh. Therefore, under current rates, PNM South residential customers receive a credit on their bills for per kWh use over 450 kWh a month, which increases as PNM South residential customers use more energy per month. Simply eliminating the CAR rate without increasing any rates would increase a PNM South residential customer’s bill for use above 450 kWh. While the PRC endorses inclining block rates, increases to the second and third block rates for the Residential Class have to be tempered because of the Recommended Decision Case No. 15-00261-UT 227 disproportionately higher bill impact on PNM South residential customers who use over about 750 kWh per month. To best achieve the PRC’s directive in the 2010 PNM Rate Case to minimize bill impact on PNM North and South residential customers who use between 350 kWh and 1,000 kWh per month, the following inclining block rates should be adopted: Summer Rates Block 1 Block 2 Block 3 $0.0736225 Block 1 $0.1171582 Block 2 $0.1412434 Block 3 Non-Summer Rates $0.0736225 $0.1010912 $0.1149609 For an average-use PNM North residential customer using 585 kWh/month, these proposed rates would result in a seasonally-weighted 7.24% bill increase. This is roughly the same as the 8.4% revenue increase to the Residential Class as a whole. For an average-use PNM South residential customer using 531 kWh/month, these proposed rates would result in a seasonally-weighted (0.85%) bill decrease. However, the seasonally-weighted bill would increase from 1.04% at 600 kWh/month up to 8.42% at 1,000 kWh/month. For a PNM North customer, the percentage increase in bill decreases when use increases above about 585/kWh per month, which doesn’t send a desirable price signal. However, a rate design that would increase the percentage increase in a PNM North residential customer’s bill when use increases above 585/kWh per month would cause too much of a percentage increase in a PNM South residential customer’s bill for use at higher levels. See Attachment G. 4. SMALL POWER CLASS PNM proposes to increase the customer charge for the Rate 2A Small Power Class from $8.46 to $17.87. Staff, consistent with its recommendation for the residential customer charge, recommends a smaller increase than proposed by PNM for the Samll Power customer charge. Recommended Decision Case No. 15-00261-UT 228 Staff recommends increasing this customer charge from $8.46 to $11.00, a $2.54 or 30% increase. Leyba-Tercero Direct 9. PNM responds, consistently with its opposition to Staff and the AG’s proposed $2.00 increase in the Residential Class customer charge, that Staff’s proposed increase “does not go far enough to be considered gradual.” Chan Rebuttal 47. A just and reasonable Class 2A customer charge is $15.83. Combined with the Hearing Examiner’s proposed Class 2A variable charges, the percentage increase in an average-use Class 2A customer’s bill is about 6%. The increase in the customer charge, as a percentage of an average-use Class 2A customer’s total bill, is about 3%, which is about the same as the increase in the customer charge for a Class 1A customer, as a percentage of an average-use Class 1A customer’s total bill. See Attachment G. 5. IRRIGATION CLASS PNM proposes to increase the customer charge for the Rate 10A Irrigation Class from $8.19 to $30.03 per month, which would collect 50% of the customer-related costs for this Class. PNM’s proposal was unopposed, but is not appropriate under the Hearing Examiner’s recommended revenue requirement. A just and reasonable customer charge for the Rate 10A Irrigation Class is $23.80. Combined with the Hearing Examiner’s proposed Class 10A variable charges, the percentage increase in an average-use Class 10A customer’s bill is about 6%. See Attachment G. 6. WATER & SEWAGE CLASS PNM proposes to decrease the customer charge for the Rate 11B, Water and Sewage Time of Use Rate, from $491.60 to $327.75. This proposal was unopposed, but is not appropriate under the Hearing Examiner’s recommended revenue requirement and banding proposal. A just and reasonable customer charge for the Rate 11B Water and Sewage Class is $442.44. See Attachment G. Recommended Decision Case No. 15-00261-UT 229 D. DATA GATHERING PROPOSALS CCAE witness Howat’s recommendation that PNM be required to file within six months of issuance of a final order in this case, data to gauge the state of low-income and general residential home energy security in PNM’s service territory, see Howat Direct 13-19, should be rejected. So too should the AG’s suggestion that the PRC explore the impacts of PNM’s rates on low-income households. While gathering this data might be desirable, such a requirement should apply to all electric utilities. CCAE and the AG should make their requests through proposed rulemakings. In fact, the state commissions identified by Mr. Howat in his Direct Testimony that require the data collection that he proposes, have adopted the requirement by regulation. Id. E. NON –RESIDENTIAL THREE-TIER RATES 1. LOAD FACTOR AND DEMAND CHARGES IN GENERAL The amount of power required to operate electric plant is measured in terms of kilowatt (kW) demand. Peak demands are important to a utility because they dictate the maximum amount of power a utility must generate or buy to meet the needs of its customers. There are daily, seasonal, and annual peak loads. Load factor is an expression of how much energy was used in a time period versus how much energy would have been used if the power had been left on during a period of peak demand. If all electric loads are turned on fully and never turned off, the load factor is 1.0 or 100%. A higher/“good” load factor (a higher percentage) implies a more constant rate of electrical use because kW/demand is held to a minimum relative to overall use. The demand charge rate structure rewards customers for improving their load factor. The idea behind a demand charge rate structure is that customers who create higher demand are charged more for it. Those who do not create peaks but maintain a relatively level demand are Recommended Decision Case No. 15-00261-UT 230 known to have a high load factor. Pricing policies are designed to pass savings on to these customers. High load factor customers use their maximum kW demand more efficiently by using more kWh per kW. A customer whose use increases the system load factor allows the overall cost of purchased power to be less. Proposed Tariff Filing by City of Newberry, Docket No. 910354-EM, 1991 WL 11686799 (Fla. P.S.C. 8-21-91). When, for a particular class, demand charges do not recover all of the Class’ demandrelated costs and variable charges recover more than the Class’ energy-related costs, higher load-factor customers subsidize lower load-factor customers. Townsend Direct 6. A customer with a low load factor pays less than the costs incurred, while a customer with a high load factor pays more than the costs incurred. Therefore, within a single class, customers who are smaller than average or have lower load factors than average are subsidized by the other customers in the class. This occurs because they do not use enough kWhs to collect all the demand costs they impose on the system. Impact of Electric Rate Billing Changes on Agriculture, 90 P.U.R.4th 1, 1988 WL 391348, *5 (Conn. Dep’t PUC 1-6-88). One of the simplest ways of improving load factor is to “shave the peaks.” “Shaving” means having a portion of the electrical load operating at peak times of the day shifted to nonpeak times. For example, instead of operating 10 machines at 11:00 a.m., it would be advantageous to have three operating at 9:00 a.m., five at 10:00 a.m., and two at 11:30 p.m. 2. PNM’S PROPOSAL IN GENERAL PNM proposes to modify its demand charges for all customer classes under a three-part tariff 54 to move demand rates closer to or at the full cost of service level. Ortiz Direct 31. The seven rate classes with a demand charge are: x x 54 Rate 3B Rate 3C General Power General Power — Low Load Factor A three-part tariff is comprised of a customer, demand and variable charge. Chan Direct 57 n.26. Recommended Decision Case No. 15-00261-UT 231 x x x x x Rate 4B Rate 5B Rate 15B Rate 30B Rate 33B Large Power Large Service >=8,000 kW Universities Manufacturing Station Power Reynolds Direct 5-6. Exh. JCA-5 provides summary of PNM’s current and proposed demand charges: PNM’s redesign achieves more cost-based demand rates by: 1. No longer recovering any demand-related costs through the customer charge, and only collecting customer-related costs through the customer charge Recommended Decision Case No. 15-00261-UT 232 2. Adjusting the demand rates for each class to more closely reflect each Class’ capacity-related costs according to PNM’s ECCOSS. Reynolds Direct 4-5. PNM proposes these changes to send a more accurate price signal to customers of the demand-related costs necessary to meet the peak demands of the affected rate classes. Aguirre Direct 60. While PNM’s goal is to move to a demand charge that fully reflects all of the Company’s capacity-related costs and send a more accurate price signal, PNM says there are reasons why it should not design a demand charge that collects all capacity-related costs for all customer classes. PNM says it must also consider: (1) the rate impacts for customers within each rate class with different load factors; and (2) preserving the underlying integrity of PNM’s existing rate schedules by preventing unintended customer migration. Aguirre Direct 11, 30. PNM limits its proposed demand rates to the lower of the cost-based level or the cost level allocated to each rate class after banding. This means that for classes receiving a subsidy through the banding process, PNM does not propose a demand charge higher than what is indicated after the banding is applied. For the rate classes not receiving a subsidy, PNM caps the demand charges at the cost-based level, resulting in true cost-based demand rates for those rate classes. Aguirre Direct 12. The following table shows the percentage of demand-related costs for each Class that would be recovered through each Class’ demand charges under PNM’s proposed rates: Rate No. 3B 3C 4B 5B 15B 30B 33B Recommended Decision Case No. 15-00261-UT Summer 89% 30% 85% 100% 80% 80% 100% 233 Non-Summer 25% 100% 88% Reynolds Direct 10. Staff’s supports PNM’s proposed redesign of these Customer Classes. Staff supports moving toward more cost-based recovery of demand-related costs, as tempered by gradualism, and explains that collecting all demand-related costs through the kW charge “sends a better price signal and incentive to its customers to increase their load factor.” Reynolds Direct 11. PNM in Rebuttal Testimony agreed to add clarifying language to the tariffs regarding minimum demand as suggested by Staff. See Reynolds Direct 12; Aguirre Rebuttal 55-56. The City/County oppose increasing demand charges for General Power 3B/3C customers. They argue that PNM’s true reason for increasing demand charges is to recover revenue that it losing because of declining energy sales. Ankum Direct 44-47. 3. INTERVENORS SERVED UNDER THREE-PART TARIFFS WITH DEMAND CHARGES NMIEC’s members are largely served under Rates 4B, 15B, 30, and would be served under new proposed Rate 35B if approved. Phillips Direct 40. Kroger receives service under PNM Rates 3B, 4B and 2A. About 92% of Kroger’s load is served under Rate 3B, 7% under Rate 4B, and 1% under Rate 2A. Kroger Exh. 3. Walmart receives service under PNM Rates 2A, 3B, 3C, and 4B. About 87% of Walmart’s load is served under Rate 4B, 10% under Rate 3B, and the remaining under Rates 2A and 2C. Tr. (4-27-16) 3136 (Chriss). ABCWUA receives service under several PNM rates. ABCWUA witness Herz focused his Direct Testimony on Rates 4B and 11B. Herz Direct 3-4. The City/County receive service under several PNM rates. City/County witness Ankum focused his Direct Testimony on Rates 3B/3C, and 20 (Streetlighting). Ankum Direct 13, 21. 