500 LEE STREET EAST SUITE I600 * t?O. BOX 553 * CHARLESTON, WEST VIRGINIA 25322 TELEPHONE: 304-340- IO00 * TELECOPIER: 304-340-1 I30 ~.jac~son~e~~~coi~ DIRECT TELEPHONE: (304) 340-1008 DIRECT TELECOPIER: (304) 340-1 080 E-Mail: [email protected] State Bar No. 12293

October 12,2018

VIA HAND DELIVERY Confidential Materials Enclosed Ingrid Ferrell, Executive Secretary Public Service Commission of WV 201 Brooks Street Charleston, West Virginia 25323

Re: West Virginia-American Water Company 201 8 Consolidated Water and Wastewater Rate Filings Case Nos. 18-0573-W-42T and 18-0576-8-42T

Dear Ms. Ferrell:

Enclosed are an original and twelve copies of the rebuttal testimony of the following witnesses on behalf of the Company: Brian K. Bruce, Rod P. Nevirauskas, Christina Chard, Jamie D. Hawn, Gregory P. Roach, Carolyn Mount, John R. Cox, John R. Wilde, Ann E. Bulkley, and Patrick L. Baryenbruch. Ms. Hawn’s testimony contains two confidential exhibits (Exhibits JDH-2 and JDH-5), which have been provided under seal. We have also included public redacted versions of these exhibits.

Please file this testimony and circulate the additional copies to the appropriate parties at the Commission. Please also date stamp the extra copies provided and return them with our messenger. As always, we appreciate your assistance.

Sincerely yours,

Nicklaus A. Presley

NAP/mrv Enclosures

Rob Passrnore Ralph Clark John Auville Andrew Gunnoe Chris Howard Mandi Kay Carter Jackie Roberts Ann E. Bulkley Torn White Patrick L. Baryenbruch

4850-800 1-9576.vI BridgepotxWV Charlescon,W 8 Marunrburg WV * Morgantown WV *Wheeltng.WV Denver. CO * Crawfordrvrlle,IN Evansville. IN * Lexingron. KY *Akron. OH * Rmburgh. PA Warhlogtwr. DC SI

CH ARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-5-42T

WEST VIRGINIA-AMERICAN WATER CO~PANY

Rule 42T Tariff Filing to increase Rates and Charges

REBUTTAL TESTIMONY OF BRIAN K. BRUCE

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 1of 9

1 I. INTRODUCTION AND SUMMARY 2 3 Q. Please identify yourself. 4 A. My name is Brian K. Bruce. I am the President of West Virginia-American Water Company (“West 5 Virginia-American,” “WVAW” or “the Company”). 6 7 9. Have you previously submitted testimony in this proceeding? 8 A. Yes, I previously submitted Direct Testimony in this proceeding filed on April 30, 2018. 9 10 Q. What is the purpose of your rebuttal testimony? 11 A. The purpose of my rebuttal testimony is to underscore how the ratemaking adjustments and 12 policies advocated by the Public Service Commission of West Virginia Staff (“Staff”) and the 13 Consumer Advocate Division (“CAD”) in this case adversely impact the Company’s ability to 14 achieve funding levels that best serve the long-term interests of its customers. Among other 15 things, various recommendations of Staff and CAD would result in an abrupt about-face to the 16 constructive rate policies recently adopted by the Public Service Commission of West Virginia 17 (“Commission”) in our most recently concluded case. These include, but are not limited to, 18 turning back the clock by (i) using a test year thirteen month average plant balance, (ii) refusing 19 to recognize rate base and expenses in the Addendum Period, (iii) advocating for unsupportive 20 regulation by recommending a significant reduction to the rate of return on equity granted in the 21 last case, and (iv) depriving the Company of any reasonable ability to achieve its revenue targets 22 by refusing to provide the tools to combat the Company’s revenue erosion. West Virginia- 23 American is facing persistent revenue shortfalls from declining use per customer, declining 24 customer count, and the need to rebuild legacy infrastructure and the recommendations of the 25 Staff and CAD turn a blind eye to these problems. 26 27 9. Have the Company proposals attempted to address these issues? 28 A. Yes, they have. Company witnesses Rod Nevirauskas, John Cox, Jamie Hawn and Christina Chard 29 explain how the Company’s ratemaking proposals - revenue stabilization mechanism (“RSM”) and 30 recognition of rate base, expenses and revenue through the commencement of new rates in West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Brian K. Bruce Page 2 of 9

1 February 2019 (Addendum Period) - are intended to, and will, efficiently and effectively address 2 those concerns. The RSM provides the Company a realistic opportunity to collect its authorized 3 revenue requirement, and the synchronization of rate base, revenue and expenses with new rates 4 properly recognizes the expense levels and plant that will be serving West Virginia-American’s 5 customers when the new rates take effect. 6 7 Q. What are the likely consequences if the Commission adopts a return on equity (“ROE“)that is 8 recommended by the Staff or the CAD? 9 A. As WVAW witness Ann Bulkley explains in her rebuttal testimony, in Staffs case, their 9.27% ROE

10 recommendation is 48 basis points lower than the ROE allowed in our last rate case at a time

11 when interest rates are rising along with investors’ return requirements. It is also at the very low 12 end of ROESallowed nationally where recently authorized ROES for water utilities have been from 13 9.20 percent to 10.50 percent in 2018. The CAD’S lower ROE recommendation of 8.75 percent is 14 100 basis points lower than was allowed by the Commission in the Company’s last case.

