REDACTED STATE OF MAINE PUBLIC UTILITIES COMMISSION Docket No. 2018-00194 June 17, 2019 PUBLIC UTILITIES COMMISSION Investigation into Rates and Revenue Requirements of Central Maine Power Company I. REPLY BENCH ANALYSIS INTRODUCTION In response to the Commission’s Notice of Investigation initiating this proceeding, on October 15, 2018, Central Maine Power Company (CMP or Company) filed its Direct Case. In that Direct Case, the Company, claimed that it required an increase to its revenue requirement of $22.9 million. This request was based on a historical test year of July 1, 2017 through June 30, 2018 and assumed that rates would go into effect July 1, 2019. CMP proposed to mitigate this rate increase by accelerating the amortization in the rate-effective year (i.e., the year beginning July 1, 2019) of the Company’s Unprotected Excess Deferred Income Tax (EDIT) regulatory liability, which was created as a result of the Tax Cuts and Jobs Act (TCJA) of 2017. For subsequent years (i.e., beginning July 1, 2020), the Company recommended that the mitigation remain in place by both (a) continuing the acceleration of the EDIT Regulatory Liability and (b) beginning the acceleration of the Company’s Cost of Removal Liability. In response to discovery, the Company acknowledged that its revenue requirement deficiency was calculated based on current rate levels, which included the recovery of a significant number of one-time adjustments (the most significant being the October 2017 storm costs) that were scheduled to be removed from rates on July 1, 2019 as part of the Company’s Annual Compliance Filing (ACF) procedure. See Reply Bench Analysis REDACTED 2 Docket No. 2018-00194 Central Maine Power Company, Request for New Alternative Rate Plan (ARP 2014), Docket No. 2013-00168, Order Approving Stipulation (Aug. 25, 2014). If those one-time adjustments were removed from rates as planned, the Company’s rate increase would grow to approximately $36.8 million. EXM-011-008. The schedule in this case initially called for Intervenor Testimony and the Bench Analysis to be filed by January 11, 2019 and the case to conclude on or about August 20, 2019. To accommodate the Office of the Public Advocate’s (OPA) request that it be allowed until April 2, 2019 to file testimony on its consultant’s analysis of plant additions going back to 2013, the schedule was modified and the decision date for the case was pushed back to approximately October 1, 2019. On February 22, 2019, the Staff filed its Bench Analysis (BA) in response to the Company’s Direct Case. Based on Staff’s traditional revenue requirement analysis, Staff found that the Company’s proposed revenue requirement deficiency should be reduced by approximately $20 million. In addition, based on Staff’s review of CMP’s customer service performance during the past several years, the Staff proposed a 75- to 100-basis-point downward adjustment to CMP’s Return on Equity (ROE), which further reduced the Company’s revenue requirement deficiency by $4.8 million to $6.5 million. On April 12, 2019, the Company filed its Rebuttal Case, which consisted of the following testimonies: Policy Rebuttal Testimony (Policy Reb.), Revenue Requirement Rebuttal Testimony, (Rev. Req. Reb.), Operations and Resiliency Testimony (Ops. and Res. Reb.); Return on Equity and Capital Structure Rebuttal (ROE Reb.); Customer Service Rebuttal Testimony (Cust. Serv. Reb.); and Operations and Capital Investment – Plant Additions. In addition, on May 15, 2019 the Company submitted Supplemental Reply Bench Analysis REDACTED 3 Docket No. 2018-00194 Rebuttal Testimony on SmartCare Implementation (SmartCare Reb.). The Company’s Rebuttal Case revises its proposed revenue requirement increase to $35.6 million. Table 1 below breaks down the $12.7 million increase from the Company’s initial filing: Table 1 Table 1 – Revenue Requirement Changes – Initial Filing to Rebuttal Dollars in Millions 1 Unmoderated Revenue Requirement Increase as filed October 15, 2018: $ 22.9 2 3 4 5 6 7 Vegetation Management Tier 1 Storm Adjust Rate Year to October 1, 2019 Electric Operations Additional Union Positions All other Total of Rebuttal Updates 8 Revenue Requirement Increase 5.2 4.1 2.9 1.7 (1.2) 12.7 $ 35.6 Policy Reb. at 5. Once again, the Company’s revenue requirement increase assumed that the one-time adjustments that would ordinarily come out of rates with the ACF process remained even after the costs associated with such items were recovered. If such items were removed from rates as part of the ACF, the Company’s distribution rate increase would grow to $49.55 million, or approximately 19%. EXM-011-004. On June 7, 2019, the Company filed a revenue requirement update which incorporated a number of changes that had been discussed at the technical conference on May 23, 2019, including the effect of removing the one-time ACF adjustments from rates on July 1, 2019, as well as several other minor changes. These changes to the Company’s proposed revenue requirement increase are set forth in the table below: Reply Bench Analysis REDACTED 4 Docket No. 2018-00194 Table 2 Central Maine Power Company Distribution “Post-Rebuttal” Changes to Unmoderated Revenue Requirement Dollars in Thousands 1 Unmoderated Revenue Requirement Increase at Rebuttal April 12th Impact of Final 2018 Allocation Factors on: 2 Payroll 3 Pension/OPEB/401k 4 Medical 5 Rate Base (reduced allocation of Common Plant) 6 Property Taxes 7 All other O & M (Items grown at inflation, not specific forecast) 8 Subtotal Allocation Factor Related $ 35,556 $ (350) (142) ( 63) (238) (289) (170) $ (1,251) 9 Remove One-Time Revenue Adjustments (July 2019 ACF) 10 Correct ASC 715 to reflect October Rate Year 11 Move CSS Savings Delay to July 2019 ACF 12 Revised NEB Estimate 13 Updated Vegetation Management ODR-012-029 14 Remove Customer Service Guarantee Payments ODR-012-025 15 Cash Working Capital Impacts of Above $ 13,950 304 (488) (1,512) 40 (20) (22) 16 Total Updates $ 11,001 17 Updated Unmoderated Revenue Requirement Increase $ 46,557 The Staff now files this Reply Bench Analysis, which provides additional analysis and preliminary recommendations in response to the Company’s Rebuttal Case. The Staff’s ultimate recommendations to the Commission will be contained in the Examiners’ Report scheduled to be issued on September 12, 2019. In providing this Reply Bench Analysis, Staff note that while many aspects of the Company’s Rebuttal Case were in response or rebuttal to the Bench Analysis, the December 20, 2018 Report of The Liberty Consulting Group (the Liberty Report), or testimony of the OPA, certain parts of Reply Bench Analysis REDACTED 5 Docket No. 2018-00194 the Company’s Rebuttal Case—such as the Company’s Electric Operations staffing proposal and its detailed Resiliency Plan—were brand new and not in response to any prior testimony. The Staff was forced to conduct its discovery and analysis of these aspects of the Company’s case within the compressed schedule established for rebuttal positions. II. POLICY A. Mitigation In the initial Bench Analysis, the Staff did not put forth a specific mitigation proposal. Instead, the Staff noted that the determination of whether mitigation through the acceleration of the amortization of regulatory liabilities was wise would depend on the revenue requirement that ultimately comes out of this process, the potential impact of any revenue requirement deficiency on current rates, and the impact that accelerated amortizations would have on future rates. In its Rebuttal Case, the Company expressed agreement with the Staff’s position on this issue, as well as with the OPA’s position that the mitigation should not be funded through an acceleration of the Cost of Removal Liability since that would lead to an increase in the Company’s depreciation rates. But the Company continued to believe that some rate mitigation was necessary in this case. Therefore, the Company proposed that any residential rate increase in this case be limited to the rate of inflation, which it estimated to be 2.21% in 2019. The Company did not, however, propose a specific mitigation mechanism in its Rebuttal Case. Given the size of the Company’s proposed increase to its revenue requirement, Staff is skeptical that it will be possible to implement mitigation in this case without Reply Bench Analysis REDACTED 6 Docket No. 2018-00194 imposing a significant burden on future ratepayers. The Staff is also concerned that a mitigation mechanism would tend to mask the real impact of the rate increase granted as well as the cost impacts of the Company’s proposals in this case. B. Rate Year/Timing of Rate Change As a result of the schedule change in this case, the Company has proposed that the rate effective year now be based on the revised expected date of the Commission’s decision October 1, 2019. The Staff does not oppose the Company’s proposed revision. C. Storm Cost Recovery 1. CMP’s Position In its initial testimony, CMP sought to lower the threshold for Tier 2 storms from $3.5 million to $1.5 million while maintaining the current amount in rates for Tier 1 storm costs ($4.0 million). Under CMP’s proposal, storms with costs exceeding $1.5 million would be considered Tier 2 storms and recovered through the storm cost reserve fund. In the Bench Analysis, Staff proposed to maintain the existing Tier 1 threshold but to increase the amount in rates for the Tier 1 storms based on historic costs. In its Rebuttal, CMP responded that it could accept the approach in the Staff’s Bench Analysis of maintaining the Tier 1 threshold and increasing the amount in rates for Tier 1 storms provided that (a) the baseline amount is adjusted to include all of CMP’s 2018 Tier 1 storm costs and (b) a “circuit breaker” mechanism is included (further described below). Including all of CMP’s 2018 Tier 1 storm costs in the calculation would increase the amount in rates for Tier 1 storms from $6.4 million, as proposed in the Bench Analysis, to $8.1 million. Policy Reb. at 13. Reply Bench Analysis REDACTED 7 Docket No. 2018-00194 In its Rebuttal, CMP reasons that while the Staff’s approach in the Bench Analysis would improve CMP’s ability to equitably recover costs over time, storm costs are variable, and inequities can occur year to year. CMP’s proposed “circuit breaker” mechanism is designed to mitigate significantly inequitable annual results for either CMP or its customers. The mechanism would establish a symmetrical dead-band of plus or minus 20% around the $8.1 million funding level included in rates. Any year’s Tier 1 storm costs that are higher or lower than the dead-band levels would be reconciled in rates the following year in favor of either CMP or its customers. Policy Reb. at 14–15. As CMP describes it: If CMP experiences Tier 1 storms costs in excess of $9.76 million (120% × $8,135,959) in any year, then the Company would defer and recover amounts in excess of this level through a rate adjustment in the next year. Conversely, if the Company experienced Tier 1 costs of less than $6.51 million (80% × $8,135,959) in any year, any excess recovery beyond this level would be deferred as a liability and returned to customers in a rate adjustment in the next year. Id. at 15. CMP also proposed to reset the Tier II and III storm reserve account to zero after December 31, 2019.1 Any reserve balance remaining at December 31, 2019, after sharing would be collected in the 2020 annual compliance filing. Id at 16. 2. Staff’s Reply Table 3 below presents the Tier 1 amounts expended by CMP during the 2014– 2018 period. 1The Tier II / III Storm Reserve balance at December 31, 2018 reflects net storm costs of $8.65 million. REDACTED 8 Reply Bench Analysis Docket No. 2018-00194 Table 3 Tier 1 Storm Costs ($000s) 1 2 3 4 5 6 Year 2014 2015 2016 2017 2018 TOTAL #Months 6 12 12 12 12 54 7 8 Per Month Average Annual Average Tier 1 $ 891 812 3,171 11,026 20,713 36,612 678 8,136 Staff accepts CMP’s proposal to include the full year of 2018 in the calculation of the annual average and thus include $8.1 million for Tier I storms in rates. Staff would note that this represents an increase of $4.1 million, or approximately double what was included in rates in Docket No. 2013-00168. The circuit breaker mechanism proposed by CMP would limit CMP’s exposure in years with numerous Tier 1 storms. CMP cites the storm totals—$11 million in 2017 and $20 million in 2018—experienced in the past two years as evidence of need for the Company to cap exposure. CMP also illustrates that in 2014 and 2015 ratepayers would have benefited from a circuit breaker mechanism. Given that the amounts in rates for Tier I storms will increase considerably, Staff believes that there is value in a mechanism that also has the potential to return some benefit to ratepayers. As noted in the Bench Analysis, the Commission in the past has encouraged the development of storm cost recovery mechanisms that are not binary in nature and that retain CMP’s incentives to control storm costs. BA at 102. Given these preferences, REDACTED 9 Reply Bench Analysis Docket No. 2018-00194 the Staff proposes to increase the dead-band to +/- 25% and to have the amounts over or under the dead-band be shared on a 50/50 basis between the Company and its ratepayers. Using the past four years of Tier I costs as an example and assuming a funding level of $8.1 million, the Staff’s circuit breaker approach would have resulted in both the Company and ratepayers sharing approximately $1.6 million over the period. The details of the sharing are set forth below. Table 4 Year Tier I Storm Cost Applicable Dead-Band CMP 50/50 Ratepayer 50/50 Storm Cost Sharing ($’s in Thousands) 2015 2016 2017 812 3,171 11,026 6,102 6,102 10,170 2,645 1,466 (428) 2,645 1,466 (428) 2018 20,713 10,170 (5,272) (5,272) TOTAL (1,589) (1,589) Finally, Staff accepts CMP’s proposal to rebalance the storm reserve account based on the balance at the end of calendar year 2019. Depending on the magnitude of the balance, Staff may suggest mitigating the effect of the recovery of the reserve balance over a two-year period. D. Affiliate Service Charges 1. Company’s Position In its Direct Case, CMP proposed that an adjustment to the test year be made for charges from its affiliates, primarily Avangrid Service Corporation (ASC), that exceeded the current cap of $31.4 million for affiliate charges. CMP stated that CMP’s distribution share of the excess charges was approximately $5.4 million. Rev. Req. Dir. at 13. CMP stated that it would be providing a market rate study in this case to support its use of cost-based affiliate charges in accordance with the Company’s commitments initially Reply Bench Analysis REDACTED 10 Docket No. 2018-00194 made in Central Maine Power Company, et. al., Request for Approval of Affiliated Interest Transaction for Two Service Agreements with Energy East Management Corporation, Docket No. 2001-00178, and more recently made in CMP’s last rate case, Docket No. 2013-00168. The Company’s initial filing did not actually contain the study but noted that it had retained Thomas Flaherty from PwC to prepare the study. Policy Dir. at 11. The Company also requested that the Commission remove the current cap on affiliate service charges in this proceeding. In the Bench Analysis, the Staff noted that even if PwC’s market rate study was completed in sufficient time to be considered in this case, Mr. Flaherty has acknowledged that the study would answer only the question of whether the charges to CMP by ASC were reasonable when compared to market rates. It would not answer the question of whether those services are cost effective when compared to the utility itself performing the services. Staff stated that the lack of such analysis in PwC’s anticipated study, or in CMP’s filing in this case, made it difficult, if not impossible, to judge the reasonableness of CMP’s requested increase in affiliate service charges. Therefore, Staff recommended that CMP’s proposed adjustment to the test year not be accepted. In addition, Staff did not support the elimination of the cap on affiliate service charges since Staff believed that the cap provides a useful tool in monitoring the level and reasonableness of affiliate service costs that CMP incurs. In its Rebuttal Case, CMP disputes Staff’s proposed exclusion of the $5,376,000 of distribution-related affiliates service charges from revenue requirements. CMP argues that all of these charges up to the current cap reflect charges to CMP for actual services provided by ASC and Avangrid Management Company, LLC (AMC). These Reply Bench Analysis REDACTED 11 Docket No. 2018-00194 services are in necessary areas, including accounting, finance, legal services, and information and technology. The Staff bases its disallowance for these charges on its belief that CMP has not met its burden of proof to demonstrate the reasonableness of the charges that exceed the current cap. Policy Reb. at 26. The Company does not dispute the fact that the collaborative process to conduct the market study contemplated in Docket No. 2013-00168 was not completed before this case was initiated. The Company notes, however, that it did not initiate this rate proceeding (and thus the timing of the rate proceeding was outside of its control), and that the required market study will be submitted later in the proceeding, thus satisfying the Company’s obligation. The Company also argues that in the last two rate cases it has presented testimony that demonstrates the reasonableness of the Company’s charges. First, in Docket No. 2007-00215, the Company submitted testimony from Eric Stinneford and Paul Dumais asserting that the provision of services, such as treasury and accounting and payroll, by CMP’s service company affiliates resulted in significant benefits to customers. In addition, the Company’s benchmarking study submitted by PA Consulting Group in the Company’s last rate case demonstrated that CMP’s affiliate service charges were generally lower, or in line with, similar charges borne by other utilities. Policy Reb. at 29. 