4. RATE 3B/3C — GENERAL POWER PNM refers to “Rate 3B/3C” as a single “Rate” with two “Schedules:” 3B and 3C. Aguirre Rebuttal 32. These Schedules comprise the General Power Service TOU Class, which generally Recommended Decision Case No. 15-00261-UT 234 applies to customers with on-peak demand of at least 50 kW or monthly energy consumption of at least 15,000 kWh. Schedule 3B is PNM’s “General Power Service TOU Rate.” Schedule 3C is PNM’s “General Power Service (Low Load Factor) TOU Rate.” The primary difference between Schedule 3B and 3C customers is not their load, but their load factor. Kroger’s Initial Brief 1-2. In the 2007 PNM Rate Case, the PRC approved PNM’s proposal to replace the thenexisting demand-based three-block On-Peak Rate structure under Class 3B with a single energy rate for all On-Peak Energy. Kroger supported eliminating the three-block rate but argued that the 3B demand charge should recover all demand-related costs for that Class, plus a proportionate share of any inter-class subsidy assigned to that Class. Kroger proposed this change because PNM’s redesign of Rate 3B resulted in higher load factor customers, such as Kroger, being allocated a greater percent of the rate increase. Final Order, ¶¶ 100-03. The PRC agreed that PNM’s redesign would result in proportionately larger rate increases for higher load-factor customers than for lower load-factor customers. It explained, “Any rate design that recovers some fixed costs in a rate’s variable charge will very likely cause a proportionately greater increase in the rates of higher-load factor customers than lower-load factor customers.” However, the PRC found that a rate design that results in a higher rate increase to high-load factors customers can be reasonable “if the rates reasonably reflect the costs of providing service to all customers of a class, and achieve other Commission policy goals such as avoidance of rate shock or the encouragement of conservation.” Id., ¶ 104-05. While rejecting Kroger’s proposal, the PRC said that Kroger raised significant issues that should be addressed in PNM’s next rate case. The PRC ordered: PNM should include in its next general rate case an alternative to Rate 3B that has a demand and variable charge that are more proportional to the fixed and variable charges that are incurred to provide that service and that are consistent with the rates of PNM’s other rate classes. Id., ¶ 106. Recommended Decision Case No. 15-00261-UT 235 In response, in its 2008 Rate Case, PNM proposed, and the PRC approved, Schedule 3C, which is a “low load factor carve-out of the Schedule 3B customers.” Schedule 3B is designed for customers with load factors higher than 35%, and Schedule 3C is designed for customers with load factors of 35% or lower. Schedule 3B has higher demand charges and lower variable charges than Rate 3C. Townsend Direct 2. Schedule 3C was designed for many small businesses that operate 8 to 10 hours daily during the work week, whereas Schedule 3B was designed for businesses that operate longer hours with multiple shifts, 6 to 7 days per week. Aguirre Rebuttal 37-38. In this case, Kroger witness Townsend said that creating Schedule 3C was only a first step and the next step should be “to allow the remaining customers on 3B to have rates that closely align with costs,” which is the intent of Kroger’s proposal in this case and is discussed below. Tr. (4-28-16) 3180 (Townsend). PNM proposes the same customer charges for Schedules 3B and 3C, but proposes different demand and variable charges for the two Schedules. PNM did not separately incorporate Schedules 3B and 3C into its ECCOSS because of sampling limits. Aguirre Rebuttal 32. Under PNM’s ECCOSS, using the incorrect apples-to-oranges comparison, Schedules 3B and 3C, collectively, would receive a 14% non-fuel revenue increase. Chan Direct 45. PNM’s ECCOSS shows that Schedules 3B and 3C, collectively, would receive a non-fuel revenue increase of only 2.29% if PNM collected its alleged revenue deficiency from each Class based on each Class’ cost responsibility. Exh. SC-9, p.2, row 14, column F, to Chan Direct. Therefore, PNM’s proposed revenue allocation results in Schedules 3B/3C subsidizing other Classes by $22.7 million. Townsend Direct 3-4. The City/County point out that peak demand by General Power customers has decreased significantly since 2010, so these customers are not driving the need for capital investments. Dr. Ankum said that while he is “not saying that cross- Recommended Decision Case No. 15-00261-UT 236 subsidization of residential rate payers is inappropriate,” it is bad public policy for the City/County to shoulder this burden. Ankum Direct 42-43. Kroger alleges that there is a significant intraclass subsidy between Schedules 3B and 3C. While PNM did not separately incorporate the two Schedules into its ECCOSS, Kroger witness Townsend allocated costs between the Schedules by allocating the combined Schedules’ revenue requirement by category (customer-related, demand-related, energy-related) based on each Schedule’s billing determinants for each category. For example, Mr. Townsend allocated demand-related costs for the combined Schedules based on each Schedule’s billable kW. Kroger argues that, without considering interclass subsidies, PNM’s proposed Schedule 3C rates would under-recover the costs attributable to Schedule 3C by $3.6 million and PNM’s proposed Schedule 3B rates would over-recover the costs attributable to Schedule 3B by $2.9 million. Additionally, Kroger asserts that Schedule 3B is paying the entire $22.7 million interclass subsidy borne collectively by Rate 3B/3C. Therefore, Kroger asserts that under PNM’s proposed rates, Schedule 3B would pay $26.3 million more than its own cost of service. Townsend Direct 3-4. Kroger does not object to the total revenue allocation to Rate 3B/3C, but it recommends one change to mitigate the alleged intraclass subsidy between Schedules 3B and 3C. Kroger recommends increasing the revenues collected fr0m Schedule 3C and decreasing the revenues collected from Schedule 3B, but capping the Schedule 3C revenue increase at twice the system average revenue increase, so that each Schedule’s rates more closely reflect each Schedule’s cost of service. Under Kroger’s proposal and PNM’s proposed revenue requirement, Schedule 3C would receive a 28.37% non-fuel revenue increase. To completely cover its own cost of service, Schedule 3C would have to receive a 28.98% non-fuel revenue increase. Revenues from Schedule 3B would still recover the entire $22.7 million interclass subsidy. Kroger recommends decreasing the Schedule 3B per kWh charges and increasing the Schedule 3C per kWh charges Recommended Decision Case No. 15-00261-UT 237 to the extent necessary to achieve this result. If the PRC decides to maintain the Schedule 3C subsidy in an amount greater than the amount recommended by Kroger, Kroger argues this subsidy should not be borne by only Schedule 3B customers. Townsend Direct 5. PNM opposes Kroger’s proposal to increase the revenues collected from Schedule 3C and decrease the revenues collected from Schedule 3B because it would be too great of a Schedule 3C rate increase. PNM acknowledges a potential subsidy from Schedule 3B to Schedule 3C, but says that Kroger’s calculations of the separate costs to serve each Schedule do not necessarily result in an equalized rate of return because PNM’s ECCOSS does not separately determine costs of service for Rates 3B and 3C because of sampling limitations for these Classes. Mr. Aguirre said that billing determinants only represent the units served for purposes of cost causation and do not reflect cost causation. PNM’s Initial Response Brief 123. In his Rebuttal Testimony filed on February 22, 2016, Mr. Aguirre said that PNM had “already taken steps to establish a separate load research sample for Rate 3B and Rate 3C in order to determine in a future rate case the cost to serve each rate schedule separately and potentially identify and address any intra-class subsidization issues that may arise.” He said that PNM would have to collect “a full year’s worth of data to be collected before a cost basis can be established for the two general power schedules.” Aguirre Rebuttal 34. It is not clear how PNM developed separate cost-based rates for Schedules 3B and 3C without independently evaluating each Schedule in its ECCOSS. However, because no party or Staff has challenged PNM’s proposed 3B and 3C Rate Schedules on that specific ground, PNM’s proposed Schedules should be approved, but such approval should not be viewed as precedent. Kroger’s proposal to increase the revenues collected fr0m Schedule 3C and decrease the revenues collected from Schedule 3B, but cap the Schedule 3C revenue increase should be denied. In its next base rate case filing, PNM shall base its proposed 3B and 3C Rate Schedules Recommended Decision Case No. 15-00261-UT 238 on separate cost of service studies for each Schedule and shall propose a rate design that moves toward eliminating any intraclass subsidy, consistent with gradualism. Kroger and Walmart both recommend that Schedule 3B’s secondary demand charge be increased to collect 100% of demand-related costs and that Schedule 3B’s variable charges be decreased proportionally. This would result in recovering the entire subsidy borne by Schedule 3B customers through variable charges. The purpose of this change is to reduce the subsidy within Schedule 3B flowing from higher-load factor customers to lower-load factor customers that is a result of not collecting all demand-related costs through the demand charge. Townsend Direct 6-7; Chriss Direct 20. Kroger and Walmart’s recommendation to increase Schedule 3B’s secondary demand charge to collect 100% of demand-related costs and decrease Schedule 3B’s variable charges proportionally should be approved. 5. RATE 4B — LARGE POWER Rate 4B is PNM’s Large Power Service TOU Rate, applicable to customers with a 500 kW minimum demand. Currently, the Rate 4B demand charges collect about 52% of that Class’ demand-related costs. PNM’s proposed rate 4B demand charges would collect 85% of that Class’ demandrelated costs. Aguirre Rebuttal 37. The following table compares PNM’s existing Rate 4B rates with its proposed Rate 4B rates as modified to collect all fuel costs through the FPPCAC: Rate 4B - Summer Current Proposed Customer $8,735 $605.13 Charge On-Peak $0.0704373 $0.0343509 kWh Off-Peak $0.0365815 $0.0160027 kWh Recommended Decision Case No. 15-00261-UT 239 Rate 4B – Non-Summer Current Proposed Customer $7,100 $605.13 Charge On-Peak $0.0553112 $0.0231511 kWh Off-Peak $0.00365815 $0.0160027 kWh Demand Charge Demand Included in Customer Charge FPPCAC $17.47 $28.56 500 0 $0.0090100 $0.0216856 Demand Charge Demand Included in Customer Charge FPPCAC $14.2 $20.45 0 0 $0.0090100 $0.0216856 Exh. JCA-11 to Aguirre 3-14-16 Supp. Walmart and NMIEC recommend that the Rate 4B demand charges collect 100% of that Class’ demand-related costs. Chriss Direct 6; Phillips Direct 42. This would result in a $33.60/kW charge for the summer months and a $24.06/kW charge for the non-summer months. Chriss Direct 19-20. Walmart argues that recovering demand-related costs through variable charges shifts demand cost responsibility from lower load-factor customers to higher load-factor customers who use PNM facilities more efficiently. Walmart witness Chriss explained: In essence, two customers can have the same level of coincident demand and cause the utility to incur the same amount of fixed cost, but because one customer uses more kWh than the other, that customer will pay more of the demand cost than the customer that uses fewer kWh. This results in misallocation of cost responsibility as higher load factor customers overpay for the demand-related costs incurred by the Company to serve them and are essentially penalized for more efficiently using the Company’s system. Chriss Direct 19. On the other hand, ABCWUA recommends that the Rate 4B demand charges collect “something less than the 85% proposed by the Company.” Herz Direct 10. Under PNM’s proposal, the average revenue increase for Rate 4B customers would be 14.17%. The average revenue increase for ABCWUA’s accounts served under Rate 4B, however, would be 44.9%.55 This is because the ABCWUA accounts have an average load factor of 14%, while the Class 4B average load factor is about 70%. Herz Direct 9; Aguirre Rebuttal 37. These percentages are taken from page 8 of ABCWUA witness Herz’s Direct Testimony and are based on PNM’s incorrect comparison of the percentage revenue increase. 55 Recommended Decision Case No. 15-00261-UT 240 PNM includes 220 Rate 4B customers in its Test Period. Exh. SC-3, p.7 to Chan Direct. About ten of these 220 customers have load factors of 35% or lower; only 2 of the 220 customers have load factors of 18% or lower. Aguirre Rebuttal 38. PNM opposes both competing recommendations. It correctly observes that setting demand rates to collect 100% of demand-related costs would have a more significant impact on customers with low load factors, while setting demand rates to collect less than 85% of demandrelated costs would contribute to intra-class subsidization. Aguirre Rebuttal 36. PNM’s proposal to collect 85% of demand-related costs through the demand charge should be approved because it recognizes the diversity of load factors in the General Power class and strikes a balance between the competing interests. The low percentage of customers within Rate 4B with load factors of 35% or lower (4.5%) does not justify creation at this time of a separate low load-factor schedule within Rate 3B. Aguirre Rebuttal 38. 6. RATE 5B — LARGE SERVICE Rate 5B is PNM’s Large Service >=8,000 kW Rate. PNM proposes to set the demand charges to collect 100% of demand-related costs. Aguirre Direct 32. This proposal is unopposed and should be adopted. 