2. Staff’s Reply Several points need to be made in response to the Company’s argument that it has demonstrated the reasonableness of its affiliate service charges in past cases. The Company is correct that it did, in Docket No. 2007-00215, provide testimony on the Reply Bench Analysis REDACTED 12 Docket No. 2018-00194 issue of savings from shared services. But the Stinneford/Dumais testimony the Company referred to was never adopted or endorsed by the Commission. Moreover, at the time of that testimony, CMP’s affiliate service charges were capped at $25 million. Central Maine Power Company, Request for Approval of Affiliated Interest Transaction to Increase Dollar Limit for the Energy East Shared Services Corp. & Energy East Management Corp. Support Services Agreement with Certain Energy East Affiliates, Docket No. 2004-00435, Order Approving Stipulation (Oct. 25, 2004). Thus, even if the Commission’s inclusion of affiliate service costs in rates in Docket No. 2007-00215 can be seen as an endorsement that the costs for those services were reasonable, that does not mean that the additional costs the Company seeks to put into rates at this time, and which are in excess of the current cap, are reasonable. The Staff’s position in this case is that the Company has not demonstrated the cost-effectiveness of the incremental charges it wishes to include in rates. The costeffectiveness standard encompasses both a cost and a quality component. In other words, the Company should be required to show that the costs for the affiliate services it wishes to include in rates are less than what the utility’s cost would be for the same services if it did not purchase them from the affiliate. In addition, the Company should show that the savings are not being achieved by sacrificing the quality of services provided to Maine ratepayers. The other piece of evidence the Company cites to in its Rebuttal Case — the PA study submitted in Docket No. 2013-00168 — was also never accepted or adopted by the Commission. While the PA study purported to show that the service company charges incurred by CMP were in line with amounts charged by affiliates of other Reply Bench Analysis REDACTED 13 Docket No. 2018-00194 utilities, the PA study did not address the issue of whether the services provided by CMP’s affiliates were being done in a cost-effective manner. Nor did It — or could It — show that the additional service costs now being incurred by CMP which are above the cap are reasonable or cost effective. As discussed in Section III of this Reply Bench Analysis and Section III of the Staff’s initial Bench Analysis, CMP has had significant issues with its customer service since 2016. The drop-off in service seems to have coincided with CMP’s parent company’s merger with United Illuminating and the formation of Avangrid. Shortly after the merger, several core supervisory and management functions were transferred from the utility to the service company. The drop in the quality of customer service — as well as other conduct raised in this case, including the Company’s decision not to seek an increase in the affiliate service charge cap prior to (or at least close in time to) actually incurring costs in excess of the cap, and its sudden decision in this case at the rebuttal stage that it needed to address aging workforce issues and the backlog of more than 8,000 distribution-line inspection defects — are all troubling signs of what appears to be an indifference to the quality of service being provided to Maine ratepayers and to its responsiveness to the utility’s regulator. The Staff believes that these issues can best be addressed through a management audit of CMP, AMC, and ASC to determine whether CMP’s management structure, including how management services are provided by AMC and ASC, is appropriate and in the interests of Maine ratepayers. The Staff believes that this step is essential in determining whether the incremental service company costs that CMP now seeks to include in rates are just and reasonable. Reply Bench Analysis E. REDACTED 14 Docket No. 2018-00194 Management Efficiency Adjustment 1. Company’s Position In the Bench Analysis, the Staff proposed a 75- to 100-basis-point downward adjustment to the otherwise allowed Return on Equity (ROE) be made in this case based on the persistent and substantial nature of CMP’s failure to provide reasonable and adequate service to its customers. The Company responded to the Staff’s proposed adjustment by arguing that the proposed adjustment was unduly punitive, was unwarranted given the Company’s recent improvement in performance; and would likely be viewed unfavorably by CMP’s credit rating agencies. 2. Staff’s Reply The Staff continues to believe that its proposed ROE adjustment is warranted given the Company’s poor customer service performance starting in 2016, and that the proposed adjustment is consistent with past Commission precedent. See Emera Maine, Proposed Increase in Rates, Docket No. 2015-00360, Order Part II at 81–85 (Dec. 22, 2016); and Central Maine Power Company, Application for Fuel Cost Adjustment Pursuant to Chapter 34 and Establishment of Short-Term Energy Only Rates for Small Power Producers Less Than 1 Mw Pursuant to Chapter 36 (Investigation of Contracts), Docket No. 1992-00102, Order at 59 (Oct. 28, 1993). F. Revenue Decoupling Mechanism 1. Company’s Position In its February 22, 2019 Bench Analysis, Staff identified three concerns regarding a renewal of a revenue decoupling mechanism (RDM) for CMP. First, Staff noted that, Reply Bench Analysis REDACTED 15 Docket No. 2018-00194 due to the billing issues and errors that are currently under investigation, it may not be possible to set accurate revenue targets. Second, Staff noted its concern that an RDM could mask and mitigate (from the Company’s perspective) management errors related to under-billing, given that resulting revenue shortfalls would be automatically recovered from other customers by operation of the RDM. Third, Staff recommended that, if the RDM were to be renewed, it should be greatly simplified. Under the Staff’s suggested simplification, targets would be set and reconciled based on billing units rather than on revenue. BA at 116–20. In its Rebuttal Case, CMP provided “corrected” sales and customer forecasts, which it used to adjust its rate-year revenue projections and RDM targets. CMP also indicated its willingness to adopt the more simplified RDM structure suggested by the Staff. CMP proposed to use its forecast of customer growth 2 to establish the RDM targets, and then reconcile to the targets based on actual customer growth. Policy Reb. at 33. In the response to EXM-011-022 (Attachment 1), CMP provided a detailed example of how its RDM would operate. 2. Staff’s Reply Staff’s final recommendation on the RDM will be provided in the Examiners’ Report. However, at this point, Staff notes that the purpose of using forecasted customer growth to adjust the RDM targets is unclear, and the use of forecasted customer growth could introduce unnecessary complexity. Actual customer growth for 2 In its response to EXM-011-022, CMP included its customer growth forecast for the period 2020-2029, indicating that it is based on the “sales and load forecast submitted in Docket No. 2018-00194” EXM-011-022, Att. 1. Staff requests the Company to provide a cite to where the forecast was submitted. Reply Bench Analysis REDACTED 16 Docket No. 2018-00194 the prior calendar year would be known well in advance of when an RDM adjustment would be made, presumably on July 1 of each year, thus allowing for the RDM targets to be established and reconciled without the need to use any forecast of customer growth. III. CUSTOMER SERVICE A. Causes of Customer Service Issues 1. Company’s Position In its Bench Analysis, Staff extensively discussed customer service problems the Company had experienced since April of 2016 and which continued after its implementation of SmartCare in October of 2017 (these problems were also discussed in the Liberty Report). In its Rebuttal Testimony, the Company acknowledges that its customer service during the period following SmartCare implementation (i.e., beginning November 1, 2017) was not up to the Company’s standards. Cust. Serv. Reb. at 3. The Company goes on to explain that this poor performance was the result of extraordinary circumstances that occurred at or shortly after the time SmartCare went live on October 30, 2017. According to the Company, these extraordinary circumstances included: • Largest outage in Company history: A significant coastal storm, referred to as a “bomb cyclone,” devastated CMP’s service territory beginning on the evening of Sunday, October 29, 2017, producing very high winds and flash flooding (the “October 2017 Storm”). Reply Bench Analysis • REDACTED 17 Docket No. 2018-00194 Extreme cold weather: Beginning in November 2017 and continuing for much of the winter of 2017–2018, extreme cold impacted customers’ energy usage with many customers experiencing bills that were higher than they expected. New users of heat pumps seemed especially vulnerable due to confusion about their optimal operation in extreme cold. • Supply price increase: Electricity supply price increases contributed to higher winter bills. The Standard Offer price increased by 18% on January 1, 2018 for residential and small business customers. Competitive Electricity Providers also raised prices, seemingly without the knowledge of many of their customers. • Frequent, intense and sensational media coverage: Media attention began with a number of people who took to social media to voice frustration and grew to include regular news segments from investigative reporters. This media attention and resulting customer engagement drove increased calls and complaints. In fact, in the days following intense media coverage, CMP experienced year-over-year increases in volumes of calls to Customer Service Representatives (“CSRs”) of up to 80%. Cust. Serv. Reb. at 4–5. In its rebuttal, the Company added that while it planned for both an increase in customer inquiries associated with its newly-designed bill and longer calls as CSRs Reply Bench Analysis REDACTED 18 Docket No. 2018-00194 adapted to the new SmartCare system, it did not anticipate or staff for the increased customer engagement that resulted from these combined extraordinary events. The Company further stated that while these events have affected both customers and the Company and resulted in performance that is below the Company’s normal levels for some customers, this combination of events was indeed unprecedented and temporary. Id. at 5. The Company also provided a chronology of the eight formal communications discussed in the Bench Analysis that it received from the CASD between April 2016 and January 2018 in which the CASD expressed concern with the Company’s customer service. The Company explained that, among other things, it created a Reinforced Customer Service Quality Plan to focus on immediate actions to address the CASD’s compliance concerns. The Reinforced Customer Service Quality Plan included three areas of focused improvement, and 35 specific planned enhancements, in the categories of people, processes, and technology. Id. at 8. In addition to the Reinforced Customer Service Quality Plan, the Company explained other actions it took to address the customer service problems raised in the CASD communications. 2. Staff’s Reply The Company attributes the customer service problems discussed in the Bench Analysis to the October 2017 storm, the cold weather experienced during the winter of 2017–2018, the supply price increase implemented on January 1, 2018, and the intense media coverage of the Company’s customer service and billing problems. All of these Reply Bench Analysis REDACTED 19 Docket No. 2018-00194 events occurred at or shortly after the implementation of SmartCare in October 2017. The Company did not provide any explanation of the customer service problems it was experiencing prior to October 2017. The Company’s testimony showed that it reacted to each of the specific problems outlined in the CASD communications and explained what specific actions the Company took to address those specific problems, such as providing training to call-center staff. But the testimony also showed that the Company did not examine or identify the underlying cause of the customer service problems. At some point between the time the Company received the first letter from the CASD and when it received the last letter from the CASD, the Company should have acknowledged that it had some serious problems with its customer service that required further examination. Because the Company failed to acknowledge the underlying problem with its customer service after receiving the letters from the CASD and even after the implementation of SmartCare, it failed to conduct a root-cause analysis of these customer service problems. This allowed the underlying problem(s) to persist and worsen, resulting in the problems the CASD outlined in eight communications to the Company. In fact, the Company did not acknowledge it had a problem with its customer service until after Liberty issued its findings in December 2018. The Liberty Report identified multiple problems it had identified with the Company’s customer service and its implementation of the SmartCare system. Liberty Report at 73–98. In January 2019, Douglas Herling sent customers a letter acknowledging Liberty’s findings that (1) the Company did not dedicate enough staff, training or management oversight to ensure that subsequent customer service responses would be satisfactory, and (2) the Reply Bench Analysis REDACTED 20 Docket No. 2018-00194 Company failed to deliver the level of service that customers expected and deserved when the Company introduced SmartCare. This letter was sent a full year after the problems with SmartCare first became apparent and more than three years after the CASD first sent the Company a letter expressing concerns with the Company’s customer service. Had the Company acknowledged its customer service problems sooner, it could have, and should have, conducted a root-cause analysis of the problems. Had the Company taken such action, it could have mitigated and possibly avoided some of the customer service problems Liberty identified. B. Proposed Service Quality Metrics 1. Percent of Business Calls Answered Within 30 Seconds In the Bench Analysis, Staff recommended that a service-quality metric be established to measure the “percentage of calls answered by a live customer service representative” and that the benchmark for this metric be “80% of calls answered within 30 seconds” (80/30). Staff noted that this was the same metric and benchmark for call answer performance used in the Company’s last Alternative Rate Plan (ARP 2008). Staff also recommended that this benchmark and metric apply to both CMP’s business line and its contractor line and that performance be measured separately for each line. BA at 95. a. Company’s Position The Company stated in its testimony that it has been achieving a standard of 80% of calls answered within 45 seconds (80/45) since April 2013 and that the change to 45 seconds (from 30 seconds) was the result of Staff’s request for the Company to Reply Bench Analysis REDACTED 21 Docket No. 2018-00194 accommodate promotional activities associated with the Green Power Supply Option.3 The Company assumes that the same definition established in ARP 2008 for this service level metric would also apply to the proposed metrics. That definition was: Calls completed in the Company’s Integrated Voice Response System (IVR) + calls answered by agents in 30 seconds + total agent calls abandoned within 30 seconds ___________________________________________________________ Calls completed in the IVR + total agent calls answered + total agent calls abandoned Cust. Serv. Reb. at 38. The Company stated that it can agree to an 80/30 call-answer target proposed by Staff, but CMP will need more customer-service representatives (CSRs) to achieve that performance level. Id. The Company further stated that it conducted a workload analysis to determine the incremental headcount required to meet the proposed target of 80/30. The workload analysis determined that CMP would need another nine fulltime CSRs. This is an increase over the current target of 93 CSRs. Id. b. Staff’s Reply During the 2009 session, the Legislature enacted An Act To Establish the Communitybased Renewable Energy Pilot Program (Act), P.L. 2009, ch. 329. Part B of the Act required the Commission to arrange for a green power offer that is composed of green power supply and to ensure that the green power offer is available to residential and small commercial electricity customers. In recognition of CMP’s agreement to have its customer service department provide support in the implementation of the statewide green power offer, the Commission modified the Company’s business-calls-answered service-quality metric contained in its ARP so that the metric would be 80% of calls answered within 45 seconds for the remainder of the ARP. Central Maine Power Company, Chapter 120 Information (Post ARP 2000 Information Transmission and Distribution Utility Revenue Requirements and Rate Design, and Request for Alternative Rate Plan), Docket No. 2007-00215(II), Order of Modification at 1 (Apr. 10, 2013). 3 Reply Bench Analysis REDACTED 22 Docket No. 2018-00194 Regarding the formula for measuring the Company’s call-answer performance, Staff recommends that the Commission eliminate “calls completed in the IVR” from both the numerator and the denominator of the formula. As stated in the Bench Analysis, the purpose of this service-quality metric is to measure the “percentage of calls answered by a live customer service representative” (emphasis added). Calls handled by the IVR are not “answered by a live representative” and thus will always be “answered” within any designated timeframe. Therefore, including IVR handled calls does not add any qualitative aspect to the metric. Further, because the calls are not answered by a live person, including them in the formula serves no purpose other than to provide a cushion for the Company to answer fewer calls by a live person within the 30-second standard, thus diminishing the value of the metric itself and defeating its purpose. In response to questions by Staff, the Company explained that in its “workload analysis,” it used a goal of “65% of calls answered within 30 seconds” for calls handled by its Maine Call Center because the Company could still meet the 80/30 benchmark due to the inclusion of the IVR calls and calls handled by the outsourced vendor in the formula. Tr. 46 (May 21, 2019). According to the Company, 26% of all calls received by CMP in 2018 were by the IVR. ODR-012-018. This is a significant amount of calls that would automatically meet the call-answer metric and dilute the significance of the calls that were answered by a live person. Further, the Company testified that its “workload analysis” (which determined the need for nine additional CSRs in the Company’s internal call center to answer 80 percent of calls within 30 seconds) did not include calls handled by the IVR. Tr. 49–51 (May 21, 2019). Reply Bench Analysis REDACTED 23 Docket No. 2018-00194 It is the Staff’s opinion that customers should receive the same level of service whether the call is handled by an outsourced vendor or by the Company itself. The vast majority of calls where a customer needs to speak to a live person should be answered by that person within 30 seconds. Thus, it would be unnecessary and inappropriate to include IVR calls in the 80/30 call-answer metric. 2. Percent of Contractor Calls Answered Within 30 seconds a. Company’s Position The Company disagrees with Staff’s proposed metric to measure the answer rate for calls to the Company’s contractor line. The Company explains that it offers a contractor line as a courtesy to contractors and that it is a relatively small group of customers with unique needs from the Company and maintaining a separate contractor SQI target would create an inefficient use of resources because of the comparatively low volume of contractor calls. Cust. Serv. Reb. at 39. The Company further states that a requirement to maintain enough of the most seasoned CSRs to achieve 80/30 performance on the contractor line would lead to inefficient utilization of these most valued resources to the detriment of all other customers. The Company also stated its belief that an SQI metric for the “percent of business calls answered” that includes contractor calls is the best way to serve all customers because it will have a larger pool of representatives available to serve all customers’ needs. Id. b. Staff Reply Staff did not realize when it proposed this metric that callers to the Company’s business line are currently offered the opportunity to be forwarded to the “contractor line,” similar to customers that dial the contractor line directly. The Company explained Reply Bench Analysis REDACTED 24 Docket No. 2018-00194 that today, if a person calls the business line and selects the “contractor option” (option #4), the call is routed to the contractor line and handled by the small group of trained contractor CSRs. In the future, the Company plans to further train the call center CSRs so that they can handle those calls, thus eliminating the need for a separate “contractor line.” Further, because these calls will be received over the business line, they will be included in the “business line” call answering metric. Tr. 57–58 (May 21, 2019). Staff finds this to be a reasonable course of action and agrees to the elimination of the separately proposed call-answer metric for the Company’s contractor line with the understanding that the Company must properly train its CSRs to handle contractor calls and these calls must be included in the calls-answered metric for the Company’s business line. 3. Call-Abandonment Rate a. Company Position In the Bench Analysis, Staff recommended that a service-quality metric be established to measure the percentage of calls received by the Company’s business line that are abandoned before answering and that the benchmark for the metric be 7%. BA, Appendix B at 1. The Company states that it can agree to the proposed servicequality metric for calls abandoned and the associated benchmark of 7%, provided that the Commission approves an additional nine CSRs. The Company also agrees with Staff’s recommendation that it be allowed to exclude “major event days” as determined using the IEEE 2.5 Beta approach, from the calculation. The Company also suggested an exclusion in the event of a work stoppage. Cust. Serv. Reb. at 39. b. Staff Reply REDACTED 25 Reply Bench Analysis Docket No. 2018-00194 Staff’s position regarding the additional personnel is discussed above in section VI (C) (5). Staff does not agree with the Company’s suggestion that the SQI calculation should specifically exclude work-stoppage days. Rather, Staff suggests that the Company may petition the Commission in the event of a work stoppage that the Company feels will detrimentally affect its ability to answer customer calls. 4. Bill Error Rate a. Company’s Position In the Bench Analysis, Staff recommended that a service-quality metric be established to measure the percentage of bills the Company issues inaccurately each year, and that the benchmark for this metric be 0.4%. BA at 95 -96. This is the same metric that was contained in Bangor Hydro Electric Company’s last Alternative Rate Plan. Bangor Hydro Electric Company, Request for Approval of Alternative Rate Plan, Docket No. 2001-410, Order Approving Stipulation (June 11, 2002). Staff’s proposed Bill Error Rate metric considered a bill erroneous if: it contains an incorrect rate or charge; lacks a proper charge, fee or tax; no monthly bill is issued when one should have been issued; or the total amount due is not correct. Estimated bills are not considered erroneous if issued in compliance with Chapter 815, section 8(L) or a Commission Order approving an alternate meter read schedule. BA, Appendix B at 2. The Company argues that in Emera Maine’s Service Quality Index Annual Report for 2017, the data definitions and detailed calculations for the Bill Error Rate do not include delayed bills nor do they include estimated bills. The Company further states that Emera Maine’s Bill Error Index tracks the percentage of bills that are issued incorrectly to customers compared with the total number of customer bills issued, and Reply Bench Analysis REDACTED 26 Docket No. 2018-00194 that the bills in this calculation (a) are actually received by the customer and (b) do not include estimated bills. Cust. Serv. Reb. at 41. Based on the above, the Company recommends that the Commission adopt the following definition for a “bill error:” A bill is considered erroneous if it contains an incorrect rate or charge; lacks a proper charge, fee or tax; or the total amount due is not correct. The “bill error rate” will be calculated by dividing the number of erroneous bills issued in a calendar year by the total number of bills issued in the same calendar year. This calculation is based on actual bills issued and not on accounts. Id. The Company also noted that the inclusion of “delayed bills” in the Staff’s proposed Bill Error metric runs counter to the Company’s process for reviewing the accuracy of bills and thus should not be included in the Bill Error metric. The Company argues that its process for reviewing the accuracy of bills is designed to reduce bill errors and thus delayed bills created through this process should not be included in the Bill Error metric. Id. The Company also argues that estimated bills should be excluded from the Bill Error metric. The Company states that it has an established process for estimating bills and that an estimated bill does not represent a billing error. Id. The Company further proposes that the following errors or events be excluded from the Bill Error Rate metric: (a) Errors resulting from incorrect or misinformation received from a customer; (b) Errors resulting from incorrect or misinformation received from a supplier; (c) Storm events; and Reply Bench Analysis REDACTED 27 Docket No. 2018-00194 (d) New rate structures or mandated collections policies/plans (e.g., Interim Payment Policy) (delivery or supply) for the first two months of implementation. Id. at 41–42. Finally, the Company proposes that the benchmark for the Bill Error metric should be based on more current information and circumstances. The Company argues that Staff’s suggested benchmark was based on negotiations in Bangor Hydro’s 2001 Alternate Rate Plan proceeding and that, because the case was resolved through stipulation, there’s no documentation of the basis for the 0.4% benchmark. Given this uncertainty and the assertion that the Company has tracked customer-facing bill errors since the SmartCare system went live, the Company recommends a Bill Error Rate benchmark based on the most current average of the actual 12-month rolling Bill Error Rate performance, adjusted by one standard deviation above this average because the billing system is still relatively new. Id. at 42. The Company states that, based on the bill error information it has tracked, the average of the actual 12-month rolling Bill Error Rate performance from April 2018 to March 2019 is 0.58% and the standard deviation of this performance is 0.40%. Therefore, based on its proposed definition, the Company recommends that the benchmark be established at 0.98%. Id. at 42–43. b. Staff’s Reply Staff agrees with the Company that events identified as (a) and (b) above should be excluded from the Bill Error metric because these events are beyond the Company’s control. But Staff disagrees with the Company’s contention that events related to (c) Reply Bench Analysis REDACTED 28 Docket No. 2018-00194 and (d) should be excluded from the metric. With regard to (c), the Company argues that if a billing change needs to be made on a day when crucial staff are instead deployed to storm duty, and thus the change cannot be made and erroneous bills are issued, the Company would exclude that day. Tr. 65 (May 21, 2019). It is Staff’s position that storm events should not affect the issuance of bills because the bill issuance process is an automated process. Staff finds that the scenario that the Company poses — where a scheduled billing change is not possible because the resources necessary to implement the billing change are working on storm restoration— is implausible. Coding changes in the Company’s billing system for a pending billing change are typically completed well in advance of the actual date the change takes effect, and thus should not be affected by a storm. Further, the calculation proposed by the Company is asymmetric: it provides relief for days when the Company fails to issue a correct bill, yet it includes those same days when the Company issues accurate bills. With regard to (d), this event is within the control of the Company and is a necessary and common occurrence. Utilities need to have the ability to modify their billing system to accommodate new and modified rate structures from time to time and to issue accurate bills afterward. Thus, it is Staff’s position that items (c) and (d) should be included in the Bill Error metric. The Staff also disagrees with the Company’s position on the exclusion of delayed bills and estimated bills from the metric. The purpose of the proposed Service Quality Index is to help the Commission to determine when the Company’s customer service returns to reasonable and adequate levels so that the Company may receive relief from the ROE adjustment. Thus, the service-quality metrics need to qualitatively measure Reply Bench Analysis REDACTED 29 Docket No. 2018-00194 the specific areas where the Company’s customer service has been inadequate. The Company has had specific problems with issuing bills (delayed bills) and estimated bills. For this reason, it is appropriate for these two issues to be included in the Bill Error metric. Further, the Company’s argument that the inclusion of “delayed bills” runs counter to its “billing accuracy process” is unconvincing. The “billing accuracy process” is designed to catch bill problems before the bill is issued so that the error can be fixed and a correct bill issued. The failure to issue a bill or the issuance of an estimated bill is an indicator that the Company either has insufficient staff to work the billing exceptions being created by the system or that the system is creating excessive billing exceptions – both of which are problems the Company must address. For these reasons, Staff recommend that both delayed bills and estimated bills be included in the proposed Bill Error metric. Staff also disagrees with the Company’s benchmark calculation. The purpose of this Service Quality Index, including this particular metric, is to help the Commission decide when the Company’s customer service has returned to reasonable and adequate levels such that the Company may seek relief from the ROE adjustment. The period the Company recommends the benchmark be based on is a period during which the Company’s billing performance was not reasonable or adequate. Thus, a benchmark that uses the Company’s billing performance during this period would reflect unreasonable and inadequate service, defeating the purpose of the proposed Bill Error metric. The Company itself acknowledged this during a Technical Conference when the Company representative stated “admittedly 2018 is not going to be the best source of information, but it’s either 2018 or be a little of '17. And if we had more information now, Reply Bench Analysis REDACTED 30 Docket No. 2018-00194 the -- as each month goes on, we have a little bit better data and better determination.” Tr. 64 (May 21, 2019). Further, it would be inappropriate to add a standard deviation to the Company’s actual performance, which already represents unreasonable and inadequate service. Finally, if the Commission were to adopt the benchmark recommended by the Company, the Company would meet the metric at the time of the metric’s adoption because the benchmark represents the Company’s performance over the previous 12 months. Again, this would defeat the purpose of the metric, which is to create an incentive for the Company to improve the quality of its bill processing. In light of this, Staff continues to recommend 0.4% as a benchmark for the Bill Error metric. 5. Field Service Appointments a. Company’s Position In the Bench Analysis, Staff stated that the Company’s response to customer requests for field services has been inadequate. Staff thus recommended that the Company make a proposal for tracking the various types of customer requests for field services and ultimately establish a service-quality metric to measure the Company’s performance in responding to these requests. BA at 96. CMP stated that it agrees that the development of a metric to track the Company’s response to customer requests for field services is appropriate and that it commits to begin tracking the amount of time the Company takes to respond to different types of request for field services to develop an appropriate metric. But because the Company has not tracked this information historically, it cannot propose an appropriate baseline to measure its performance in this area. As such, the Company believes that a REDACTED 31 Reply Bench Analysis Docket No. 2018-00194 field-service response-time metric should not play any role in the Commission’s future determination of whether any downward ROE adjustment for management efficiency should be eliminated or reduced based on CMP’s customer-service performance. Pol. Reb. at 23. b. Staff’s Reply Because it would take at least a year to establish a meaningful benchmark for this proposed metric and then another year of “reasonable service” before the ROE adjustment could be lifted, Staff agrees with the Company that the Field Service Appointment metric proposed in the Bench Analysis should not be a part of the Company’s SQI. Instead, Staff proposes that the Company begin tracking its response times to field-service appointments as recommended by the Company and report its performance of this metric as part of its annual rate filing. C. Operation of the Service Quality Index 1. Company’s Position The Company agrees with Staff’s recommendation that service-quality metrics be adopted to aid the Commission in determining when CMP’s customer service in the areas of call-center operations and billing functions has returned to “reasonable and adequate levels,” and also agrees that the Company should be permitted to seek relief from any downward ROE adjustment based on its performance in the areas measured over any rolling 12-month period. The Company objects, however, to the “all or nothing” approach recommended by Staff for when the Company may seek relief from the ROE adjustment. Pol. Reb. at 24. Reply Bench Analysis REDACTED 32 Docket No. 2018-00194 The Company argues that the Commission should instead either (1) impose a Service Quality Index similar to the one contained in ARP 2008;4 or (2) retain the ROE adjustment, but allow the Company to seek immediate relief from some or all of the ROE adjustment as soon as its performance on any of the metrics over a rolling 12month period averages at or above the specified benchmark. In support of the first option, the Company states that the SQI contained in ARP 2008 was developed so that each service-quality metric was independently measured and any resulting one-year financial adjustment depended on how poorly CMP fared in meeting the metric. The Company states that this type of structure reflects the importance of meeting the service-quality metrics, but does not result in a full adjustment if only one metric is missed by a small margin. Id. Under the second option, the Company could receive immediate relief from some or all of the ROE adjustment as soon as its performance for any one of the metrics over a rolling 12-month period averages at least the specified benchmark. At a minimum, the Company argues, the Commission should allow it to reduce the downward adjustment by at least one-third for each of the three applicable metrics when the Company’s performance averages at least the specified benchmark for the applicable 12-month period.5 Finally, the Company proposes that any downward ROE adjustment remain in Central Maine Power Company, Chapter 120 Information Post ARP 2000 Transmission and Distribution Utility Revenue Requirements and Rate Design, and Request for Alternative Rate Plan, Docket No. 2007-00215, Order Approving Stipulation (July 1, 2008). 4 The Staff recommended the immediate adoption of four metrics, with a fifth metric being added at a future point in time. Thus, it is unclear why the Company references a downward adjustment of “one third” in its rebuttal. 5 Reply Bench Analysis REDACTED 33 Docket No. 2018-00194 effect only until CMP is able to meet the applicable customer service metric. Id. at 24 25. 2. Staff’s Reply As noted in the Policy section of this Reply Bench Analysis, Staff continues to believe that the ROE adjustment is the appropriate mechanism to address the Company’s inadequate and unreasonable customer service. Assuming that the Commission accepts this recommendation, the Company’s proposal to seek relief from the adjustment on a piecemeal basis is inconsistent with the intent of the SQI. The rate adjustment is intended to address the inadequate and unreasonable customer service discussed in the Liberty Report and in the Bench Analysis. The metrics are intended to aid the Commission in determining when the Company’s inadequate and unreasonable service has returned to reasonable and adequate levels. Accepting the Company’s recommendation that it be allowed to seek relief from the ROE adjustment if the Company meets any one of the metrics would mean that each component had a specific value in the Commission’s ROE adjustment. The Staff does not believe that this is the case. Rather, the adjustment reflects poor customer service, which needs to be addressed comprehensively rather than piecemeal. Therefore, the Staff continues to recommend that the SQI be met in its entirety before the Company could seek or obtain relief from the ROE adjustment. IV. OPERATIONS AND RESILIENCY A. Vegetation Management 1. CMP’s Position Reply Bench Analysis REDACTED 34 Docket No. 2018-00194 In its initial testimony, the Company proposed the continuation of its five-year cycle trim program along with the continuation of its Incidental, Hot Spot pruning and Enhanced Tree Trimming program on a somewhat expanded basis. CMP’s revenue requirement request was only slightly more than the $20.3 million which was currently in rates. At the time of its filing, however, the Company had not yet selected a contractor to perform the vegetation management services and had not finalized the amount for the cycle trim services or the details of its enhanced program. The Staff in its Bench Analysis accepted the Company’s proposed spending level with the understanding that the matter would be revisited once CMP had finalized the price and scope of services. In addition, as part of the Bench Analysis, the Staff recommended that the Company undertake a study of the current state of vegetation on its distribution system, similar to a study that was done by ECI for CMP in 2014, in order to assess the need for and effectiveness of any additional enhanced vegetation management programs. In its Rebuttal Case, CMP updated the pricing for its cycle trim and enhanced vegetation management programs based on the Company’s ongoing negotiations with its selected contractors. CMP notes that the Bench Analysis accepted CMP’s initial estimate for its vegetation management program subject to finalized pricing terms with the chosen contractors. For the cycle trim program and enhanced vegetation program similar in scope to the 2013- 2018 period, CMP includes [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] as compared to the [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] included in rates in Docket No. 2013-00168. REDACTED 35 Reply Bench Analysis Docket No. 2018-00194 CMP states this increase is a result of higher labor and fuel prices. Ops. and Res. Reb. at 4. In addition to the programs listed above, CMP also propose [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] for enhanced ground to sky vegetation management and hazard tree removal costs associated with the specific proposed resiliency investment circuits. CMP suggests a three-year recovery period to mitigate the rate impact. CMP proposes to symmetrically reconcile the vegetation management costs with what is included in rates in this case. The Company would defer any difference between the rate allowance and the actual costs on an annual basis which would be reconciled as part of the Annual Compliance Filing process. As proposed, the vegetation management deferral balance would accrue carrying costs and CMP suggests that any deferral balance be returned, or recovered, if the deferral balance exceeds a substantial level such as $5 million dollars. EXM-016-010. As part of CMP’s Rebuttal case, CMP notes its agreement with Staff’s proposal to have ECI do an updated Distribution System Workload Study. CMP states that this study would be done later this year. CMP, however, proposes that the enhanced ground to sky and the enhanced Hazard Tree Removal programs being proposed as part of the Company’s Resiliency Plan be allowed to go forward at this time. 2. Staff’s Reply a. Cycle Trim Previously, the Staff has expressed its support for the continuation of the fiveyear cycle trim program. The information in this case provided by CMP indicated that Reply Bench Analysis REDACTED 36 Docket No. 2018-00194 the contractors retained to perform the cycle trim during the next five-year cycle were solicited through a standard RFP process and were chosen through reasonable selection criteria. Therefore, Staff accepts CMP’s revised pricing of [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] annually to continue the five- year cycle trim program. Staff also accepts the [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] associated with shifting to an October 1 rate year. b. Ancillary Trim Program CMP has requested funding to maintain what Staff will refer to as the Ancillary Trim Program which consists of the Incidental, Hot Spot and Basic Enhanced Trimming programs as committed to in Docket 2013-00168. The revised cost of the Ancillary Trim Program is expected to be [BEGIN CONFIDENTIAL] [END CONFIDENTIAL]. This amount is based on an average of the 2014-2017 actual spend escalated for the higher prices included in the new vegetation management contracts. Ops. and Res. Reb. at 6. CMP did not include 2018 in the average as the enhanced program spending was significantly underspent as a result of storm events. The Staff does not oppose CMP’s proposal to include [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] for Ancillary Trim Program costs in rates in this proceeding. The Staff, however, believes that the 2018 under spend should be carried over and be 6 The vegetation management forecasts for calendar years 2019 and 2020 are $21.034 million and $22.228 million, respectively. The forecast for a rate year beginning July 1, 2019 would be one half of 2019 and one-half of 2020, or $21.631 million. Shifting the start of the rate-year to October 1, 2019 shifts the weighting of the calendar year inputs to one-quarter of 2019 and three quarters of 2020, or $21.930 million, an increase of $299 thousand. EXM 016-006. Reply Bench Analysis REDACTED 37 Docket No. 2018-00194 used by CMP as part of either the Ancillary Trim Program or as the Enganced Resiliency Plan Program. The Stipulation in Docket No. 2013-00168 provides that: “CMP commits to the continuation of the cycle trim and the distribution line inspection programs on a five-year cycle basis and to fund the Hazard tree, Hot Spot and Incidental work components of the vegetation management programs at the levels included in revenue requirements in this case. (See Attachment 2, page 5.)” Attachment 2, page 5 of the Stipulation sets forth the following levels of funding for the Ancillary Programs included in rates as follows: Table 5 Dollars in Thousands 1 2 3 4 Program Hot Spot Incidental Enhanced Line Clearance Total Ancillary Program Rate Year Amounts 1,029 895 1,800 3,724 Again, Attachment 2, page 5 notes CMP’s commitment to fund the Ancillary Program at the levels listed above. In 2016 and 2017 CMP’s underspending on the Ancillary Trim Program was $56,000 and $475,000 respectively. As part of the Stipulation which resolved CMP’s 2018 Annual Compliance Filing case, Docket No. 2018-00069, the Company agreed to use the cumulative 2016-2017 underspend as an offset to its otherwise recoverable October 2017 storm costs. In 2018, the Company’s underspending on the Ancillary Program continued and grew to $2,043,450. The Staff believes that, in accordance with the Stipulation in Docket No. 2013-00168, this underspending should be used to fund Reply Bench Analysis REDACTED 38 Docket No. 2018-00194 the Company’s basic enhanced trimming program in the upcoming year or used to fund a pilot of the Company’s Enhanced Ground to Sky program discussed below. CMP has acknowledged that it underspent on the Ancillary Program in 2018. Ops. and Res. Reb. at 5. However, CMP argues that in 2018 it spent more on the cycle trim program than what was included in rates because of the 2017 October Storm. In response, Staff would first note that the Stipulation in Docket No. 2013-00168 called for the Company to complete the five-year cycle trim program over the course of the 20142018 cycle period. The Staff has recognized all along that there may be yearly variation in the annual cycle trim spending but as long as CMP met its commitment to complete the program it would have met its commitment under the Stipulation. The Stipulation included $16,649,000 in rates for the cycle trim. With the exception of 2018, CMP underspent this amount every year and when the five years are looked at in total, CMP spent $79,963,000 on the cycle trim program while it collected $83,245,000 in rates. CMP has stated that it has trimmed all of its circuits in accordance with the fiveyear cycle trim schedule. Accordingly, the Staff believes that the Company has met its commitment under the prior rate case Stipulation and that it should be allowed to retain the savings it was able to achieve, $3,282,000, during the 2014-2018 period. However, the fact that the Company spent more on the cycle trim program in 2018 than it had in prior years, or the amount that was allowed in rates for that one particular year, does not excuse its significant underspending on its Ancillary Program in 2018. CMP should be required to apply its underspending to the Ancillary Program or to the Enhanced Ground to Sky Program during the coming rate effective year. c. Reconciliation of Vegetation Management Costs Reply Bench Analysis REDACTED 39 Docket No. 2018-00194 CMP’s proposed plan to reconcile vegetation management costs through the annual compliance process