7. RATE 15B — UNIVERSITIES Rate 15B is PNM’s Large Service for Public Universities >=8,000 kW Rate. PNM proposes to set the summer demand charges to collect 80% of demand-related costs and the non-summer demand charges to collect 100% of demand-related costs. NMIEC proposes setting the demand rates to collect 100% of demand-related costs. Phillips Direct 42. While PNM agrees in principle with NMIEC’s proposal, Mr. Aguirre said that it could result in unintended consequences. Setting demand rates to recover 100% of the demand- Recommended Decision Case No. 15-00261-UT 241 related costs could cause disproportional seasonal impacts for Rate 15B and 30B and the overstatement of demand-related costs for Rate 15B. Aguirre Rebuttal 41. PNM proposes a more reasonable gradual increase to demand rates and its proposal should be adopted. 8. RATE 30B — MANUFACTURING Rate 30B is PNM’s Large Service for Manufacturing >=30,000 kW Rate. PNM proposes to set the summer demand charges to collect 80% of demand-related costs and the non-summer demand charges to collect 88% of demand-related costs. This proposal avoids a disproportional increase in summer bills for this Class. Aguirre Direct 33. NMIEC proposes setting the demand rates to collect 100% of demand-related costs, consistent with its proposal for Rate 30B. Phillips Direct 42. NMIEC’s proposal should be rejected for the same reason it is rejected for Rate 30B. 9. RATE 33B — STATION POWER Rate 33B is PNM’s Large Service for Station Power TOU >=30,000 kW Rate. PNM proposes to set the demand charges for this Class to collect 100% of demand-related costs. Aguirre Direct 34. This proposal is unopposed and should be adopted. F. PROPOSED RATE 35B: NEW HIGHER LOAD FACTOR LARGE CUSTOMER CLASS PNM propose to add a new tariff — Rate 35B — to provide service to “Very Large” customers with a minimum monthly demand of 3,000 kW and a minimum 75% load factor.56 PNM estimates that five of its existing Rate 4B customers would be eligible for service under Rate 35B. Dauphinais Direct 36. Rate 4B is PNM’s Large Service Time-of-Use Rate. Its eligibility requirement is a minimum 500 kW demand. PNM proposes Rate 35B for its few customers with sufficiently 56 In its Direct Testimony, PNM proposed a minimum 80% load factor. In Rebuttal Testimony, PNM agreed to a 75% minimum load factor as recommended by NMIEC witness Dauphinais. Chan Rebuttal 52. Recommended Decision Case No. 15-00261-UT 242 higher demands and capacity factors to justify their own rate. PNM’s cost to serve these customers is less because: 1. There are economies of scale associated with serving customers with larger demands 2. They take service at a higher, primary substation level 3. They consume a larger share of their electricity during lower-cost off-peak time periods. Dauphinais Direct 34-35. PNM initially proposed demand rates to recover about 96% of the allocated demandrelated costs for Rate 35B. Chan Direct 57. In Rebuttal Testimony, PNM agreed with NMIEC’s recommendation to set demand rates at 100% of the cost-based level as long as the proposed rate does not exceed the cost level set after banding. Aguirre Rebuttal 41; PNM’s Initial Brief 291-92. Staff concluded that PNM provided use data that “reasonably suggests that the migrating customers have a distinct set of characteristics to form a separate rate class.” Staff supports creation of Rate 35B. Reynolds Direct 17. PNM’s proposed Rate 35B, in the form of Exh. SC-2 to Ms. Chan’s Rebuttal Testimony, should be approved. This version of proposed Rate 35B contains clarifying language regarding minimum demand as suggested by Staff. See Reynolds Direct 12, 18; Aguirre Rebuttal 55-56. G. TIME OF USE RATES The intent of time of use (TOU) rates is to set rates based on the cost to serve during a particular time period and provide an economic incentive to customers to shift their use from peak periods — periods that require use of a significant portion of a utility’s capacity. PNM’s definition of the TOU Period has not changed since 1986. PNM says that its TOU rates have not been cost-based in many years, and as such, the current price differentials for these alternative rate schedules are misaligned with the actual costs that the Company incurs. Aguirre Rebuttal 6-7. Recommended Decision Case No. 15-00261-UT 243 Currently, only 128 of PNM’s 453,396 residential customers, or 0.028%, participate in PNM’s TOU rate for the Residential Class — Rate 1B. For residential customers, the average TOU national participation rate is 1.9%. In New Mexico, Central New Mexico Electric Cooperative’s and Sierra Electric Cooperative’s residential TOU participation rates are 12.3% and 9.2%, respectively. Only 523 of PNM’s Small Power customers, or 1.02%, participate in PNM’s TOU rate for the Small Power Class — Rate 2B. For commercial customers, the average TOU national participation rate is 12.4%. Howe Direct 17-18. PNM proposes to change its TOU on-peak hours from 8 a.m. to 8 p.m., Monday through Friday (Current TOU Period) to 10 a.m. to 10 p.m., Monday through Friday (Proposed TOU Period). According to PNM witness Aguirre, the Proposed TOU Period more accurately reflects the period in which PNM’s generation and delivery costs are higher. Since at least 2007, PNM’s monthly system CPs frequently have occurred in current off-peak hours later in the day, particularly in the non-summer months. Based on an eight-year average of PNM’s actual system peaks, (i) the probability of PNM’s system peak occurring during the Current TOU Period is 61%; (ii) the probability of PNM’s system peak occurring during the Proposed TOU Period is 70%. Aguirre Direct 21 & Exh. JCA-8 to Aguirre Direct. PNM also proposes to change its TOU rates to more accurately reflect PNM’s variations in energy costs. PNM based its proposed rate variances among seasonal TOU periods on the differences observed for PNM’s hourly energy costs using historical data. Specifically, PNM used system lambdas to determine these rate variances, which is a marginal cost approach that represents the higher of (1) the cost of PNM’s highest-price energy purchase made in the wholesale generation market in a given hour; or (2) the cost of PNM’s highest-priced generation resource dispatched in a given hour. PNM’s analysis showed that for its system as a whole, average summer on-peak energy costs are 25% higher than non-summer on-peak energy costs and 50% higher than off-peak energy costs. PNM proposes to set the volumetric summer on- Recommended Decision Case No. 15-00261-UT 244 peak rates about 25% higher than non-summer on-peak rates and about 50% higher than offpeak rates. Aguirre Rebuttal 24-25. WRA looked at PNM’s TOU proposal in detail and submitted compelling evidence that PNM’s proposal would be ineffective in reducing on-peak energy use by residential and small power customers. For example, under PNM’s proposal, even if a residential customer could shift 100% of its use off-peak, that customer would need to use over about 1,360 kWh per month to be better off under PNM’s proposed residential TOU rate. Tr. (4-22-16) 2433 (Monroy). Dr. Howe particularly identified five problems that he sees with PNM’s TOU proposal. Howe Direct 20. When asked whether it would be acceptable to address his concerns with PNM’s TOU rate design proposal in PNM’s next rate case, Dr. Howe said, “Partially.” If the PRC were to defer consideration of PNM’s TOU rate design, Dr. Howe recommended that the PRC appoint a facilitator to convene PNM and the interested Intervenors in this case to discuss how to design effective TOU rates, before PNM files its next rate case. Tr. (4-25-16) 2674. Relying on PNM’s allegedly flawed TOU rate design, WRA and CCAE urge the PRC to completely reject PNM’s proposed rate designs for its Residential and Small Power Classes and, in the interim, set rates for these Classes by prorating the approved revenue requirement to each of the existing rate components. Howe Direct 16-17; CCAE’s Initial Posthearing Brief 55. ABCWUA argues that the PRC should defer changing PNM’s TOU Period until more analysis is performed. Mr. Herz suggests that such analysis better demonstrate a correlation between PNM’s increased costs and its proposed TOU Period and be presented by season. Herz Direct 11-14. NEE argues for rejection of PNM’s change in the TOU Period because PNM didn’t show that the shift would benefit ratepayers. NEE’s Initial Posthearing Brief 40-41. CFRE agrees. CFRE’s Posthearing Response Brief 18. Recommended Decision Case No. 15-00261-UT 245 The City/County argues that PNM’s proposed TOU rates are badly designed and should be redesigned. City/County’s Initial Posthearing Brief 17. In its Posthearing Response Brief, PNM said that it “understands the intervenor concerns related to peak usage on PNM’s system” and is willing to maintain its existing TOU rate structure while it studies a modified TOU pricing structure. PNM says that “[b]y agreeing to study a modified TOU pricing structure, the only caveat is that PNM requests that the Commission mandate that TOU pricing be cost based and not solely based on an arbitrary noncost based principle.” PNM’s Posthearing Response Brief 137-38. It is not clear whether PNM is agreeing to maintain both its existing TOU Period and its TOU pricing structure, or just its TOU pricing structure. Developing an effective TOU rate design is important and requires significant study. The evidence suggests that PNM’s current TOU rate design is not effective and that its proposed TOU rate design would not be any more effective. PNM’s existing TOU rate design — including its existing TOU Period and TOU pricing structure — should not be changed at this time. In its final order in this case, the PRC should appoint a facilitator to meet with PNM and interested parties to discuss proposals for an effective TOU rate design, and set a date for the first meeting. This recommendation is consistent with PRC’s directive in the 2015 EPE Rate Case for EPE to develop proposals to encourage greater participation in its TOU rates. Final Order 89, ¶ 214. The PRC rejected Intervenor proposals to change the on-peak hours and months in EPE’s Residential Time-of-Use Rate because it would result in even higher variable charges than the variable charges resulting from the Hearing Examiner’s recommendation. Id., Final Order 59, ¶¶ 131. The PRC agreed with the Hearing Examiner that subscribership to the Residential TOU Rate “is far too low and should be rectified.” It ordered EPE to investigate reasonable changes to increase participation and reduce peak demand and to propose those changes in its next rate case. Id., Final Order 59-60, ¶¶ 131-33. Recommended Decision Case No. 15-00261-UT 246 WRA and CCAE’s recommendation to completely reject PNM’s proposed rate designs for its Residential and Small Power Classes at this time should not be adopted. H. STREETLIGHTING TARIFF In the 2010 PNM Rate Case, PNM agreed to “enter into discussions with customers of Rate 20 on cost allocation, rate design, maintenance, re-lamping and energy efficiency.” Amended Stipulation, ¶ 28. PNM had such discussions and, as a result, proposes to redesign its Streetlighting Tariff — Rate 20. 1. PROPOSED C ONTINUATION OF CONSOLIDATION ADJUSTMENT RIDER The 2010 PNM Rate Case consolidated rates for PNM North and PNM South customers, except customers taking service under Rate 20. As such, PNM North customers pay different prices for lights and poles than PNM South customers. Also, for PNM North customers, Rate 20 has separate light and pole rates. For PNM South customers, Rate 20 has a bundled rate for lights and poles. PNM proposes to single set of rates under Rate 20 to apply to both PNM North and South customers. To mitigate extreme impacts to PNM South customers, PNM proposes to continue to apply the Consolidation Adjustment Rider (CAR) — Rate Rider No. 35 — to PNM South Rate 20 customers. Aguirre Direct 42. The CAR would limit the Rate 20 increase to 14.1% (under PNM’s apples to oranges comparison of revenues). Aguirre Direct 49. 2. SIMPLIFY TARIFF PNM proposes to simplify Rate 20 in several ways. First, PNM proposes to eliminate overhead and underground categories. Second, PNM proposes to eliminate two lighting options that are no longer installed in the field: 1. 2. 250W Mercury Vapor Underpass Light 150W High-Pressure Sodium Streetlight Third, PNM proposes to combine two 400W High-Pressure Sodium lighting options into one lighting option. Recommended Decision Case No. 15-00261-UT 247 Fourth, PNM proposes to correct and standardize Tariff language. Aguirre Direct 44-45. 3. INCREASED CUSTOMER CHOICE Several customers asked PNM to offer a high-efficiency lighting option. In response, PNM proposes to offer the following PNM-owned lighting options: x 39W LED Light x 118W LED Light x 257W LED Light Additionally, PNM proposes to add a new section to the Tariff applicable to customerowned and maintained lighting that is not specific to any light type and therefore “freely permits high-efficiency lighting installations by the customer.” Aguirre Direct 46. 4. MAINTENANCE FOR CUSTOMER-OWNED LIGHTS Currently, some Rate 20 customers who own their own lighting have to pay a fee for PNM maintenance even if they want to do their own maintenance. PNM’s proposes to eliminate this restriction by allowing customers to separately contract for PNM maintenance of customerowned lights. Aguirre Direct 47. 5. POSITIONS OF STAFF AND INTERVENORS Staff supports PNM’s proposed revisions to Rate 20. Reynolds Direct 20. In his Direct Testimony, CCAE witness Bickford recommended that Rate 20 be revised to: x x x x Account for the salvage value of replacement streetlights Include a wider range of streetlamps Offer streetlamps with a longer expected life span Offer a range of operational characteristics In his Rebuttal Testimony and at the hearing, Mr. Bickford said that, after he filed his Direct Testimony, he recognized that PNM does not have technical and cost information necessary to change its proposed Tariff in this case. To effectuate development of this data, Mr. Bickford recommended that the PRC order PNM to “convene a stakeholder group consisting of Recommended Decision Case No. 15-00261-UT 248 the intervenors in this case plus any interested municipalities, to investigate the opportunities and costs of replacing existing streetlights with LED streetlights.” Bickford Rebuttal 6. Mr. Bickford recommended that PNM’s proposed changes to Rate 20 be approved. Tr. (4-21-16) 2082. At the hearing, PNM witness Aguirre said that PNM agrees to meet with stakeholders to investigate issues related to LED replacement and to begin collecting data necessary to necessary to evaluate or propose Mr. Bickford’s recommendations. Tr. (4-22-16) 2494-97. In its Initial Posthearing Brief, CCAE makes a more specific request that request the Commission order PNM to convene stakeholder meetings (inviting intervenors to this case and other interested stakeholders, such as municipalities) to discuss at a minimum the following issues for inclusion in PNM’s next rate case: conversion of high pressure sodium lighting to LED lighting, including a) 100,000 hour lights; b) metering and advanced lighting control options at the request of the customer; c) salvage values and recovery of undepreciated assets; d) expanded lighting range options for conversion; and e) installation allowances. CCAE also requests that PNM be directed to develop the cost and technical data necessary to develop a tariff that includes these items. CCAE’s Initial Posthearing Brief 32. In its Response Brief, PNM says that it welcomes productive stakeholder discussions related to LED lighting options, but does not believe that the PRC needs to order it to do so because PNM has demonstrated its willingness to work through as many issues as possible. PNM’s Response Brief 140-41. The evidence does show that PNM has worked with the City of Albuquerque to address the City’s concerns with Rate 20. Nevertheless, PNM should be directed to do what is requested by CCAE to ensure that CCAE’s concerns are addressed. In his Direct Testimony, Mr. Bickford recommended that language in Rate 20 be revised to make clear that installed streetlights can be converted to streetlights. Bickford Direct 7. In his Rebuttal Testimony, Mr. Aguirre said that PNM’s proposed Rate 20 does allow for Recommended Decision Case No. 15-00261-UT 249 conversion of streetlights. Aguirre Rebuttal 44. At the hearing, CCAE’s attorney asked Mr. Aguirre if he objected to revising language in the “Special Conditions” section of Rate 20 on page 5, subsection 1(a), to read: Upon request from the Customer, the Company shall convert or install Company owned street lighting fixtures at its own expense . . . . Mr. Aguirre responded that he “didn’t see a problem with doing that” but wanted to review the change to ensure it would not create any inconsistency. Tr. (4-22-16) 2492-93. In its Initial Posthearing Brief, CCAE asks the PRC to order PNM to revise the language in the “Special Conditions” section of Rate 20 as it proposed to Mr. Aguirre at the hearing. CCAE also asks the PRC to order PNM to revise the “Special Conditions” section of Rate 20 on page 6, subsection III, to read “Relocations, conversions and changes, other than normal operation . . . .” CCAE’s Initial Posthearing Brief 31-32. PNM did not respond to these suggested revisions in its Response Brief. PNM should be ordered to make these suggested revisions to Rate 20. Tony Gurule filed Rebuttal Testimony on behalf of the City/County responding to Mr. Bickford’s Direct Testimony. Mr. Gurule agreed “for the most part” with Mr. Bickford’s Testimony. Mr. Gurule said that Mr. Bickford’s testimony focuses on the conversion of PNMlighting to LED. Mr. Gurule said a purpose of his Rebuttal Testimony was to discuss conversion of City-owned lighting to LED. Gurule Rebuttal 2. In its Posthearing Response Brief, the City/County say, “The Commission should approve PNM’s proposed Rate 20 with its proposed volumetric rate for customer-owned LED Streetlighting.” City/County’s Posthearing Response Brief 5. It is not clear whether the City recommends any other PRC action relating to PNM’s Rate 20. In its Posthearing Brief, the City/County, citing to page 309 of PNM’s Initial Posthearing Brief, asserts that “PNM attempts to raise false issues, possibly to delay implementation [of the City’s LED Streetlighting program].” City/County’s Posthearing Response Brief 5. Page 309 of PNM’s Initial Posthearing Recommended Decision Case No. 15-00261-UT 250 Brief responds to what PNM calls the City’s proposal “that PNM adopt certain, specific types of photocell controls which could function as ‘smart’ meters that would be installed on each individual LED streetlight.” PNM suggests that the City’s prop0sal might require revision of PRC Rules. In its Posthearing Response Brief, the City/County deny that there are any regulatory gaps and say, “The Commission will of course initiate rule-making as warranted.” City/County’s Posthearing Response Brief 5. It appears that the City/County and PNM are continuing productive discussions relating LED streetlights, and that the issue alluded to is not ripe for review. PNM’s Initial Posthearing Brief 308-09 PNM’s proposed revisions to Rate 20, as revised by CCAE’s recommendations, should be approved, and PNM should be ordered to convene stakeholder meetings immediately to discuss the matters identified by CCAE. I. BILL COMPARISONS Comparisons between current bills under existing rates and projected bills under the Hearing Examiner’s proposed rates are in Attachments G (Non-Capped, Non-Exempt Customers), H (Exempt Customers), and I (Large Capped Customers). J. CITY/COUNTY’S PROPOSAL FOR SEPARATE RATE CLASSES OR COST-BASED RATES UNDER EXISTING RATE CLASSES City/County witness Ankum filed Direct Testimony in which he said that his “overall recommendation” is that the PRC require PNM to create separate rate classes for the City and County for Streetlighting and General Power 3B/3C with “cost based rates that are free of support for other rate classes, ratepayers or services.” He referred to these proposed rates as “special economic development rates for Streetlighting and General Power[.]” He made other recommendations that “complement” his overall recommendation. Ankum Direct 13-14. Dr. Ankum’s additional recommendations are discussed in other Sections of this Recommended Decision. Recommended Decision Case No. 15-00261-UT 251 Dr. Ankum’s rationale for creating separate rate classes for the City/County is his argument that the services provided by the City/County promote economic development and job growth and therefore the City/County should not be burdened with subsidizing other PNM ratepayers. Id. at 15-21. He summed up his recommendation as follows: Specifically, I am recommending that the Commission recognize the City’s and County’s roles, and those of other similar governmental entities, as engines of economic growth and ensure that their electric services are provided at cost-based rates that are not artificially inflated to support other rate payers. In short, I recommend that the City and County pay their own freight — but no more than that. Id. at 21 (footnote omitted & emphasis in original). The PUA prohibits PNM from offering unreasonable preferences, advantages or differences in rates between localities. NMSA 1978, § 62-8-6. The PUA prohibits PNM from creating separate rate classes for the City/County for the reasons argued by Dr. Ankum. PNM can create different rate classes based on differing costs of service, but the City/County presented no evidence that the cost of serving them differs from the cost of serving other ratepayers in Rate Classes 3B/3C and 20. PNM’s Initial Posthearing Brief 329. In their Initial Posthearing Brief, the City/County make a related argument that the PRC should not approve “any increases in Rates 3B (general power) or 20 (Streetlighting) as applied to the City and County.” City/County’s Initial Posthearing Brief 10 (emphasis added). In support of this argument, the City/County argue that their electric use has not contributed to any need for increased capacity, including peak demand. To support this claim, the City/County point to Dr. Ankum’s Direct Testimony and Mr. Gurule’s Rebuttal Testimony. Id. Dr. Ankum’s Direct Testimony focuses on the customer class level. Ankum Direct 14, 30-33. Mr. Gurule only offers that the City’s electricity use has decreased given certain programs. Gurule Rebuttal 3-4. None of this information shows that the City/County’s “on-peak demand has not increased.” Moreover, the information at the customer class level does not rise to evidence that the City/County themselves are not “responsible for any of the claimed rate increases …” and Recommended Decision Case No. 15-00261-UT 252 certainly does not justify the creation of a separate customer class with no rate increases. PNM’s Posthearing Response Brief 145. K. RATE 16 — SPECIAL CHARGES Under Rate 16 (Special Charges), PNM collects miscellaneous charges from customers in exchange for performing services not covered under typical electric service tariffs. The purpose of these charges is to recover the reasonable cost that PNM incurs to provide these services. Aguirre Direct 50. PNM proposes several additions or modifications to Rate 16: 1. 2. 3. New charges for the provision of services that are currently not included in Rate 16 Modified charges for existing services based on updated cost data Wording changes to Rate 16 to clarify how the existing charges are applied PNM’s Initial Posthearing Brief 332. The table below shows the new and modified charges. Aguirre Direct 52. Recommended Decision Case No. 15-00261-UT 253 PNM’s proposed changes to Rate 16 are unopposed, and Staff supports the proposed changes. Ms. Leyba-Tercero reviewed PNM’s supporting cost data for its proposed charges and had no reason to dispute the data. She said that it is not in the public interest to price the services below cost because the services are provided to a minority of PNM customers and other PNM customers should not subsidize their costs. Leyba-Tercero Direct 13-17. L. RATE 23 — SPECIAL CONTRACT SERVICE PNM proposes to eliminate Rate 23 — Special Contract Service — because no customer has signed up to receive service under this Rate since its creation in 2003. Aguirre Direct 62?? Walmart takes no position on PNM’s proposed elimination of Rate 23, but recommends that if the PRC does eliminate the Rate, the PRC state that its elimination does not prejudice future commissions in considering alternative supply options for large customers. Walmart witness Chriss recommends that PNM be required to work with interested stakeholders to develop energy supply options, with a particular focus on renewables. Chriss Direct 6, 20-23. PNM’s proposal to eliminate Rate 23 should be approved, subject to recommendation. M. ECONOMIC DEVELOPMENT RIDER Section 62-6-26 of the PUA, titled “Economic development rates for gas and electric utilities; authorization,” took effect on June 19, 2015. It allows the PRC to approve special rates or tariffs “in order to prevent the loss of customers, to encourage customers to expand present facilities and operations in New Mexico and to attract new customers where necessary or appropriate to promote economic development in New Mexico.” NMSA 1978, § 62-6-26(A). The number of PNM’s Large Power Customers, served under Rate 4B, has been declining as shown in the table below: Number of Large Power Customers 6/2010 12/2012 11/2013 6/2014 3/2015 251 231 229 221 220 Ortiz Direct 41-42. Recommended Decision Case No. 15-00261-UT 254 PNM proposes a new economic development rider — Rider No. 45 — “to encourage new industry to locate in New Mexico and incentivize