would reconcile all costs included for vegetation management. Since the implementation of the enhanced vegetation management program established in Docket No. 2013-00168, Staff has taken the position that funds for vegetation management need to be prudently spent removing threats to the reliability of the distribution system. Returning unspent funds to customers from this program is not the Staff’s preference. CMP has identified tree contact as the primary cause of outages. Through the annual compliance process Staff has been quite clear that it expects funds earmarked for the enhanced program to be used to remove trees. In order to ensure that spending on the enhanced program occurs, the Staff is willing to accept the Company’s Reconciliation Proposal as it concerns the costs for this aspect of the Ancillary Program. With regard to the Cycle Trim Program, the Staff believes that the costs and commitments associated with this aspect of the Company’s vegetation management program are sufficiently clear and that standard ratemaking can be utilized for this aspect of the program and that a reconciliation mechanism may add unneeded rate volatility. B. Resiliency Plan 1. Overview As part of its Operations and Capital Investment Testimony, the Company submitted what it referred to as its “Capital Resiliency Plan”. Ops. and Cap. Dir. at Exh. CAP-3. In it, the Company described the impact of Maine weather on the ability to provide reliable service and the likely impact that climate change will have on service interruptions. The Company stated that it endeavors to decrease the number of Reply Bench Analysis REDACTED 40 Docket No. 2018-00194 outages experienced by customers and the duration of those outages by developing a portfolio of programs with distinct projects that will harden the system and increase resiliency. Id. CMP provided the following cost estimates for the investments associated with its Resiliency Plan over the 2019-2020 time-frame. Table 6 Description 2019 Forecast 2020 Forecast Total $1,000 $5,000 $6,000 1 Circuit Segmentation 2 Feeder Ties 2,000 5,000 7,000 3 Circuit Upgrades 2,000 2,000 4,000 4 Incremental Automation 3,000 4,000 7,000 5 Total $8,000 $16,000 $24,000 As part of discovery, the Company was asked to provide CMP’s actual Resiliency Plan which formed the basis of Exhibit CAP-3. The Company responded that CAP-3 provided an overview of the Resiliency Plan. The Company stated that the Resiliency Projects were currently undergoing an internal review prior to finalizing projects and sequencing them into a ten-year plan. The Company stated that it expected to have more detailed project and circuit information by the first quarter of 2019. EXM-002-050. The Company was also asked to provide a cost-benefit analysis performed to develop the Resiliency Plan. The Company responded that it had not yet performed any formal financial cost-benefit analysis to develop CMP’s Resiliency Plan. The Company noted that while there was a preliminary projected impact on reliability as measured by circuit SAIFI, specific improvements will be determined after final actions of the different programs are defined. In addition, the Company noted that as the Reply Bench Analysis REDACTED 41 Docket No. 2018-00194 Company vets each planning level proposal into detailed design, a formalized costbenefit analysis will be performed. EXM-002-051. In the Staff’s initial Bench Analysis, the Staff recommended that the Company’s proposal to add $16 million of discrete “Resiliency Plant Additions” to rate base between the interim year and the rate effective year not be accepted since the Company’s proposal did not contain any details on the design, timing and amounts of the Company’s Resiliency Plan investments and, therefore, failed to meet the Commission’s “known and measurable” standard. In addition, the Staff noted that the Company’s description of the investments to be made under the program (automation, circuit tying and circuit segmentation) were already CMP programs and, therefore, the Resiliency Investments should be considered to be covered by the plant addition attrition allowance which added between $153.5 and $138.7 million to rate base during the interim and rate effective years depending on whether the Company’s or the Staff’s attrition proposals were accepted. As part of its Rebuttal Case, the Company submits its actual Resiliency Plan which identified the 12 worst performing circuits on the Company’s distribution system and proposed an Enhanced Ground to Sky and Hazard Tree Removal Program for such circuits. In addition, the Company’s Operation and Resiliency Rebuttal Testimony contains for the first time an “Electric Operations Staffing” proposal which would add 75 Electric Operations personnel to its Revenue Requirement in this case in order to address its aging workforce and Company workload requirements. The Staff addresses each of the components of the Company’s Operations and Resiliency proposals below. 2. Enhanced Ground to Sky and Hazard Tree REDACTED 42 Reply Bench Analysis a. Docket No. 2018-00194 Company Position In addition to the established vegetation management programs discussed above, CMP has requested an additional [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] to implement an enhanced ground to sky program and to more aggressively target hazard trees as part of its proposed Resiliency Plan. For 2020, CMP proposes to perform ground to sky and enhanced hazard tree removal on the 3phase portion of the 12 identified worst performing circuits. Ops. and Res. Reb. at 7. Specifically, CMP estimates the cost for conducting the additional ground-to-sky trim work to be [BEGIN CONFIDENTIAL] [END CONFIDENTIAL], and the cost of the Enhanced Hazard Tree Program to be [BEGIN CONFIDENTIAL] [END CONFIDENTIAL]. CMP also estimates that there would be an offsetting reduction in the lump sum cycle trim expense for these circuits of [BEGIN CONFIDENTIAL] [END CONFIDENTIAL]. The Company has proposed to establish a deferral for the net resiliency vegetation spending in order to minimize the rate impacts of the first year costs of the program and to amortize recovery of such costs over three years. Rev. Req. Reb. at 11. b. Staff Reply The Staff has several concerns with the Company’s additional resiliency related vegetation spending. First, as noted above, the Company’s proposal would cover 249 circuit miles. This represents about 6% of CMP’s three-phase distribution system. Based on the per mile costs to do the initial 12 circuits, CMP would need to spend approximately [BEGIN CONFIDENTIAL] [END CONFIDENTIAL] to do its entire system. Second, the Company assumes in its cost estimates for the Enhanced Reply Bench Analysis REDACTED 43 Docket No. 2018-00194 Ground to Sky Program that it will not need to expend additional money to perform the additional clearing above the current clearance zone after the initial cycle. It is not clear to Staff why additional ground to sky trimming costs would not be incurred on a going forward basis in order to prevent the over hanging branches, or canopy, from regrowing. As noted above, the Company has accepted Staff’s proposal to engage ECI to do a vegetation study of its current system and the effectiveness of its programs. It would seem logical then that before CMP commit to the enhanced ground to sky program and incur significant costs, it have the benefit of the ECI study. Finally, Staff would note that CMP has historically performed both ground to sky and hazard tree removal on its worst performing circuits through its existing vegetation management program. Using the 2014-2017 time-period, the Company spent an average of $1.5 million a year targeting trees on the worst performing circuits. EXM016-013 Attachment 2. As discussed above, in 2018, CMP underspent on its basic Ancillary Program by $2.1 million. To the extent that CMP believes that it would be beneficial to conduct a pilot program that could be done in conjunction with the ECI study, this level of spending when coupled with CMP’s existing spending on worst performing circuits should provide CMP with ample opportunity to conduct an effective pilot program. 3. 2019 and 2020 Resiliency Plan Investments a. CMP Position CMP determines that capital spending on the 12 identified worst performing circuits would be capped at $25.7 million during the 2019 to 2020 time period with $8.2 Reply Bench Analysis REDACTED 44 Docket No. 2018-00194 million to be spent in 2019 and $17.5 million to be spent in 2020. The overall capital spending is broken down between Hardening and Topology with Automation as follows: Table 7 Topology, with Automation Total Year Hardening Lines Substations Other Top. Automation CapEx 2019 $1.67 $4.81 - $1.05 $0.68 $8.20 2020 $3.90 $7.58 - $3.20 $2.80 $17.48 Ops. and Res. Reb. at 27. With regard to the hardening component of the program, CMP will replace target poles that meet either of the following two criteria: 0 Poles that have been in service for at least 75 years; and/or 0 Poles that fail inspections performed in accordance with Distribution Line Inspection (DLI) program. Cumulatively, the Company projects replacing 945 poles on the 12 selected circuits with 214 poles being replaced due to age (over 75 years old) and 731 poles being replaced based on DLI inspection results. The Company estimates spending approximately $15.6 million on such pole replacements in the 2019-2020 timeframe. Id. at 105-106. The other component of the infrastructure hardening program is the use of tree wire. The Resiliency Plan states: Tree wire will only be installed on a targeted basis as part of 2019-2020 Resiliency Plan and only if identi?ed during the detailed engineering and project planning phase as a prudent addition to the circuit-specific resiliency plan. Thus, tree wire will be applied based on observed field conditions. Reply Bench Analysis REDACTED 45 Docket No. 2018-00194 Ops. and Res. Reb. at OAR-REB-3, pgs. 19 and 20. The Company, however, has not projected any specific investment for Tree Wire as part of the 2019-2020 Resiliency Plan. With regard to the Automation and Topology aspect of the 2019-2020 program, the Resiliency Plan includes schematics of each of the worst performing circuits selected as well as planned automation and topology actions. In commenting on its actions with regards to Circuit 834D2 in Monson, CMP notes: CMP has not yet performed detailed engineering or field work to confirm this plan or any other plan. While CMP is confident that the general outline of each circuit plan will be achievable, it expects the precise design to be refined with the effort to begin planning the specific projects necessary to change the topology of the circuit. In the case of Circuit 834D2, it is also conceivable that load or other circumstances will change between now and 2023, when CMP begins to implement the major investments. CMP’s phased approach from initial plans to engineering plans to contracting for resources will also provide the flexibility needed to confirm the most optimal solutions and achieve engineering and construction efficiencies. Id. at 29. b. Staff’s Reply As discussed above, in the initial Bench Analysis, the Staff indicated that it was not inclined to recommend approval of Company’s proposed resiliency investments because the investments did not satisfy the “known and measurable” standard and that such investments could be seen as having been already funded through the attrition allowance provided in revenue requirements. Although the Company has supplied much additional and new information as part of its Rebuttal Case, the Staff continues to believe that the rationale for not making a specific adjustment to rate base for the Resiliency Plan investments is still valid and applicable. Reply Bench Analysis REDACTED 46 Docket No. 2018-00194 As discussed above, the Company itself has recognized that investment plans for the designated circuits are fluid and not final. In addition, it does not appear that any investments on the proposed projects have actually been made to date, nor is there a concrete plan as to when such investments will be made or when the projects will be completed. ODR-012-068. The Staff would note that the Commission recently addressed a similar type proposal in Emera Maine, Request for Approval of Proposed Rate Increase, Docket No. 2017-00198, Order (June 28, 2018) In that case, the Commission rejected Emera Maine’s proposal to include in rate base amounts for an “undesignated customer experience fund” on the grounds that it did not qualify as a known and measurable adjustment to the test year. Id. at 15-17. Similarly, the Company’s Resiliency Plan investments do not qualify as a “known and measurable” change to be allowed in rate base. In addition, the proposed resiliency investments can be seen as already being covered by the attrition allowance. During the past five years the Company has addressed worst performing circuits as part of its Reliability Improvement Plans which are regularly submitted as part of the Company’s ACF. During that period the Company spent $37 million on capital investments in the identified worst performing circuits. EXM-016-013. In addition, investments for automation and circuit tying circuits are not new types of investments but are made regularly to improve reliability. While it may be necessary for the Company to reprioritize certain investments in order to accommodate the specific investments that the Company contemplates making as part of its 20192020 Resiliency Program, it should be able to accomplish such a reprioritization as part of its capital planning process and within the amounts allotted for the attrition allowance. Reply Bench Analysis REDACTED 47 Docket No. 2018-00194 The Staff has several other concerns with the Resiliency Plan as filed with the Company’s Rebuttal Case that are worth noting at this time. First, is the Company’s proposal to replace poles at 75 years regardless of condition. In its Resiliency Plan, the Company states that: DLI Inspections are conducted within a year of circuit tree trimming and includes a quality check to ensure that tree trimming was performed in accordance with contractual specifications. Any equipment, including target poles, that is identified for repair or replacement is prioritized by criticality of failure and subsequently remedied by the Company. Ops. & Res. Reb. Exh. 3 at pg. 17. It is not clear to Staff why a pole which was regularly inspected and not considered in need of replacement would be based solely on its age, be automatically replaced. It is also not clear to Staff whether the Company plans to apply this new criteria to just the 12 worst-performing circuits targeted in the Resiliency Program or to all circuits on its system. If so, Staff would note that there are currently approximately 4,000 poles on CMP’s system which are 75 years or older. Based on CMP’s estimated cost of $6,000 per pole, this new proposal would result in an additional $24.0 million of investment. C. Electric Operations Staffing 1. Company’s Position As part of the Company’s Operations and Resiliency Rebuttal Testimony, the Company submitted an Electric Operations Staffing Plan which it states was developed subsequent to its initial October filing. Ops. and Res. Reb. at 10. The Staffing Plan has two components. The first is the aging workforce component under which the Company plans to add 32 new positions and eliminate the staffing reduction of 20 workers which was included in the Company’s initial filing. In support of this proposal, the Company REDACTED 48 Reply Bench Analysis Docket No. 2018-00194 notes that it has a number of employees that are already eligible for retirement and that number will substantially increase in Electric Operations in the next few years. Id. The second component of the Staffing Plan is to address the workload requirements. Under this component of the Staffing Plan, CMP proposes hiring additional workers to the Electric Operations staff. In support of this proposal, CMP has performed a preliminary assessment of workload requirements. Based on the initial results of this assessment, the Company proposes a net increase of 23 workers. The additional positions to be added by the Company’s staffing plan are set forth below. Table 8 Proposed Electric Operations Staffing Position 9/30/2019 9/30/2020 Total