existing customers to further invest in their business in this State.” Chan Direct 72-73. The proposed Rider authorizes a discount (the EDR Discount) to participating customers. To be eligible for Rider No. 45, a customer must: x x x x Receive service under one of the following Classes: 4B, 5B, 30B, and 35B Have at least 500 kW of New Demand, if a New Retail Customer Have at least 200 kW of Incremental Demand, if an Existing Retail Customer Make at least 50% of its sales from sources outside of New Mexico. Exh. SC-14 to Chan Direct. The Standard EDR Duration is four years. At a customer’s request, PNM shall file an application with the PRC requesting that the EDR Discount apply for an additional year. This is consistent with the time limits in § 62-6-26 when a utility does not have excess capacity. NMSA 1978, §§ 62-6-26(C) & (D). The proposed EDR Discount is: x During the first year, a maximum 50% discount per month on its Base Tariff Demand Charges x During the second year, a maximum 35% discount per month on its Base Tariff Demand Charges x During the third and fourth years, a maximum 20% discount per month on its Base Tariff Demand Charges x During the fifth year, a maximum 10% discount per month on its Base Tariff Demand Charges Total estimated billings based on charges including the EDR Discount cannot fall below the Incremental Cost of providing service. Id. PNM says that the Tariff would not increase costs to other customers because the proposed EDR Discounts are developed to recover at least the incremental cost to serve such customers. Chan Direct 77. Staff witness Reynolds said, “Given that New Mexico appears to be lagging other states in its economic recovery, it is in the public interest that an offering of an economic development tariff be considered as proposed by PNM.” Reynolds Direct 12. However, Mr. Reynolds points Recommended Decision Case No. 15-00261-UT 255 out a significant problem with PNM’s proposed Rider No. 45: it is inconsistent with Section 626-26(D)(1)’s explicit requirement that the PRC determine the incremental cost of providing service to an economic development rate customer. When a utility does not have excess capacity, a statutory condition of approval of economic development rates is that the rates “shall not be lower than the incremental cost of providing service to the economic development rate customer as determined by the commission.” (emphasis added). PNM admits that its proposed Rider does not require PRC approval of incremental cost; rather, PNM calculates incremental cost. PNM defends this obvious inconsistency with the Statute by saying, “Requiring a long or burdensome approval process before the Commission for potential EDR customers may discourage participation in the program.” Chan Rebuttal 51-52. Regardless, neither PNM nor the PRC can eliminate the requirement included by the New Mexico Legislature that the PRC determine incremental cost. PNM’s proposed Rider No. 45 should be approved, conditioned on PNM amending the language in the Rider to require PRC approval of the “the incremental cost of providing service to the economic development rate customer.” N. TARIFF REVISIONS AND CLEAN-UP PNM proposes minor “clean-up” revisions to several of its tariffs, which are described in Rule 530 Schedule O-4. These revisions are unopposed and should be approved except to the extent they conflict with recommendations in this Order, including the recommendation WR PDLQWDLQ 310¶V H[LVWLQJ 728 3HULRG DQG 728 SULFLQJ VWUXFWXUH O. ELIMINATION OF CONSOLIDATION ADJUSTMENT RIDER PNM proposes to eliminate its Consolidation Adjustment Rider (CAR) for all customer classes except the Streetlighting class. The PRC approved PNM’s in the 2010 PNM Rate Case was created to assist with the accelerated consolidation of PNM South and North tariffs. In that case, a partial consolidation schedule for PNM North and South was approved earlier than Recommended Decision Case No. 15-00261-UT 256 originally permitted by the Stipulation approved in Case No. 04-00315-UT, although a rate impact was expected for PNM North customers. The CAR was created to reduce that impact by approximately $4.1 million for PNM North customers. The CAR currently applies to PNM South customers taking service under the following rate schedules: Rate 1A – Residential; Rate 1B – Residential TOU; Rate 2A – Small Power; Rate 2B – Small Power TOU; Rate 3B/3C – General Power; Rate 10A/10B – Irrigation; and Rate 20 – Streetlighting. PNM says that eliminating the CAR is an important step towards full consolidation of PNM North and South tariffs and would send more accurate price signals relative to the costs of supplying electricity. Chan Direct 4849. This proposal is unopposed and supported by Staff. Staff witness Leyba-Tercero said that PNM’s proposal is a reasonable step toward full consolidation of PNM North and South rates. Additionally, she agreed with PNM’s assertion that elimination of the CAR would provide more accurate price signals relative to the cost of supplying electricity. Leyba-Tercero 10. PNM’s proposal to eliminate the CAR for all customer classes except the Streetlighting Class should be approved. P. MODIFICATIONS TO VOLTAGE ADJUSTMENT FACTORS PNM proposes to revise its voltage class adjustment factors, which reflect the relative energy loss rates for each class for the Test Period as compared to the Company average energy loss rate for the Test Period. Exhibit 7 to Mr. Aguirre’s Direct Testimony shows the revised loss factors. PNM-58 (Aguirre) at 19:3-8 and PNM Exhibit JCA-7. PNM’s proposed changes an unopposed and should be approved. Q. SPLIT OF DEMAND-RELATED REVENUE REQUIREMENT BETWEEN SEASONS PNM proposes to split its demand-related revenue requirement between seasons to assign its demand production costs. All other demand-related costs are considered nonseasonal in nature and, thus, were assigned proportionally based on the corresponding annual Recommended Decision Case No. 15-00261-UT 257 billing determinants within each applicable rate schedule. Exhibit 4 to Mr. Aguirre’s Direct Testimony shows the derivation of the factors used to assign demand production costs between seasons. As a result of this analysis, PNM proposes to assign approximately 38% of the demand production costs to the summer season and approximately 62% to the non-summer season. Exhibit JCA-5 to Mr. Aguirre’s Direct Testimony shows current and proposed non-volumetric charges, customer and demand charges, by rate schedule for all retail classes. Aguirre Direct 1314. PNM’s proposal is unopposed and should be approved. XXIII. PNM’S PROPOSED REVENUE BALANCING ACCOUNT A. THE EFFICIENT USE OF ENERGY ACT The Efficient Use of Energy Act (EUEA) requires electric and gas utilities to acquire all cost-effective and achievable energy efficiency and load management resources available in their service territories. For electric utilities, this requirement shall not be less than savings of 5% of 2005 total retail kWh sales to New Mexico customers in 2014 and 10% of 2005 total retail kWh sales to New Mexico customers in 2020 as a result of energy efficiency and load management programs implemented starting in 2007.57 A public utility that undertakes cost-effective energy efficiency and load management programs has the option of recovering its prudent and reasonable costs along with Commission-approved incentives for demand-side resources and load management programs through an approved tariff rider or in base rates, or by a combination of the two.58 The EUEA requires the Commission to identify regulatory disincentives or barriers for public utility expenditures on energy efficiency and load management measures and ensure that they are removed in a manner that balances the public interest, consumers’ interests and 57 58 NMSA 1978, § 62-17-5(G). Id., § 62-17-6(A). Recommended Decision Case No. 15-00261-UT 258 investors’ interests. The Commission also must provide public utilities an opportunity to earn a profit on cost-effective energy efficiency and load management resource development that, with satisfactory program performance, is financially more attractive to the utility than supply-side resources.59 Therefore, utilities have the opportunity to both remove disincentives and provide incentives. A disincentive that the EUEA intended to eliminate is the amount of fixed costs that utilities would not recover as the result of the implementation of energy efficiency and load management programs. Case No. 10-00197-UT, Final Order Disapproving Certification of Stipulation and Denying Application Without Prejudice, ¶ 6 (11-10-11). The EUEA states that disincentives to energy efficiency programs should be removed “in a manner that balances the public interest, consumers’ interests and investors’ interests.” NMSA 1978, §§ 62-17-2(E) & 62-17-3 (verify). In Attorney General I, the New Mexico Supreme Court held that disincentive removal must be based on costs. Case No. 11-00047-UT, Certification Recommending Modification of Stipulation 20 (9-13-12), adopted by Order (12-1112). PNM has an incentive in place. PNM currently recovers incentives through its Energy Efficiency Rider. PNM’s approved incentive for plan year 2015 is $1,750,000, prorated for midyear implementation beginning in June 201560; provided, however, that PNM is entitled to the incentive only if it meets or exceeds a minimum of 452,000,000 kWh of verified cumulative savings through December 31, 2015. Case No. 14-00310-UT, Certification of Stipulation 51-52, 63 (4-10-15), adopted by Final Order (4-29-15). PNM exceeded this benchmark in 2015. Tr. (419-16) 3498 (Pitts). B. HISTORY OF DECOUPLING IN NEW MEXICO Id., § 62-17-5(F). PNM’s Energy Efficiency 2015 “Plan Year” is the 7-month period of June through December 2015. Case No. 14-00310-UT, Certification of Stipulation 62. 59 60 Recommended Decision Case No. 15-00261-UT 259 Revenue decoupling is a method designed to reduce or eliminate a utility’s disincentive to promote conservation and energy efficiency by removing the link between the utility’s sales and the utility’s revenues. It does so by defining a target revenue and placing over and undercollections compared to the target in a deferred account for refund or recovery in a later period. Hansen Direct 3. In 2012, the PRC rejected a proposal to declare revenue decoupling the mechanism of choice to eliminate disincentives to energy efficiency programs. In July 2012, in Case No. 1200250-UT, the PRC issued a notice of proposed rulemaking (NOPR) to revise Rule 17.7.2. The proposed rule attached to the NOPR contained a subsection titled “Revenue Decoupling” that stated in part, “Revenue decoupling shall be the mechanism used to remove an electric utility’s regulatory disincentives to promote and engage in energy efficiency measures.” Case No. 1200250-UT, Order Closing Docket in Case No. 12-00144-UT and Issuing New NOPR, Attachment A, §17.7.2.19 (7-26-12). However, the PRC did not publish the proposed rule but held workshops to consider the proposed rule. In October 2012, following the workshops, the PRC voted to not publish a NOPR to revise Rule 17.7.2. Case No. 12-00250-UT, Douglas Howe’s Dissent (11-212). No further action occurred in Case No. 12-00250-UT. In October 2013, the PRC, in Case No. 13-00310-UT, issued and published a NOPR to revise Rule 17.7.2. The proposed rule contained only the following provision regarding removal of disincentives: 17.7.2.17 REGULATORY DISINCENTIVES: The commission shall, upon petition or its own motion, identify regulatory disincentives or barriers for public utility expenditures on energy efficiency and load management measures and ensure that they are removed in a manner that balances the public interest, consumers’ interest and investors’ interests. Case No. 11-00310-UT, Order Initiating Proposed Rulemaking, Staff’s Exhibit A, § 17.7.2.17. In its Final Order in Case No. 13-00310-UT, the PRC adopted the proposed Subsection 17.7.2.17 but added a second sentence to the Subsection. Subsection 17.7.2.17 as adopted reads: Recommended Decision Case No. 15-00261-UT 260 17.7.2.17 REGULATORY DISINCENTIVES: The commission shall, upon petition or its own motion, identify regulatory disincentives or barriers for public utility expenditures on energy efficiency and load management measures and ensure that they are removed in a manner that balances the public interest, consumers’ interest and investors’ interests. Public utility petitions for regulatory disincentive removal shall be supported by testimony and exhibits. Case No. 13-00310-UT, Final Order Repealing and Replacing Rule 17.7.2 NMAC, Exh. A1, § 17.7.2.17. The PRC’s existing Energy Efficiency Rule, 17.7.2.9 NMAC, which took effect on January 1, 2015, does not change the language in 17.7.2.17 NMAC adopted in Case No. 13-00310-UT. In Case No. 06-00210-UT, the Commission rejected a decoupling proposal by PNM for its then-natural gas utility. The mechanism proposed in that case adjusted PNM’s revenues for deviations between expected and actual sales. It did not adjust revenues if the actual number of customers differed from the base line or assumed number of customers. The Hearing Examiner in Case No. 06-00210-UT found PNM’s proposal overbroad because it insulated PNM from revenue losses resulting from causes other than PNM’s energy efficiency programs.61 Recommended Decision 115 (5-23-07). He also found that the proposal was designed to create a financial windfall because it would compensate PNM for lost revenues due to declines in per customer use even if PNM’s total revenues remained stable or grew because of customer growth. Id. at 116-17. He concluded that the proposal would not make PNM financially neutral, as required by the EUEA, but would instead insulate PNM from a large variety of business risks including the risk of customer response to higher prices. Id. at 119. He noted that PNM sought this protection without any reduction in its authorized return on equity. Id. at 122. The Hearing Examiner not only recommended rejection of PNM’s specific decoupling proposal, but found “that decoupling is not in the public interest and is inconsistent with the Public Utility Act.” Id. at 125. 