Lineworker Apprentice 12 12 24 Lineworker 1/C - 13 13 Substation Apprentice 4 4 8 Field Planner 6 - 6 Line Clerk 4 - 4 Total 26 29 55 Source: Rev. Req. Reb. at Exh. REB-RRP 3-18, Schedule D. 2. Staff’s Reply The Staff has a number of concerns with the Company’s current Staffing Proposal. As noted above, the Company initially, proposed the elimination of 20 Electric and Operations positions which were to be vacated. In its Rebuttal Case, the Company made a u-turn on this position and said that not only did these positions need to be filled but that it needed to add 32 additional positions based on the Company’s aging workforce. Reply Bench Analysis REDACTED 49 Docket No. 2018-00194 The issue of the Company’s aging workforce is not a new one. In the Company’s last rate case, Docket No. 2013-00168, the Company’s Revenue Requirement Panel testified: Over the next six years, CMP faces a significant challenge with its employee base and aging workforce. The Company will have approximately 33% of its workforce eligible for retirement in 2014 and 53% by 2019. CMP intends to manage its workforce challenge during ARP2014 by a combination of programs to ensure knowledge transfer, employee development, training and recruitment. The Company intends to hire a combination of qualified workers and entry level employees. The particular areas of focus are T&D line workers, substations maintenance testing, control center operators and engineering. Rev. Req. Test at 39, Docket No. 2013-00168 (May 13, 2013) Given the Company’s recognition of its aging workforce problem back in 2013, it is baffling why it would initially propose not to fill 20 Electric Operations positions and then, in mid-case, decide to address the issue by reversing positions on the initial proposal and proposing to add 32 more positions. The Company’s proposal to add 23 positions to address workload requirements also raises a number of questions. The Company’s proposal is supported by its Preliminary Resource Plan which was provided in response to EXM-016-014 and OPA010-024. The Resource Plan dated May 1, 2019 indicates that the need for the additional crews is driven by Distribution Line Inspection repair backlog, the overwhelming number of which (6,588 out of 7,702) are DLI 3s, the lowest repair priority. ODR-012-040. As noted in the Bench Analysis, it was Staff’s understanding that DLI 3 defects only needed to be addressed during the next inspection cycle. B.A. at 104. At the technical conference on the Company’s Rebuttal Case, the Company’s witness testified that DLI 3s in fact needed to be addressed within two years of detection. Tr. 57-58 (May 22, 2019). The Company’s response to ODR-012-067 Reply Bench Analysis REDACTED 50 Docket No. 2018-00194 indicates that the requirement for DLI 3s to be repaired within two years after the inspection has been in effect since 2008. ODR-012-067. If this is the case, and the defects were identified and categorized under the assumption that they would be addressed within two years the question becomes, why did the Company wait until now to address this issue. While the Staff has not had the opportunity to fully vet the Resource Plan Model, a preliminary review of the plan raises questions about the resource needs identified by the Company. First, the DLI 3 backlog includes 3,883 DLI 3s which are currently telephone company responsibility. While the Staff recognize that it would be advantageous if CMP could assume ownership and responsibility for these poles and that talks on this issue are ongoing, as of this date this has not happened. The Resource Plan Model also assumes resource requirements based on 2018 storm activity. As noted in section II (C), 2018 was a particularly bad year for storms. If a three year average was used for storm resource requirements, as it was for other resource requirements in the model, the amount of storm resource requirements drops from 66,236 hours to 50,866 which is the equivalent of about 13 positions. In addition, the Resource Plan assumes a backlog of 8,500 DLI repairs. In response to discovery, the Company has identified 7,702 DLIs past due. Based on the Company’s assumption of 16.5 man hours to repair a defect, ODR-012-035, the decrease in defects translates into the equivalent of 11 FTEs for one year. Finally, the Company has not quantified or included in revenue requirements the offsetting savings which would result from hiring the additional workers in terms of external crews which would no longer be needed to be retained for storms, and the Reply Bench Analysis REDACTED 51 Docket No. 2018-00194 overtime avoided by having additional workers available and the outage costs avoided by having these workers address defective poles. In deciding what part of the Company’s Electric Operations Staffing Plan should be accepted for purposes of setting rates in this case, the Staff is guided by the “known and measurable standard” for test year adjustments. As most recently discussed in Northern Utilities, Inc. d/b/a Unitil, Request for Approval of Rate Change Pursuant to Section 307, Docket No. 2017-00065, Order at 32 (Feb. 18, 2018) when making adjustments to the test year, the Commission has applied a strict known and measurable standard. “To be ‘known,’ any change [the] test year must be reasonably certain as to whether and when it will occur. To be . . . ‘measurable,’ the amount of change must be reasonably certain.” Camden and Rockland, Maine and Wanakah Water Companies, Proposed Increase in Rates, Docket No. 93-145, Order (Part II) at 9 (July 12, 1994). With regard to the Company’s position to fill the 20 vacant positions, through this reversal the Company is essentially going back to the staffing as embedded in the historical test year. Given the Company’s acknowleged difficulty in keeping up with its DLI repairs, the Staff accepts this change in position as both reasonable and also consistent with the test year. With regard to the new positions to be added, the Company has provided information that they have retained workers for 17 of the new positions. ODR-012-024. Given the Company’s aging workforce issues and DLI issues discussed above, the Staff believes that the costs associated with these new hires are reasonable, consistent with the known and measurable standard and thus should be included in revenue requirements. Reply Bench Analysis REDACTED 52 Docket No. 2018-00194 As to the base salaries for the filled positions, the Company would include in the revenue requirement wages that are consistent with the currently effective union agreement. Staff agrees with the proposed incremental Electric Operations salary levels for the allowable positions.7 CMP also, however, has included an amount for variable compensation, which it calculated as 6% of the forecasted payroll expense. The Company based the 6% amount on the maximum payout amount allowed under the union contract. However, the actual average payout for performance years 2016–2018 for the Electric Operations group was only 4%. ODR-012-063, Att. 1. It is Staff’s position that variable compensation should reflect actual historic payouts, and should thus be calculated as 4% of payroll. Finally, Staff notes that the transmission and distribution allocators as well as the O&M percentage used by the Company in REB-RRP 3-18 (calculating the revenue requirement for the incremental electric operations staff) are not consistent with the other payroll calculations shown in REB-RRP 3-9. Specifically, the Company updated the distribution wage allocator to 75.43% as noted in section V(A), but it did not update that allocator for the electric operations staffing. Similarly, in rebuttal testimony the Company stated that it has updated the O&M allocation percentage from 73.85% to 69.71%, but in schedule REB-RRP 3-18 the Company uses 71.56%. Rev. Req. Reb. at 5. These errors will need to be corrected. 7 In the response to ODR-012-045, the Company reports that the starting wage for the Line Worker Apprentices – Travel who started on June 3, 2019 is $26.71 per hour. This amount is the April 28, 2018 wage, however, rather than the April 28, 2019 wage of $27.51 per hour that the Company used in its revenue requirement calculation. See EXM-012-012, Att. 1 at 20. The Company will need to reconcile this discrepancy. Reply Bench Analysis V. REDACTED 53 Docket No. 2018-00194 COST OF CAPITAL In its Rebuttal Testimony, CMP’s return on equity (ROE) witness, Ms. Bulkley, responds to the Staff’s Bench Analysis and the testimony of the OPA’s ROE witness, provides a discussion of recent capital-market conditions, and summarizes her conclusions and recommendations. Ms. Bulkley’s testimony does not update the analysis presented in her Direct Testimony, which was based on market data from late August 2018. In rebuttal, she continues to support her initial analyses and recommendation of an ROE of 10.00% and an equity ratio of 55%. A. Market Conditions and Updated Analyses As the Staff has stated and as the Company’s ROE Testimony has emphasized, market conditions and expectations are fundamental to utility cost-of-equity analysis. The primary models the Commission relies upon—the discounted cash flow model (DCF) and the capital asset pricing model (CAPM)—are market-based analytical approaches that typically incorporate current market-related data. As such, the results of any analysis done at a point in time become unreliable as underlying credit and equity market conditions change. Thus, it is appropriate to consider market conditions prevailing as of the date of the analyses presented and to update those analyses as those conditions change. Although numerous factors influence ROE estimation models, interest rates—especially long-term rates—and equity-market conditions have considerable influence on ROE models. As illustrated in Figure 1 below, over the past three decades, average authorized ROEs for electric utilities have declined from above 12% in 1990 to 9.60% in 2018. At the same time, the yield on 30-year Treasuries has Reply Bench Analysis REDACTED 54 Docket No. 2018-00194 declined from almost 9% to approximately 3%, and the Dow Jones Industrial Average (DJIA) has increased tenfold from approximately 2,600 at the end of 1990 to almost 26,000 today. Figure 1 Electric and Gas Authorized ROEs More recently, and relevant to the analyses in this case, as shown in Table 9 below, long-term interest rates have declined since both the time of the Company’s ROE analysis and the date of the Bench Analysis. Of note, as of June 4, 2019, the yield on 30-year Treasuries was 42 basis points lower than at the time of the Company’s analysis and 44 basis points lower than on the date of the Staff’s Bench Analysis. REDACTED 55 Reply Bench Analysis Docket No. 2018-00194 Table 9 Selected Interest Rates 10-Year Treasury 30-Year Treasury Moody's Seasoned Baa Corporate Bond Yield (monthly) Moody's Seasoned Aaa Corporate Bond Yield (monthly) Company Testimony 8/31/2018 Staff Bench Analysis 2/13/2019 Staff Reply Bench Analysis 6/4/2019 Change 8/31 to 6/4 Basis Points 2.86% 3.02% 4.77% 3.88% 2.71% 3.04% 4.95% 3.79% 2.12% 2.60% 4.63% 3.67% (74) (42) (14) (21) Source: FRED, accessed on June 6, 2019. In contrast, as shown in Table 10 below, key measures associated with equity markets—the other key input to the DCF analytical model—are relatively unchanged since the Company’s analysis and the Bench Analysis. Table 10 Selected Equity Market Indices Dow Jones Industrial Average Dow Jones Utility Average S&P 500 Company Testimony 8/31/2018 Staff Bench Analysis 2/13/2019 Staff Reply Bench Analysis 6/4/2019 Change 8/31 to 6/4 Points 25,965 726 2,901 25,543 737 2,753 25,332 793 2,803 (633) 67 (98) Source: Yahoo!Finance, accessed on June 6, 2019. Given these changes, Staff has updated the DCF and CAPM analysis to reflect market data as of June 4, 2019. B. Proxy Group Selection As explained in the Bench Analysis, to select a comparable group of companies upon which to base its ROE analysis, Staff started with a group of domestic utilities classified by Value Line as Electric Utilities or Natural Gas Distribution Companies and Reply Bench Analysis REDACTED 56 Docket No. 2018-00194 then applied a set of screening criteria. In this updated analysis, Staff continues to rely on the same screening criteria and the proxy group from the Bench Analysis but reviewed recent news announcements for recent developments. Based on that review, Staff eliminated El Paso Electric Company due to recent merger activity. 8 Staff’s updated proxy group as used in this Reply Bench Analysis consists of the 21 companies listed in Table 11. 8 El Paso Electric to be bought by an infrastructure fund for $2.78 billion, Reuters, June 3, 2019, https://www.reuters.com/article/us-el-paso-electric-m-a-infrastructure-i/el-pasoelectric-to-be-bought-by-an-infrastructure-fund-for-2-78-billion-idUSKCN1T41K8. REDACTED 57 Reply Bench Analysis Docket No. 2018-00194 Table 11 Staff’s Updated Proxy Group Company Alliant Energy Corporation LNT Ameren Corporation AEE Atmos Energy Corporation ATO CMS Energy Corporation CMS Consolidated Edison, Inc. ED DTE Energy Company DTE Edison International EIX Evergy, Inc. EVRG Eversource Energy ES IDACORP, Inc. IDA New Jersey Resources Corporation NJR NiSource, Inc. C. Ticker NI Northwest Natural Gas Company NWN NorthWestern Corporation NWE ONE Gas, Inc. OGS PNM Resources, Inc. PNM Portland General Electric Company POR PPL Corporation PPL Public Service Enterprise Group, Inc PEG Spire, Inc. Xcel Energy Inc. SR XEL Constant-Growth Discounted Cash Flow As done in the Bench Analysis and consistent with Commission practice and orders, Staff uses a constant-growth DCF approach to the cost of equity analysis. In its simplest form, a DCF estimates the cost of equity capital using the following formula: Reply Bench Analysis REDACTED 58 Docket No. 2018-00194 K = D/P + g where: K = cost of equity capital D/P = dividend yield g = long-term expected growth rate The updated DCF calculations are provided in Exhibit RBA-ROE-1. To summarize those calculations: the current quarterly dividend for each utility as of June 4, 2019 was converted to a forward dividend using the analysts’ estimate of future growth. Staff calculated the dividend-yield component of the model by dividing the resulting forward dividend by the share price for each utility, using a 50-day moving average of closing share price and a 200-day moving average of closing share price for each utility as reported by Yahoo! Finance on June 4, 2019. The growth component for each utility is the analysts’ projected 5-year growth in earnings per share, also as reported by Yahoo! Finance on June 4, 2019. This updated analysis produces an indicated ROE that ranges from 4.94% to 10.05%, with a midpoint of 7.50%. One of the criticisms in Ms. Bulkley’s Rebuttal Testimony is that Staff gave no consideration to whether the DCF results are reasonable estimates of the cost of capital. ROE Reb. at 19. Specifically, Ms. Bulkley questions the ROE range presented by Staff in the Bench Analysis for its failure to eliminate individual results below 7.00%, a result sufficiently above the cost of debt to be reasonable. To address that criticism, in Exhibit RBA-ROE-2 Staff replicates its DCF analysis and excludes any individual company result less than 7.00%. This analysis produces an indicated ROE that ranges from 7.10% to 10.05%, with a midpoint of 8.58%. Reply Bench Analysis REDACTED 59 Docket No. 2018-00194 Finally, in her Rebuttal Testimony, Ms. Bulkley states that “[r]easonable adjustments can be made to the individual company indicated ROE results produced by Staff, by applying the lower thresholds relied on by Dr. Griffing and myself that would result in DCF returns in the range of 9.71 percent to 9.95 percent.” ROE Reb. at 25 and Ex. REB-AEB-1. In the attached Exhibit RBA-ROE-3, Staff updates the analysis presented in Ms. Bulkley’s Rebuttal Testimony to reflect current market conditions and data. As appropriate for individual companies in the proxy group, Staff updates the current dividend, current growth estimates, and the closing market-price information. Staff also eliminates El Paso Electric from this analysis. This updated analysis currently produces ROE midpoints of 8.39% to 8.59%. To summarize, incorporating current market data into the DCF analyses supports a determination that an appropriate ROE falls within the range of 7.10% to 10.05%. Prior to any adjustment for flotation costs, which is discussed below, the midpoint of the ROE range is approximately 8.60%. D. Capital Asset Pricing Model As the Commission has previously recognized, results from a CAPM analysis provide a useful check on the DCF analysis. The general form of the CAPM model is: REDACTED 60 Reply Bench Analysis Docket No. 2018-00194 K = Rf + β (Rm – Rf) where: Rf = risk free rate Rm = return on market β = beta Rm – Rf = market risk premium As a check to the results of the updated DCF analysis, Staff has also made adjustments to selected inputs to the CAPM analysis to reflect the change in prevailing interest rates, the 30-year Treasury rate used as the risk-free rate, and changes in the Betas for the proxy-group companies as reported by Value Line. As shown in Table 12 below, those adjustments reduce the mean ROE indicated by the CAPM analysis to slightly above 10%, a reduction of 42 basis points from the date of the Bench Analysis and 159 basis points from the date of the Company’s initial testimony. Table 12 Updated CAPM Results Company Testimony 8/31/2018 Staff Bench Analysis 2/13/2019 Staff Reply Bench Analysis 6/4/2019 Change 8/31 to 6/4 Basis Points CAPM ROE Result - Low CAPM ROE Result - High CAPM ROE Result - Mean 11.30% 11.47% 11.67% 9.15% 11.59% 10.50% 8.40% 11.51% 10.08% (290) 4 (159) 30-year Treasury yield - 30 day average 30-year Treasury yield - one day Average Valueline Beta 3.05% 3.02% 0.676 3.04% 3.04% 0.633 2.80% 2.60% 0.605 (25) (42) (710) Consistent with the Commission's preference and the methodology used in the Bench Analysis, Staff uses a current Treasury rate rather than a forecast of interest rates. Staff calculates the 30-day average of the 30-year Treasury rate as of June 4, Reply Bench Analysis REDACTED 61 Docket No. 2018-00194 2019 and uses 2.80% as the risk-free rate. In her Rebuttal Testimony, Ms. Bulkley criticizes Staff’s reliance on current interest rates and asserts that it is important to consider investors’ expectations on interest rates and prospective market conditions and their effect on the CAPM results. ROE Reb. at 26–27. More generally, Ms. Bulkley states that investors expect continued increases in interest rates on both government and corporate utility bonds over the next few years. Id. at 94–95. Staff continues to have the same concerns as expressed in the Bench Analysis about the reliability of forecast interest rates, even in the near term. As an illustration, in her Rebuttal Testimony, Ms. Bulkley presents a graphic indicating that, by Q2 2019, the 30-year Treasury rate would be above 3.00%, the 10-year Treasury rate would be slightly below 3.00%, and the yield on Moody’s A Utility Bonds would be well above 4.00%. Id. at 95, Fig. 15. But, as noted in Figure 1, as of June 4, 2019, the 30-year Treasury rate was 2.60% and the 10-year Treasury rate was 2.12%. As of June 13, 2019, the yield on Moody’s A Utility Bonds was 4.00%.9 Staff thus continues to believe that the current risk-free rate should be used in the CAPM analysis, and therefore does not incorporate interest rate forecasts. E. Capital Structure The Company did not provide additional analysis of the proposed capital structure in its Rebuttal Testimony, and Staff does not update the capital structure analysis in this Reply Bench Analysis. Staff does note, however, that its recommended 50% common equity ratio is consistent with common equity ratios authorized in other Moody’s Analytics, Credit Trends, accessed June 13, 2019 https://credittrends.moodys.com/. 9 Reply Bench Analysis REDACTED 62 Docket No. 2018-00194 jurisdictions in 2017, 2018, and 2019. As noted by Regulatory Research Associates: To offset the negative cash flow impact of federal tax reform, many utilities sought higher common equity ratios, and the average authorized equity ratios adopted by utility commissions in the first three months of 2019 were modestly higher than the levels observed in 2018 and 2017. The average authorized equity ratio for electric utility cases nationwide was 49.51% in the first three months of 2019, 49.02% in full year 2018 and 48.90% in 2017. ... The aforementioned averages include allowed equity ratios adopted by utility commissions in…jurisdictions that authorize capital structures that include cost-free items or tax credit balances. Excluding these jurisdictions, the average authorized equity ratio for electric utility cases nationwide was 50.86% for the first three months of 2019, 50.53% in cases decided during 2018 and 50.02% in 2017. RRA Regulatory Focus, Major Rate Case Decisions—January–March 2019, April 11, 2019. F. Flotation Costs As noted in the Bench Analysis, the Commission has permitted a flotation cost adjustment in prior rate cases. Staff has not updated the flotation-cost adjustment from the Bench Analysis and continues to accept the flotation-cost adjustment of 11 basis points as presented in Ms. Bulkley’s initial testimony. G. ROE and Weighted-Average Cost of Capital (WACC) Updating the analyses to incorporate current market data produces DCF results that fall within a range of 7.10% to 10.05%, with a midpoint of approximately 8.60%. The updated CAPM analysis produces a range of 8.40% to 11.51%. Staff does not believe CMP’s risk profile is sufficiently different from that of the proxy group to warrant moving the Company’s ROE away from the midpoint of the DCF range. The addition of the adjustment for flotation costs results in Staff’s recommended ROE of 8.75% and a pre-tax WACC of 8.28%, as shown in Table 13 below. Reply Bench Analysis REDACTED 63 Docket No. 2018-00194 Table 13 Staff’s Recommended Capital Structure, Costs, and ROE Capitalization Percentage Common Equity Preferred Stock Long Term Debt Short Term Debt 2.82% 100.00% Total VI. 50.00% 0.02% 47.16% Cost 8.75% 6.00% 4.45% 3.50% After -Tax Weighted Pre-Tax Weighted Cost Cost 4.38% 0.00% 2.10% 6.08% 0.00% 2.10% 0.10% 0.10% 6.57% 8.28% REVENUE REQUIREMENT A. T&D Allocations In its Bench Analysis, the Staff proposed that the 2018 allocators—which allocate costs between transmission and distribution when they are not directly charged to one or the other—be used to better reflect the rate-year split between transmission and distribution costs. BA at 9–10. CMP concurred with Staff’s approach and updates its allocation factors in its Rebuttal Case. Rev. Req. Reb. at 1. At the May 23, 2019 technical conference, the Company presented updated allocation factors based on its finalized FERC Form 1. Tr. 50–51 (May 23, 2019). These allocation factors were incorporated into the Company’s updated Revenue Requirement calculation submitted on June 7, 2019. The finalized 2018 distribution allocation factors as presented by the Company are: Customer – 50.78%; Wage – 75.43%; Plant – 42.46%. These allocators are consistent with the data presented to the Staff in the Company’s annual transmission update. Staff accepts these 2018 allocators to calculate the rate-year split between transmission and distribution costs. REDACTED 64 Reply Bench Analysis B. Docket No. 2018-00194 Rate Base 1. Plant Addition Attrition Analysis a. CMP’s Position As CMP noted in its Direct Testimony, the attrition technique is a trend analysis that calculates a compound annual growth rate (CAGR) based on historical plant balances, adjusted for atypical additions. Rev. Req. Dir. at 26. CMP used its June 30 plant balances from June 30, 2013 through June 30, 2018 to arrive at a 5-year CAGR of 4.54%. CMP then applied this CAGR to the June 30, 2018 plant balance to arrive at its proposed plant additions of $75.1 million for the transition year (12 months ending June 30, 2019) and $78.5 million for the rate effective year (12 months ending June 30, 2020). In the Bench Analysis, Staff excluded 2018 plant additions—approximately $97.8 million—because they were significantly higher than plant additions from previous years and were considered anomalous. This adjustment by Staff resulted in an estimated 4-year CAGR of 4.11%. In its Rebuttal Case, CMP states that it did not agree with Staff’s “asymmetrical” approach to determining plant trend rates, where Staff only excluded as anomalous plant additions that appeared to be high when compared to other years. Rev. Req. Reb. at 3. In addition, as part of its Rebuttal, CMP updates its plant trend analysis by first modifying the study year to a 12-month year ending on December 31 and then updating the analysis to capture actual plant balances through December 31, 2018. Id. The resulting 5-year CAGR was 4.26%, which is lower than the 4.54% CAGR initially proposed by CMP but higher than the 4.11% CAGR proposed by Staff in the Bench Analysis. Reply Bench Analysis b. REDACTED 65 Docket No. 2018-00194 Staff’s Reply Staff has reviewed CMP’s revised calculation and believes CMP’s recalculation of the CAGR on a calendar-year basis is reasonable. In addition, based on our review of the year-over-year levels of plant additions, Staff also accepts the inclusion of 2017 (now the highest year in the analysis) given the overall dispersion of investment spending over the years: several low years, a few mid-level years, and one high year. Staff, however, believes that one modification to CMP’s analysis is appropriate. In response to a data request from the Staff, CMP provided the amount of distribution plant additions placed in service for the following projects: (1) Maine Power Reliability Project (MPRP), (2) Midcoast Reinforcements, (3) Berwick Reinforcements, (4) Meadow Road, and (5) the Brunswick/Topsham Brightline Project. EXM-009-001. Based on the Company’s responses, Staff believes that two specific projects—the $12.1 million of investment costs associated with the MPRP and Brunswick/Topsham Brightline projects—should not be included in the trend analysis as these projects are extraordinary, one-time investments. In the case of MPRP related costs, the Staff notes that the MPRP was a $1.1 billion, once-in-a-generation, statewide transmission project. The Brightline costs are also extraordinary since this investment was a result of a NERC reclassification of how bulk transmission was defined—another rare event. The exclusion of these extraordinary items results in a 5-year CAGR of 4.11%, which is identical to the CAGR proposed by Staff in its initial Bench Analysis. The results of Staff’s analysis are presented in Table 14 below. REDACTED 66 Reply Bench Analysis Docket No. 2018-00194 Table 14 Staff’s Attrition Calculation ($ millions) Distribution Oct-17 Storm Plant 6 12/31/2013 12/31/2014 12/31/2015 12/31/2016 12/31/2017 12/31/2018 7 5 Year CAGR 8 9/30/2019 9/30/2020 1 2 3 4 5 9 1,159.0 1,213.3 1,234.4 1,296.6 1,390.7 1,425.7 (24.7) (24.7) G&I Plant 272.6 275.9 301.7 323.6 408.9 438.6 ECC Additonal G&I to CRM&B Transmission Removals (b) (18.6) (18.1) (18.1) (16.3) (17.2) (17.2) (55.5) (55.5) (49.8) (51.1) (57.7) (73.3) (81.7) (87.3) (12.1) Total 1,363.1 1,420.0 1,460.3 1,530.7 1,620.5 1,667.5 Plant Additions 56.8 40.4 70.3 89.9 47.0 4.11% (a) 1,718.9 1,789.6 51.4 70.7 (a) only 9/12's of the trend rate was applied to 12/2018 balance to derive the 9/30/2019 balance. (b) Consists of (as reported in EXM-009-001) $ 10.4 MPRP (Lewiston) 1.7 Brunswick/Topsham Brightline $ 12.1 2. . Resiliency Investment See discussion in IV (B). 3. Power Tax a. Company’s Position CMP has recorded a regulatory asset for the effect of its implementation of the Power Tax software in 2011 (the Power Tax Regulatory Asset, or PTRA). The PTRA has been the subject of an ongoing separate review by consultants hired by the Commission to validate CMP’s calculation of the dollar amounts. In its Rebuttal Testimony, CMP states: “The Company maintains its position that the Commission must provide recovery of the PTRA consistent with the PLR in order to avoid a Normalization REDACTED 67 Reply Bench Analysis Docket No. 2018-00194 Rule violation.” Tax. Reb. at 6. CMP estimates that the PTRA balance as of September 30, 2019 will be $14.142 million. Rev. Req. Reb. at 17. b. Staff’s Reply CMP estimates that the PTRA includes $4.023 million related to the 2014 depreciation inconsistency. However, the depreciation inconsistency was not the result of the implementation of the Power Tax software. Instead, it was the result of CMP’s error in calculating deferred tax amounts for the book-tax timing differences after the revised depreciation rates were set in Docket No. 2013-00168. Staff believes that CMP is not entitled to collect this amount from ratepayers because any under-recovery resulted from CMP’s error, and the Commission will be correcting the error at its first opportunity to do so—i.e., this rate proceeding. Staff believes that this is consistent with the IRS’s normalization rules and the Private Letter Ruling (PLR) that addressed the need to establish the PTRA. As to the remainder of the PTRA, Staff notes that the final decision on the PTRA will be made in Docket No. 2016-00035 after the Commission’s consultants have concluded their review. Once this decision is made, the appropriate amount of the PTRA can be included in the Company’s rate base. In any case, Staff recommends that the $4.023 million related to the depreciation inconsistency be excluded from the PTRA at this time. C. Expense 1. Tax a. Company’s Position Reply Bench Analysis REDACTED 68 Docket No. 2018-00194 In its Rebuttal Testimony, CMP makes the following three changes to its calculations of excess accumulated deferred income taxes (Excess ADIT) and average rate assumption method (ARAM): • Inclusion of the final 2017 tax return data for fixed assets, including fixed-asset additions and depreciation. • Adjustment of the depreciation rates to remove the portion of the rate related to cost of removal. CMP considers cost of removal to be an unprotected deferred tax balance whose inclusion could potentially result in a violation of the normalization rules. CMP stated that it adjusted the book depreciation rates to those approved in the Stipulation in Docket No. 2013-00168, which represents the most recent figures, and excluded cost of removal from the book rates in the current protected excess deferred tax reversal schedule. • Amortization of protected deferred tax assets. CMP stated that the book and tax vintage cost data CMP used in the original calculation required additional review and alignment of vintage data. This update resulted in a new ARAM schedule for protected excess ADIT. The new ARAM for protected deferred tax assets occurs over a 31-year book life. CMP’s Rebuttal Testimony did not indicate what the effect of each of the changes above were on the revenue requirement. CMP also notes in its Rebuttal Testimony that it had determined that its unprotected, non-plant-related Excess ADIT balances needed adjustment. CMP adjusted both the unprotected Excess ADIT regulatory asset and regulatory liability, increasing the regulatory asset from $59.5 million to $106.7 million and increasing the Reply Bench Analysis REDACTED 69 Docket No. 2018-00194 regulatory liability from $51.1 million to $59.9 million. CMP stated that the unprotected Excess ADIT increase was recalculated during “the provision-to-return year-end process to account for the unfunded temporary differences that had previously been included in the total excess deferred tax balance calculated in the 2017 financial statements”. Tax. Reb. at 4. CMP also states that the process identified that in its October filing the Company determined the unprotected excess amount by calculating the difference between the total (funded and unfunded) unprotected balance and only a portion of the unfunded balance, rather than by calculating the difference between the total unprotected balance and the entire unfunded balance. CMP states that removing the entire unfunded difference from the calculation was necessary because CMP had already reduced its unfunded regulatory asset for the unfunded Excess ADIT decrease. CMP also proposes to amortize the unprotected Excess ADIT liability over a fiveyear period, instead of a 10-year period as originally proposed, if it is not used for rate mitigation. b. Staff’s Reply First, Staff is concerned about the recurring changes in the Excess ADIT balances, the classification of such balances, and the proposed amortization periods. It is understandable that changes are necessary, but even at