61 Case No. 06-00210-UT, Recommended Decision of the Hearing Examiner at 115 (5-23-07). Recommended Decision Case No. 15-00261-UT 261 In its Final Order, the Commission adopted the Hearing Examiner’s recommendation to reject PNM’s revenue decoupling proposal. The Commission described the proposal as “broad and overreaching,” and noted that (i) it would make PNM whole for past conservation efforts of consumers that have absolutely nothing to do with the enactment of the EUEA; and (ii) it fails to take account of customer growth that occurs during the time that per customer consumption declines. Final Order Partially Adopting Recommended Decision, ¶ 119 (6-29-07). The Commission did not adopt the Hearing Examiner’s wholesale rejection of revenue decoupling. The Commission said: That is not to say, however, that the Commission will not consider a welldesigned decoupling proposal that meets the criteria of the [EUEA]. The Commission welcomes appropriate measures to eliminate disincentives to investment by utilities in energy efficiency programs as contemplated by the Act. However, they must be narrowly focused to address those disincentives, and not be aimed at making the utility whole for all load losses. Id. at 120. In 2011, in a challenge to a previous version of Rule 17.7.2 in which the Commission adopted Interim Adder Rates, the New Mexico Supreme Court held that incentives and disincentives under the EUEA must be cost-based. Attorney General v. New Mexico Pub. Regulation Comm’n, 2011-NMSC-034, 150 N.M. 174 (Attorney General I). At the evidentiary hearing in the rulemaking, each of the three investor-owned utility companies in New Mexico presented evidence on the projected impact of the Interim Adder Rates, i.e., the projected revenue the utility anticipated losing due to its energy efficiency programs and the projected recovery the utility would experience if the PRC were to adopt the Interim Adder Rates. Id., ¶ 7. The PRC determined that the EUEA did not require that the Interim Adder Rates be cost-based. Id., ¶ 12. The PRC did not inquire into any of the utilities’ revenue requirements, nor any of the traditional elements of the ratemaking process. Id., ¶ 18. The New Mexico Supreme Court held that the adder rates were arbitrary because they were not evidence-based, cost-based, or utility-specific, and because the PRC failed to balance Recommended Decision Case No. 15-00261-UT 262 the interests of consumers and investors by not considering whether the additional adder revenues provided under the rule amendments would cause utilities to earn more than a reasonable return. The Court stated that it reads the EUEA in harmony with the Public Utility Act and concludes that when the Commission sets a rate, such as the stipulated adder rates, the Legislature intended that the Commission conduct the same balancing process under the EUEA to remove disincentives as it does under the Public Utility Act to determine just and reasonable rates. The Court stated that the balancing process is grounded in the consideration of a utility's costs -- whether the utility is recovering its prudently incurred costs and earning a reasonable return on its capital investments and whether ratepayers are protected from excessive rates. Id., ¶ 16. The Court said, “Without inquiring into a utility’s revenue requirements, we fail to see how the PRC could adequately balance the investors’ interests against the ratepayers’ interests when adopting [the Interim Adder Rates].” Id., ¶ 18. After issuance of the Attorney General I opinion, the PRC made clear in Case No. 1000197-UT that a condition of approval of a mechanism to remove disincentives is evidence of lost fixed costs. In rejecting a stipulated mechanism for removing disincentives to Southwestern Public Service Company, the PRC said: Thus, in order to meet the requirements of the Attorney General decision for approval of an adder or add-like mechanism to eliminate disincentives, a utility must show the amount of fixed cost that would not be recovered as the result of its energy efficiency and load management programs and that the amount to be recovered by its proposed adder-like mechanism would not recover more than that amount. Case No. 10-00197-UT, Final order Disapproving Certification of Stipulation and Denying Application Without Prejudice, ¶ 6 (11-10-11). Most recently, in Case No. 11-00047-UT, the PRC again made clear that a condition of approval of a mechanism to remove disincentives is evidence of lost fixed costs. The PRC disapproved a stipulation that provided an additional increment of revenue, an “adder,” to be recovered by El Paso Electric Company (EPE) to remove disincentives and provide incentives for Recommended Decision Case No. 15-00261-UT 263 EPE’s energy efficiency programs. Case No. 11-00047-UT, Certification of Stipulation (9-13-12), adopted by Final Order (12-11-12). The amount of adder revenues to be recovered was determined by multiplying $0.0045/kWh and $20/kW rates by the estimated lifetime savings from the energy efficiency programs implemented in each year. The adder rates did not distinguish how much of the amounts were intended to remove disincentives and how much were intended to provide an incentive. Certification of Stipulation 6. The PRC adopted the Hearing Examiner’s finding that EPE’s evidence of “net lost revenues” — the revenues EPE would not receive due to the reduced kWh sales achieved by its energy efficiency programs — was insufficient to support the adder rates. EPE stated that $0.066156/kWh of its $0.1179/kWh summer residential rate was designed to recover fixed costs. It calculated its 2012 projected lost revenues by multiplying $0.066156/kWh by its projected savings in kWh. Lacking were estimates of EPE’s fixed costs in 2012 and 2013 and revenues to recover those costs. In fact, EPE’s kWh sales had increased in recent years at a rate that exceeded the kWh sales reductions that it achieved with its energy efficiency programs. EPE’s witness said that the actual amount of EPE’s lost fixed costs — its annual fixed costs minus any reduced revenue to recover those fixed costs — could not be determined without conducting “a miniature rate case each year.” Id. at 9-10. Staff was the only party that presented evidence of EPE’s lost fixed costs. Staff estimated EPE’s annual fixed costs for its residential class for the years 2009, 2010, and 2011 by escalating EPE’s Test Period fixed costs from its most recent rate case. Staff then compared this estimate to EPE’s annual revenues in 2009, 2010, and 2011 from the portion of its kWh rate intended to recover its fixed costs and the customer charge. Staff’s evidence showed that EPE recovered more in revenues than the fixed costs it incurred in 2009, 2010, and 2011. Staff testified that if there was no evidence of lost fixed costs, the adder rates would produce only incentives. Id. 11. Recommended Decision Case No. 15-00261-UT 264 No party presented evidence showing that EPE would not recover the fixed costs that it expected to incur in 2012 and 2013: No party provided evidence that the amount of fixed costs EPE is expected to incur in 2012 and 2013 will be larger than the revenues that will be available to recover them in those years, leaving a sum of fixed costs that is un-recovered or “lost.” Id. 9, 14. In recommending rejection of the proposed rates, the Hearing Examiner explained: Revenues and lost revenues, however, are not measures of costs, and lost revenues (and lost contributions to fixed costs) are not a measure of lost fixed costs. .... A ‘lost revenues’ analysis, without the consideration of EPE’s costs and the potential for over-earning, conflicts with the New Mexico Supreme Court’s holding in Attorney General [v. NMPRC, 2011-NMSC-034].” Id. 14, 17. The PRC emphasized this point in its Final Order when it said: More importantly, however, is the underlying principle that a loss of revenue does not automatically result in a disincentive, as it may be accompanied by a corresponding reduction in expenses that may equal or exceed the loss of revenue. Depending on where the utility’s production of electricity is on its cost curve, it may well be advantageous to the utility to experience a loss of revenue. Final Order, ¶ 5. Importantly, for purposes of this case, the PRC said that “disincentive removal is best accomplished in a rate case,” which provides the opportunity to examine the impact of energy efficiency programs on a utility’s costs, revenues, and earnings. Id. at 22. It further said, “The use of a future Test Period provides a vehicle for incorporating the impact of the programs on the utility’s costs.” Id. C. PNM’S PROPOSED REVENUE BALANCING ACCOUNT (RBA) PNM’s argues that it has a disincentive to promote energy efficiency programs because it recovers significant fixed costs through its energy/volumetric rates, so that any resulting sales reductions will cause its revenues to be reduced by more than its avoided costs. Chan Direct 67. Recommended Decision Case No. 15-00261-UT 265 Therefore, every unsold kWh reduces PNM’s fixed cost recovery. In fact, PNM witness Ortiz said that PNM’s only avoided costs are its fuel costs. Tr. 4-11-16 (45). PNM estimated that it recovers 8.13¢/kWh in fixed costs through volumetric charges from residential class members and 3.25¢/kWh in fixed costs, on average, from all other classes except the Irrigation Class. This represents PNM’s per-kWh under-recovery for each kWh of saved energy. Chan Direct 67. PNM estimated the total amount of its fixed costs not recovered because of energy efficiency programs by multiplying the fixed cost that it recovers through the volumetric rate (8.13¢/kWh and 3.25¢/kWh) by projected kWh savings. It projects $24,905,720 in lost fixed recovery in 2016 from energy efficiency programs. Chan Direct 68. PNM witness Stella Chan said, “In summary, because there is a reduction in revenues as well as in fixed cost recovery stemming from energy efficiency programs due to current rate design, there is a regulatory disincentive for energy efficiency measures.” Id. at 69 (emphasis added).62 PNM says that its residential and small power classes are responsible for about 84% of its under-recovery of fixed costs. Chan Direct 68. PNM says that adoption of its proposed increases in the monthly charge for residential and small power service customers would not remove its disincentive to promote conservation and energy efficiency. PNM proposes to increase its monthly charges for residential and small power service customers to $13.14 and $17.87, respectively. PNM says that it would need to increase its monthly charges for residential and small power service customers to $61.62 and $163.72, respectively, to recover all of its fixed charges through the customer charge. Hansen Direct 7. PNM allocated the residential class a $339 million fixed revenue requirement, but Based on an independent evaluator’s report, PNM’s energy efficiency programs produced 287 GWh of cumulative savings as of 2014. The resulting revenue impact was about $21 million in 2015. Ortiz Direct 35. 