the May 21, 2019 technical conference CMP indicated that there could be more updates to these amounts in the future. At the May 26, 2019 Technical Conference, Company Witness Beeber stated “I’m prepared to say that there could be an update. There may be a portion of the amount that would change. Could be more than insignificant, but I don’t have an exact amount.” Tr. 151 (May 21, 2019). Staff is hesitant to include an amount in revenue Reply Bench Analysis REDACTED 70 Docket No. 2018-00194 requirements—an amount that will remain there for the foreseeable future—when even the Company cannot say that the amount is final. Staff does agree that it would be appropriate to adjust the Excess ADIT to the final 2017 tax return amounts. Unfortunately, it is not clear from CMP’s filings what portions of the changes proposed are due to just this factor. In response to EXM-013-003, CMP stated: Subsequent to the October 2018 filing, CMP and PwC updated the Company's Excess ADIT calculations to replace preliminary 2017 book and tax data with final 2017 tax return data. During this update the Company determined that there were some routine changes in the protected and unprotected balances. These changes were generally attributable to provision to return adjustment and impacted plant and nonplant. During this same update, the Company also determined an additional adjustment was required to remove an unfunded plant temporary difference inadvertently included in the Excess ADIT remeasurement. This latter adjustment was the primary driver of the increase in the recoverable balance of Excess ADITs. Attachment 1 provides a calculation of the change in the unprotected components. The adjustment to remove the unfunded excess inadvertently included in the funded unprotected calculation was equal to the unfunded excess adjustment. The Company and PwC are re-evaluating whether this adjustment should have equaled the entire unfunded adjustment. The Company is aggressively working to conclude its re-evaluation if this issue and will supplement this response upon concluding this analysis. Staff notes that if these amounts had been inadvertently included elsewhere in the calculation, the changes from the October 2018 filing to the April 2019 filing would not have increased the amounts recoverable from ratepayers by $47 million; instead, the changes would have just modified where the dollars resided. Second, since these amounts are related to “unfunded” items, there should be no effect on rates of any deferred taxes the Company recorded under Generally Accepted Accounting Principles (GAAP). The Commission’s ratemaking policies have allowed those items to be flowed REDACTED 71 Reply Bench Analysis Docket No. 2018-00194 through. As a result, rates would reflect the deduction estimated for the rate year but would not reflect an expense deduction for items where flow-through had been used. There is no reconciliation or true-up for these costs. Staff also does not believe that CMP has adequately supported its proposed change and thus Staff recommends that CMP not be allowed to collect the “recoverable” Excess ADIT that it has proposed. As in the past with these complicated tax issues, Staff recommends that more time be spent to obtain a full understanding of this issue so that what is ultimately included in rates is accurate and justified. If these items have been reflected in the calculation of rate base, the balances should be removed until the final review is completed. 2. Pension/OPEB a. Company’s Position CMP adopted Accounting Standards Codification (ASC) 715 in 2018. ASC 715 limited the amount of pension and OPEB expenses that can be capitalized to service costs, which results in an increase in the portion of such costs that are expensed. CMP states that it is “allowed to defer ‘GAAP Financial Accounting Standards’ under provision number 16 of Attachment 2 of the Stipulation approving rates in Docket No. 2013-00168.” Rev. Req. Reb. At 19. The revenue requirement CMP proposes in its rebuttal filing reflects the increased expense due to the adoption of ASC 715 for the rate year. In addition to this, CMP requests that it be allowed to recover the increased expense that was not previously reflected in rates for the period of January 2018 through October 2019 ($5,585,000). It has asked to recover this amount over a two-year period, at $2,770,000 per year. Id. at 20. b. Staff’s Reply Reply Bench Analysis REDACTED 72 Docket No. 2018-00194 The Commission has allowed utilities to adopt GAAP financial standards in the past and in this instance does not disagree with CMP’s adoption of ASC 715. However, Staff notes that under the prior accounting methodology, the recovery period for capitalized costs was the life of the asset, which in some cases could be over 30 years. CMP did not defer the increased costs related to the adoption of ASC 715 as a regulatory asset in 2018, even though a utility would typically defer those costs when it plans to seek recovery of costs that are not being collected in rates and where the regulatory body would generally grant recovery. It also appears that CMP did not seek recovery of these costs in its initial filing, but only sought their recovery in its rebuttal. See ODR-012-061, Att. 2 (see the tab labeled “1 Inputs” in this Excel file, which shows that the amount for Pension/OPEB was increased to $13.585 million from $11.345 million due to “updated allocation factors, plus ASC 715” (see especially cell J218)). Staff notes that in response to EXM-003-073 the Company stated: “There is no precedence of recovery for CMP, so it did not defer such costs at the time of the adoption of the FASB's ASC 715 update.” Because CMP did not defer the costs and did not request an accounting order, Staff recommends that CMP not be allowed to retroactively recover these costs. If the Commission disagrees with Staff’s position, the Staff would recommend a recovery period of ten years, or $558,000 per year. A ten-year period is consistent with the period CMP has proposed for its recoverable excess deferred income taxes. This would reduce CMP’s proposed revenue requirement by $2,211,500. 3. Rate Case Expense a. CMP’s Position Reply Bench Analysis REDACTED 73 Docket No. 2018-00194 In its initial filing, CMP estimated that it would incur approximately $2.135 million in consultancy and legal expenses in the preparation and processing of this rate proceeding. CMP proposed recovery of this amount in a one-year period. In the Bench Analysis, Staff noted that CMP failed to provide support for several elements of its consulting expenses. Specifically, CMP failed to justify the $200,000 estimate for consulting expenses related to its Policy Panel testimony and only established that 50% of the Company’s healthcare and pension consulting expenses, or $100,000, were rate case related. Therefore, Staff recommended that $300,000 be removed from CMP’s rate case consulting expenses. In addition, Staff recommended that CMP’s rate case expenses be normalized over a four-year period instead of the one-year period proposed by the Company, as four years was more representative of the typical duration between rate cases. In its Rebuttal, CMP agrees with Staff’s proposal to eliminate $300,000 in rate case consultancy expenses. However, CMP asserts that this reduction is offset by increases in rate design, shared services, vegetation management, and tax expense consultant costs. CMP also modifies its expense recovery period to two years. CMP’s updated rate case expense proposal in comparison to the original position and the Staff’s Bench Analysis proposal is presented in Table 15 below. REDACTED 74 Reply Bench Analysis Docket No. 2018-00194 Table 15 CMP’s Rate Case Expense Updates Table 6 - Rate Case Expense Updates ($000's) Oct Filing 8 Consultancy Costs ROE & Capital S*8tructure Pension and OPEB Studies Policy Taxes Rate Design Shared Services BA ECI Study - Vegetation Management Subtotal Consultancy Costs 9 Legal Costs 1 2 3 4 5 6 7 10 Total Rate Case Expense 11 Amortization Period (years) Amount in Rates 12 $ BA 150 200 200 200 250 250 1,250 $ Rebuttal 150 100 200 250 250 950 $ 150 100 450 425 375 125 1,625 885 885 885 2,135 1,835 2,510 4 459 2 $ 1,255 1 $ 2,135 $ Rev. Req. Reb. at 8, Table 6. b. Staff’s Reply As noted above, CMP contends that it incurred increased tax, rate design, shared services, and vegetation management study expenses. When asked to explain and provide an explanation for the increased rate case expenses, CMP responded that its initial filing reflected cost estimates that would be impacted by the level of discovery, participation in technical conferences, and other factors occurring during the proceeding. The increase in estimates for tax, rate design, and shared services study consultancy costs reflected in Table 15 above, resulted from the work in these areas that exceeded what the Company anticipated in its initial filing. EXM-012-018. In REDACTED 75 Reply Bench Analysis Docket No. 2018-00194 response to an oral data request at the May 2019 technical conference, CMP provided the following summary of year-to-date rate case expenses incurred, including legal fees. Table 16 Rate Case Summary of Costs through May 28, 2019 Summary – Non-Legal Consultancy Amount AON Hewitt CMP Rate Case – Pension/OPEB Concentric Resiliency-Engineering Study Concentric ROE/Cap Structure/Other 120,680 NERA Marginal Costs/Tariff Design 209,090 Nexant Resiliency/Benefit Cost Analysis 76,170 NorthBridge Group Reg Strategy/Cash Flow 38,585 PWC Service Company Market Study $ 35,000 68,311 431,283 $979,115 Legal (a) Pierce Atwood Revenue Requirements Pierce Atwood Rate Design 239,677 11,133 $250,810 Total Consultancy and Legal $1,229,925 ODR-012-064. CMP’s oral data request response indicates that it has spent only $35,000 in healthcare and pension consulting fees, significantly less than the $100,000 initially proposed for allowance in the Bench Analysis and accepted by CMP. As this issue is not being contested, it does not appear that CMP will incur additional healthcare and pension consulting expenses over the duration of the rate case. Therefore, Staff proposes that CMP’s healthcare and pension consulting fees be reduced by $65,000 to $35,000, the amount actually incurred. Reply Bench Analysis REDACTED 76 Docket No. 2018-00194 CMP asserts that its tax consulting expenses have increased from $200,000 as estimated as part of its October filing to $450,000 at the time of its Rebuttal, an increase of $250,000. CMP’s oral data request response fails to provide any support for this tax consulting costs. Without any support for this cost, the Staff believes that this cost item should be excluded from regulatory expense at this time. It is also difficult to comprehend how CMP’s tax consulting expense could have more than doubled during this time period since there was nothing in the rate case process that was so onerous as to warrant such a significant increase in costs. As set forth in Table 15, CMP’s rate design costs allegedly increased from $250,000 to $425,000. However, CMP’s oral data request response indicates that it has only spent $209,090 in rate design costs to date. ODR-012-064. Staff recognizes that these are expenses that will continue to accumulate throughout the duration of this litigation and, therefore, the rate design expense amount should be updated prior to the close of the record in this case. Finally, CMP’s oral data request response states that the Company has spent $250,810 to date in rate case legal fees. Staff recognizes that CMP’s legal fees will increase throughout the course of this litigation and therefore proposes a $500,000 rate case legal expense recovery as a placeholder until CMP’s actual rate case legal fees can be determined at the end of the rate proceeding. With regards to the normalization period for the recovery of rate case costs, the Staff believes that the Company’s normalization period of two years for all expenses is both too rapid and also fails to recognize the differing recurrence periods of the REDACTED 77 Reply Bench Analysis Docket No. 2018-00194 individual expense items. Based on the likely recurrence periods of the individual expense items, the Staff proposes the following recovery periods: Table 17 Staff’s Recommended Recovery Periods Cost Category Time Period Revenue Requirement/Rate Setting (Including Legal, ROE, Pension, Engineering) 4 years Rate Design 5 years Vegetation Management Consultant 5 years Tax Services (Related to TCJA or Power Tax) 10 years Affiliate Market Rate Study 20 years 4. Storm Cost Expense See discussion in Section II (C). 5. Customer Service Staffing In the initial Bench Analysis, the Staff proposed basing Revenue Requirement on Customer Service staffing level of 138 positions. B.A. at 25. For the reasons similar to those discussed in Section IV (C) above, the Staff believes that this number should be adjusted to 155 positions. Staff believes that this number reflects the actual hiring that has occurred to date as well the Company’s reversal of its position that it will not be filling vacancies during the rate effective year.10 Compensation for the additional 10 The record on this point is somewhat unclear and Staff will need to verify this number at the hearing. REDACTED 78 Reply Bench Analysis Docket No. 2018-00194 employees added should be based on the methodology proposed by the Staff in the initial Bench Analysis and now accepted by the Company. Rev. Req. Reb. at 5. 6. Inflation Factor a. Company’s Position CMP uses a general inflation index for cost and revenue projections where a specific forecast is not used. In its Direct Case, CMP calculated the projected increase in the Gross Domestic Product Price Index (GDP-PI) using confidential source data and projections from IHS Markit U.S. Economic Outlook—July 2018. In the Bench Analysis, Staff updated the general inflation outlook to reflect more recent data and projections and relied on publicly available projections of GDP-PI, specifically forecasts provided by the Congressional Budget Office (CBO) in January 2019. The Company does not agree with using the CBO projections, noting that the IHS forecasts have been used in three prior rate proceedings. Rev. Req. Reb. at 9. CMP does update the general inflation projection to reflect the IHS January 2019 data. Id. b. Staff’s Reply Staff notes that the use of the CBO projections of inflation is not without precedent in rate proceedings. In the most recent rate proceeding for Emera Maine, the inflation index used was the average of the CBO forecast and a forecast provided by Moody’s Blue Chip Economic Indicators Forecast. Emera Maine, Proposed Increase in Distribution Rates, Docket No. 2017-00198, Testimony and Exhibits of Kris Chahley, David Davoren and Gregg H. Therrien, October 2, 2017, Exhibit RR-69. The update provided by CMP reflecting more recent inflation projections brings the Company’s calculation of the RY1 inflator more in line with Staff’s calculations based on the CBO Reply Bench Analysis REDACTED 79 Docket No. 2018-00194 projections. It is not clear that there are significant compelling reasons to favor an approach based on CBO projections versus IHS projections. Staff recommends that the Company apply the same approach as was used in the last Emera Maine case and average the two inflation projections to calculate the RY1 inflator. 7. Injuries and Damages In response to discussions at the December 2018 technical conferences and the Staff’s Bench Analysis, CMP agrees in its Rebuttal Case that the $500,000 reserve adjustment related to underground service drop liabilities is non-recurring and can be removed from test year injuries and damages expense. Rev. Req. Reb. at 10. In addition, the Company confirmed that the reserve for Levesque class action lawsuit is not included in the historic test year. ODR-012-055. Therefore, Staff agrees with the Company’s injuries and damages test year expense amount as adjusted. Reply Bench Analysis Dated: June 17, 2019 REDACTED 80 Docket No. 2018-00194 Respectfully submitted, _______________________ Chuck Cohen Hearing Examiner Brian George Hearing Examiner _______________________ Katie M. Gray Hearing Examiner On behalf of Advisory Staff Faith Huntington Derek Davidson Christine Cook Michael Simmons Lucretia Smith Sally Merritt Margrethe Heimgartner Matthew Rolnick