62 Recommended Decision Case No. 15-00261-UT 266 would only collect about $72 million from the residential class through its proposed customer charge. Hansen Direct 8. PNM asserts that because it recovers a significant amount of fixed costs through volumetric rates, sales reductions cause PNM’s revenues to be reduced by more than its avoided costs, creating a disincentive for PNM to promote conservation and energy efficiency. Hansen Direct 7. According to PNM, “The fact that PNM recovers fixed costs through per-kWh rates means that its profitability is directly tied to its sales levels.” Hansen Direct 8; see also Cavanagh Direct 11-12. As part of the stipulation approved in PNM’s last rate case, the stipulating parties agreed that PNM would withdraw the decoupling proposal it made in its application and that PNM’s disincentives would be deemed recovered in its stipulated rates. Case No. 10-00086-UT, Certification of Stipulation 43. They also agreed: Before PNM requests Commission approval of any mechanism to address disincentives to utility energy efficiency programs, PNM and other parties shall engage in good faith consultations regarding alternative ratemaking solutions, including alternative mechanisms such as off-system sales credits, increased demand charges or reducing the recovery of fixed costs through volumetric charges for non-residential customers. PNM shall act in good faith to incorporate the suggestions of other Signatories into its filing. Any suggestions not incorporated by PNM must be specifically identified and thoroughly analyzed in its filing. Amended Stipulation, ¶ 25. PNM met with stakeholders on September 29, 2014 and November 5, 2014 to discuss potential alternatives to decoupling. Chan Direct 72. PNM considered the following alternatives to decoupling for removing its disincentive: x Increasing demand charges or reducing the recovery of fixed costs through volumetric rates for non-residential customers x Off-system sales credits x Future Test Period x More frequent rate cases x Straight Fixed Variable (SFV) rate design x Lost Revenue Adjustment Mechanisms (LRAMs) x A minimum bill provision. Recommended Decision Case No. 15-00261-UT 267 Hansen Direct 13. PNM considered, but rejected these alternatives, and, in this case, proposes a four-year pilot mechanism which it calls a “Revenue Balancing Account” or RBA, which PNM proposes to implement through its proposed Rate Rider No. 44. The RBA would apply only to residential customers taking service under Rates 1A and 1B and small power customers taking service under Rates 2A and 2B. For these groups, the PRC would authorize PNM to collect a pre-established level of revenue toward fixed cost recovery regardless of PNM’s actual sales revenues. The PRC would approve an allowed revenue per customer class. Annually, PNM would calculate the difference between the total allowed revenue per customer and the actual revenue per customer recovered through sales. The overor under-collection of fixed costs would be collected from or refunded to customers the next year. Chan Direct 64-65; Ortiz Direct 37. The actual fixed cost recovery through the volumetric rate would be calculated by multiplying monthly kWh sales by an approved per/kWh. The difference would be calculated monthly, and the annual balance would be incorporated into customer rates for the following year by dividing the balance by the forecasted sales to the customer class. Separate fixed cost recovery amounts and per kWh rates would apply to each customer group. For customers paying block or time-of-use rates, the actual fixed cost recovery would be approximated using a single fixed variable charge. Hansen Direct 26-28. Any surcharge or credit would be applied as a flat $/kWh adjustment to a customer’s variable charge. If the RBA would result in a rate case increase of more than 5% of a rate Class’ Test Period revenue, as approved in PNM’s last rate case, the amount above 5% would be deferred for recovery in a future year. PNM proposes to apply a carrying charge equal to the Customer Deposit Interest Rate to RBA deferrals. Hansen Direct 31-32. Recommended Decision Case No. 15-00261-UT 268 The RBA would not apply to General and Large Power Service customers because a relatively small percentage of the fixed costs of serving these customers would be recovered through PNM’s proposed volumetric rates. Hansen Direct 28-29. PNM proposes revenue decoupling as a four-year pilot, beginning in the month following PRC approval. Before the end of the four-year period, PNM would file a recommendation to continue, modify or terminate the mechanism. Hansen Direct 30. PNM opposes reducing its requested ROE if the RBA is approved. Hevert Direct 61-62. In her Direct Testimony, PNM witness Chan said, “The RBA also will account for compensating offsets to under-recovery of fixed costs from other factors, such as weather or increased energy consumption due to new end-uses of electricity.” Chan Direct 70. However, at the hearing, Ms. Chan admitted that the RBA does not account for differing changes in energy use. In other words, it adjusts PNM’s revenues for any deviation between expected sales and actual sales, regardless of the reason for the deviation. Tr. 4-21-16 at 225556. Thus, it protects PNM from losses in sales due to factors such as an economic downturn, an unexpectedly cool summer, or customer response to high rates. Absent decoupling, PNM’s shareholders absorb those losses. D. POSITIONS OF STAFF AND INTERVENORS Among the parties and Staff who took positions on the RBA, the following positions were taken: x CCAE supports the RBA and recommends that the PRC consider three changes to the RBA. Cavanagh Direct 26. x The AG opposes the RBA unless the PRC incorporates five identified customer protections into the RBA. Gegax Direct 20-25. x Staff opposes the RBA. Pitts Direct 4. x ABCWUA opposes the RBA. ABCWUA’s Initial Posthearing Brief 108-09. Staff’s reasons for opposing the RBA mirror the PRC’s reasons for rejecting the decoupling mechanism proposed in Case No. 06-00210-UT: Recommended Decision Case No. 15-00261-UT 269 1. Under the RBA, “any decline in load growth is attributed to the ‘success’ of energy efficiency programs without considering other factors involved.” 2. The RBA is an overly broad mechanism to address the negative impact of energy efficiency on revenues. Pitts Direct 11-14. Witnesses in support of decoupling testified extensively of policy reasons to support decoupling and adoption of decoupling nationwide. See Cavanagh Direct; Cavanagh Rebuttal 4; Gegax Direct 21-23. Mr. Cavanagh criticized Staff for failing to “acknowledge or respond to the reality of the mechanism’s extensively documented success around the country in recent decades.” Cavanagh Rebuttal 2. Notably, none of these witnesses proposed changes to the RBA to eliminate the PRC’s concerns. Instead, they disagreed with the PRC’s expressed concerns. E.g., Cavanagh Direct 1821 (arguing against narrowing the RBA to only capture decreased sales from EE); 24 (arguing that the RBA would not result in paying PNM for savings that it did not help to achieve); 25 (arguing that decreases in sales because of economic downturns are likely to result in increased rates with or without decoupling). CCAE witness Cavanagh identified as a benefit of the RBA that PNM would refund to customers gains in fixed cost recovery from unexpected increases in sales. He did not, however, address the offsetting concern of the PRC in Case No. 06-00210-UT that the RBA, like the decoupling mechanism proposed in Case No. 06-00210-UT, would create a financial windfall because it would compensate PNM for lost revenues due to declines in per customer use even if PNM’s total revenues remained stable or grew because of customer growth. E. ANALYSIS/RECOMMENDATION PNM’s RBA should be rejected for several reasons: First, the RBA is no more narrowly tailored to recover revenue losses from EE programs than the decoupling proposal that did not survive scrutiny in Case No. 06-00210-UT. It shifts Recommended Decision Case No. 15-00261-UT 270 the risks of economic cycles and weather fluctuations from utilities to ratepayers. Staff witness Pitts aptly and correctly concluded that “to create a revenue recovery mechanism that is not subtle enough to distinguish among the various impacts on energy sales is much too broad.” Pitts Rebuttal 14. A decoupling mechanism that does not address the potential for over-earning conflicts with Attorney General I. The RBA places significant financial risk on ratepayers and shields shareholders from the economic impact of recession. If another serious recession results in lower than projected sales levels, PNM would be entitled to recover lost revenues resulting from that recession. As commissions reduce utility exposure to uncontrollable or unforeseeable risks through adjustment mechanisms, incentives that once encouraged operating efficiency are reduced. Returning the risk of sales revenues and non-fuel expenses to the utility provides utility management with the right incentive and flexibility to reduce costs and better market its commodity. No parties proposed adjustments that can be made to a decoupling mechanism to reduce the shift of risks to ratepayers. For example, allowed revenue could be normalized for weather or economic conditions or allowed revenue could be adjusted based on the number of customers, which would leave utilities subject to economic conditions to a degree. Or, PNM’s return on equity could be lowered to reflect the reduced risk to shareholders resulting from decoupling adjustment. While only about 20% of state commissions that have approved decoupling have linked decoupling to a decrease in ROE, Cavanagh Direct 15, a ROE reduction should flow from approval of decoupling. Pitts Rebuttal 16-17; Gorman Direct 4-5; Chriss Direct 13. Second, PNM submitted no evidence that it would not recover the fixed costs that it expects to incur in the Test Period. To the contrary, PNM has incorporated projected lost fixed costs due to EE programs in its Test Period projected revenues. Faruqui Direct 29-34; Tr. (4-12- Recommended Decision Case No. 15-00261-UT 271 16) 309 (Ortiz). PNM admitted that if its sales forecast is completely accurate, “then there would be no decoupling adjustment.” Tr. (4-12-16) 309 (Ortiz). While there is no guarantee that PNM’s sales forecast will prove completely accurate, the financial impact of forecasting errors is minimized when the time period between rate cases is shorter. (Tr. (4-12-16) 310 (Ortiz); Hansen Surrebuttal 6-7. Tr. (4-12-16) 310 (Ortiz). PNM’s intent is to file another rate case application as soon as December 2016. Tr. (4-12-16) 310 (Ortiz). The risk of forecasting errors is minimized by use of a future Test Period (Pitts Rebuttal 15). Third, none of the decoupling proponents addressed whether distributed generation (DG) customers contribute to PNM’s lost revenues. If they do, including their use in monthly sales in the RBA calculation might be inconsistent with Section 62-13-13.2 of the PUA because the legislative intent is to assess additional costs caused by DG customers to DG customers. CCAE’s assertion that “every unsold kilowatt-hour would reduce the Company’s fixedcost recovery and undercut shareholder welfare,” Cavanagh Direct 9, is not true to the extent that kWh reductions are incorporated into PNM’s sales forecast and revenue requirement. In Case No. 11-00047-UT, the PRC indicated that incorporating reduced sales from EE in the Test Period sales forecast in a utility’s rate case is the desired way to eliminate a utility’s disincentive to EE resulting from lost fixed recovery. Approving the RBA would reduce the need to accurately calculate Test Period sales in a general rate case. CCAE argues that the RBA removes a separate disincentive to energy efficiency programs, which it identifies as a utility’s desire to promote electricity sales: once rates are set, utilities have incentives to promote electricity sales; if a utility sells more than had been expected, it earns more revenue, which it is allowed to retain. Conversely, if a utility sells less than anticipated, it bears the burden of the revenue shortfall. CCAE argues that promoting sales is not desirable when the state and the country are trying to conserve energy. Recommended Decision Case No. 15-00261-UT 272 The Undersigned is aware of no case in which the PRC has identified a utility’s desire to promote energy sales as a disincentive to EE that the EUEA intended to remove. Even if the EUEA so intended, no evidence shows that PNM itself has the ability to materially increase its sales. Under the federal Public Utility Regulatory Policies Act, electric public utilities cannot offer promotional rate structures except when they can be justified by the cost structure of utility companies. 16 U.S.C. § 2621(d)(2). Under PRC Rules, a public utility cannot recover through rates the cost of advertising that promotes increases in use of energy. 17.3.350.9(C)(1) NMAC. Additionally, PNM’s residential inclining block rate structure is inconsistent with an incentive to promote energy sales. PNM has no control over the driving factors in energy sales: interest rates, fossil fuel prices, inflation, and the economy. Recently, the Montana Public Service Commission, quoting from witness testimony, recognized: Whether or not a utility is able to achieve the authorized rate of return is a function of many independent factors between general rate filings including customers’ usage levels influenced by weather, customer-specific behaviors and preferences causing load growth or decline, an increasing or decreasing number of actual customers, utility cost increases or decreases, general economic conditions, and the level of investment in or retirement of utility assets. In re Lost Revenue Adjustment Mechanism, Docket No. D2014.6.53, Order No. 7375a, ¶ 42, 2015 WL 6128712 (Mont. P.S.C. Oct. 6, 2015). No evidence shows that the absence of decoupling has subverted efforts to promote conservation and the efficient use of energy. Rather, evidence shows that PNM has aggressively pursued EE programs. In 2015, PNM exceeded its energy efficiency goals. Tr. (4-29-16) 3498 (Pitts). PNM has a mature portfolio of EE programs, the actual costs of which it fully recovers through the EE Rider. It is more important to encourage conservation by sending consumers proper price signals than by changing a utility’s behavior. For these reasons, the RBA as proposed by PNM and as proposed with modifications by Intervenors, should be rejected. Recommended Decision Case No. 15-00261-UT 273 XXIV. PAYMENT CENTERS In 2011, PNM informed the PRC that it intended to close its eight remaining walk-in payments centers. In response, the PRC docketed an investigation in Case No. 11-00435-UT. The PRC’s Final Order in Case No. 11-00435-UT approved a stipulation subject to PNM’s agreement to one change to the stipulation, which PNM agreed to. Under the amended approved stipulation, PNM agreed to keep open its eight walk-in payment centers at least two days per week and at least through the effective date of rates adopted in PNM’s next general rate case. PNM also agreed to arrange for Western Union to accept cash or check payments of PNM bills, at no fee to customers, at 16 designated Western Union facilities in Alamogordo, Bayard, Deming, Las Vegas, Lordsburg, Ruidoso, Silver City, Albuquerque, Clayton and Santa Fe, at least through the effective date of the rates adopted in PNM’s next general rate case. The amended stipulation says: The continuation of these PNM payment centers and fee free payment facilities will be addressed by PNM in its next general rate case filing and discontinuation of these facilities will be subject to Commission approval in that proceeding. Case No. 11-00435-UT, Amended Stipulation to Conform to Commission Order, ¶ 14 (7-5-12), approved by Final Order (6-28-12). In this case, PNM recommends that it continue to operate each of the payment centers on the current operating schedule. Ortiz Direct 59. This recommendation should be adopted. XXV. 1. FINDINGS OF FACT AND CONCLUSIONS OF LAW The Findings and Conclusions in all Sections of this Recommended Decision and the Decretal Paragraphs contained in the Recommended Decision are adopted, approved, and ordered by the Commission. 2. PNM is a public utility as defined in the PUA. 3. The PRC has jurisdiction over the parties and the subject matter of this case. Recommended Decision Case No. 15-00261-UT 274 4. Reasonable, proper and adequate notice of this case has been given. 5. PNM’s currently authorized rates are not fair, just and reasonable. 6. The tariffs filed under Advice Notice No. 513 contain rates that are not fair, just and reasonable. 7. Advice Notice No. 513 is cancelled. 8. PNM should file, under a new Advice Notice, rates consistent with the cost of service and rate design determinations reflected in this Order and in the Attachments to this Order. Such rates shall be effective upon approval as to form and compliance by PRC Staff. 9. Under the same Advice Notice, PNM should file revised tariffs for Rate Rider Nos. 23 and 36 consistent with the rulings in this Order. 10. The Joint Applicants’ suggested corrections to the Transcript are uncontested and should be adopted to the extent that they correct errors in transcription by the court reporter. 11. If PNM changed its position in response to Staff/Intervenor Testimony and the change in position eliminated a contested issue, that issue is not discussed in this Recommended Decision. XXVI. A. DECRETAL PARAGRAPHS All findings and conclusions in all Sections of this Recommended Decision and the Decretal Paragraphs in this Recommended Decision are adopted, approved, and ordered by the Commission. B. PNM’s Advice Notice No. 513 is disapproved and cancelled. C. PNM shall file, under a new Advice Notice, new rates consistent with the cost of service and rate design determinations reflected in this Order and in the Attachments to this Recommended Decision Case No. 15-00261-UT 275 Order. Such rates shall be effective for billing upon approval as to form and compliance by PRC Staff.63 D. Under the same Advice Notice, PNM shall file a revised Renewable Energy Rider Tariff — Rate Rider No. 36 — consistent with the determinations in this Order. E. Under the same Advice Notice, PNM shall file a revised Fuel and Purchased Power Cost Adjustment Clause Rider — Rate Rider No. 36 — consistent with the rulings in this Order. F. PNM is authorized to continue using its Renewable Energy Rider Tariff — Rider G. PNM’s annuitization of the pension benefits of PNM’s former gas utility No. 36. operations eliminates the need to allocate pension expense between electric and gas in future rate cases.64 H. PNM shall adopt “Method A,” which makes the following changes to PNM’s current methods of billing fuel and purchased power costs and renewable costs: 1. Recovers all fuel and purchased power costs through PNM’s FPPCAC, and recovers no such costs through PNM’s base energy rates 2. Moves recovery of the NMWEC procurement costs from the FPPCAC to the Renewable Rider PNM witness Ortiz requested that new rates take effect for bills rendered, not service rendered, as of the effective date. Tr. (4-12-16) 302. Normally, new rates take effect for service rendered beginning on the effective date to avoid retroactive ratemaking issues. However, it is reasonable in this case to allow new rates to take effect for bills rendered beginning on the effective date because of the PRC’s extension of the suspension period. 64 See Ortiz Direct 65. 63 Recommended Decision Case No. 15-00261-UT 276 3. Excludes kWh generated by renewable energy from the calculation of total kWh sales in calculating PNM’s FPPCAC, thereby applying the FPPCAC Factor to only the portion of a customer’s energy use that is estimated to be generated by nonrenewable energy 4. Breaks the FPPCAC charge on a customer bill into two parts: a. One FPPCAC Factor applies to the estimated percentage of a customer’s fuel use generated by non-renewable energy b. The other FPPCAC Factor applies to the estimated percentage of a customer’s fuel use generated by renewable energy — this FPPCAC Factor is zero because no fuel use is associated with use of renewable energy. Therefore, the charge on this line of a customer’s bill will always be zero. c. Uses a different estimated percentage of fuel use generated by renewable energy for each of the three customer classes, which is then multiplied by the FPPCAC factor I. PNM is authorized to include its coal and nuclear fuel handling expenses and purchases and sales of spinning reserves in its FPPCAC calculations. J. One-hundred percent of PNM’s revenues from the chemical pretreatment of coal at the SJGS shall be credited to ratepayers. K. Parties and Staff are put on notice that the PRC will consider in PNM’s next base rate case whether PNM’s Rider No. 8 should be discontinued. If PNM proposes to continue Rider No. 8 in its next base rate case, it shall file direct testimony justifying why it should continue. L. In its renewable energy plan case to be filed in 2017, PNM shall file testimony that: 1. Addresses and corrects the remaining cost misallocation identified by Dr. Howe Recommended Decision Case No. 15-00261-UT 277 2. Discusses the merits of correcting the remaining fuel cost allocation through the method suggested by Dr. Howe 3. Discusses the merits of correcting the remaining fuel cost allocation by calculating each Large Capped Customer’s dollar cap on a net cost basis 4. Discusses the merits of any other method of correcting the remaining fuel allocation that might be suggested by PNM 5. Discusses whether each Large Capped Customer’s dollar cap should be calculated on a net cost basis to be consistent with calculating the Large Customer Adjustment on a net cost basis. M. PNM shall revise the language in the “Special Conditions” section of Rate 20 on page 5, subsection 1(a), to read: Upon request from the Customer, the Company shall convert or install Company owned street lighting fixtures at its own expense . . . . PNM shall also revise the “Special Conditions” section of Rate 20 on page 6, subsection III, to read “Relocations, conversions and changes, other than normal operation . . . N. Within 20 business days of issuance of this Final Order, PNM shall convene stakeholder meetings (inviting Intervenors to this case and other interested stakeholders, such as municipalities) to discuss at a minimum the following issues for inclusion in PNM’s next base rate case: conversion of high pressure sodium lighting to LED lighting, including a) 100,000 hour lights; b) metering and advanced lighting control options at the request of the customer; c) salvage values and recovery of undepreciated assets; d) expanded lighting range options for conversion; and e) installation allowances. PNM shall develop the cost and technical data necessary to develop a tariff that includes these items. 2 310¶V SURSRVHG PLQRU ³FOHDQ XS´ UHYLVLRQV WR VHYHUDO RI LWV WDULIIV DUH DSSURYHG H[FHSW WR WKH H[WHQW WKH\ FRQIOLFW ZLWK UHFRPPHQGDWLRQV LQ WKLV 2UGHU Recommended Decision Case No. 15-00261-UT 278 P. In its next base rate case filing, PNM shall base its proposed 3B and 3C Rate Schedules on separate cost of service studies for each Schedule and shall propose a rate design that moves toward eliminating any intraclass subsidy, consistent with gradualism. Q. Within five business days of issuance of this Final Order, the PRC shall appoint a facilitator to meet with PNM and interested parties to discuss proposals for an effective TOU rate design, and set a date for the first meeting. R. PNM’s request to eliminate Rate 23 is granted. This elimination does not prejudice future commissions in considering alternative supply options for large customers. Upon request, PNM shall work with interested stakeholders to develop energy supply options, with a particular focus on renewables. S. PNM’s proposed Rider Rider No. 45 is approved, conditioned on PNM amending the language in the Rider to require PRC approval of the “the incremental cost of providing service to the economic development rate customer.” T. Within two months of issuance of a final order in this case, PNM shall begin collecting and maintaining data necessary to determine the demand and customer components of distribution facilities. In its next base rate case filing, PNM shall identify which of the two methods identified by NARUC it proposes to use in its following rate case filing to determine the demand and customer components of distribution facilities and shall describe the procedure it has adopted to collect and maintain the data necessary to use this method. U. PNM shall continue to operate its payment centers on the current operating schedule. V. PNM’s proposal to eliminate the CAR for all customer classes except the Streetlighting Class is approved. W. PNM proposed revisions to its voltage class adjustment factors are approved. Recommended Decision Case No. 15-00261-UT 279 X. PNM's proposed split of its demand-related revenue requirement between seasons for rate design purposes is approved. Y. PNM may begin recovering previously approved regulatory assets and liabilities relating to the decommissioning of the Las Vegas Generating Station over two years. Z. PNM's proposed changes to its Special Charges are approved. AA. Suggested corrections to the Transcript are adopted to the extent that they correct errors in transcription by court reporters. BB. Any matter not specifically ruled on during the hearing or in this Final Order is disposed of consistently with this Final Order.6s CC. Issues raised in prefiled testimony or at the hearing, but not argued in post- hearing briefs, are not considered. i.2.2.36(B)(4) NMAC. DD. This Order is effective immediately. Issued at Santa Fe, New Mexico on August 4, 2016. NEW MEXICO PUBLIC REGULATION COMMISSION Qad -{( . clill· u f, _ __, Hearing Examiner 65 See State v. King, 2007-NMCA-130, if 17, 142 N.M. 699 (court may refuse to consider arguments unsupported by authority or analysis); International Minerals & Chemical Corp. v. N ew Mexico Pub. Serv. Comm'n, 1970-NMSC-032, if 8, 81 N.M. 280 (PUA requires that PRC find only the ultimate fact). Recommended Decision Case No. 15-00261-UT 280