15 Furthermore, if these low ROESare combined with Staff’s recommendation to impose an equity 16 ratio of just 45.59%,’ or the CAD’S equity ratio of 47.73%,2 the resulting low weighted returns on 17 equity would directly and adversely impact the Company’s ability to secure discretionary funding 18 levels that best serve the long-term interests of our customers. 19 20 9. Was there a significant, supportive regulatory change adopted by the Commission in the 21 Company’s last rate case? 22 A. Yes, terminal rate base recognition of actual and projected capital investments made through the 23 year immediately preceding the rate year was a key factor in the Company’s 2015 rate case. This 24 regulatory treatment significantly reduced regulatory lag for the Company as explained in the 25 rebuttal testimony of John Cox. Accordingly, and in keeping with this policy, the Company 26 presented terminal rate base treatment through its Addendum Period filing. 27

Appendix US-1. * Exhibit LA-1, Schedule D. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 3 of 9

1 Q. Do Staff and the CAD support the continuation of the Commission’s constructive regulatory 2 policy? 3 A. No, surprisingly, they do not. In an unfortunate about-face, both Staff and CAD witness Smith not 4 only removed the Addendum Period from rate base, but also used a test year thirteen-month 5 average plant balance, despite the Commission authorizing terminal rate base treatment through 6 the Addendum Period in 2015. Their justification for this removal is due to the Company being 7 awarded a DSlC in 2017. 8 9 Q. Does the fact that the Commission approved a DSlC for WAW warrant the rejection of 10 recognizing rate base through the Addendum Period? 11 A. No, it does not. Perhaps if DSlC were used and available for all of the Company’s investment this 12 might be a supportable position. That is not the case, however. The Company invests only 13 approximately 40% of its capital yearly in DSlC projects and DSlC investment is only 13% of this 14 rate request. As I will explain below, refusing to recognize plant in service will hurt the Company’s 15 ability to attract discretionary capital investment. In addition, there is no question that when the

16 Commission approved the Addendum Period rate base recognition concept, it did so knowing that 17 it had simultaneously authorized the Company to seek Commission approval of a Staff-

18 recommended surcharge mechanism very much like the DSIC. The Commission expressly found 19 that such a mechanism “would also be effective in reducing the regulatory lag. . ..” 2015 Rate 20 Case Order at 26. Staff and CAD’S reasoning does not support the Commission changing its 21 decision to authorize Addendum Period cost recovery in the last case. 22 23 9. In addition to rate base, should expenses and revenue also be recognized through the 24 Addendum Period? 25 A. Yes, absolutely they should be recognized. Regulatory lag is not only composed of the inability to 26 earn a return on our investment, it is also important that expense increases and revenue erosion

27 be recognized. The Commission is setting rates for a period that begins in February 2019. It is 28 hardly extreme to take into account all rate base, expenses and revenues for a period that ends 29 before new rates even begin. 30 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-421 Rebuttal Testimony of Brian K. Bruce Page 4 of 9

1 Q. Staff and CAD also contend that your ROE should be reduced to reflect the DSlC and RSM. Are 2 their recommendationsappropriate? 3 A. No, they are not, and Ms. Bulkley explains why these are not appropriate adjustments in her 4 testimony. I would further point out that many of our affiliated companies in the American Water 5 system already have a DSlC or RSM. These include our affiliates in Pennsylvania, Missouri, Illinois, 6 Indiana, New York, Virginia, Iowa, Tennessee, California and New Jersey. To the extent that our 7 ROE is reduced even further to reflect a mechanism that our sister companies have and which 8 does not reduce their respective ROE will, as I explain below, put us even further behind in the 9 competition for discretionary capital. 10 11 . Is the RSM important to ~AW? 12 A. Yes, the ability to collect the revenue authorized in by the Commission in our rate order is critical 13 to our ability to achieve the rate of return found reasonable in that order. If, as has been the case, 14 we cannot achieve that forecasted revenue, then we have no realistic ability to earn the rate of 15 return that the Commission finds reasonable. Given the continuing trend of declining use per 16 customer, unless the rates set reflect that trend during the period they are effective, the Company 17 will suffer revenue shortfalls. The inability to capture revenue lost through declining use per 18 customer adversely affects our earnings and, hence, our ability to compete for discretionary 19 ca pita1. 20

21 Q. Are you aware of the net result of Staff’s and the CAD’S respective ROE recommendations? 22 A. Yes, in her rebuttal testimony, Ms. Bulkley explained that Staff witness Shamblin’s recommended 23 equity ratio of 45.59 percent, in combination with her recommended ROE of 9.27 percent, would 24 provide an overall equity cost rate of 4.23 percent. Ms. Bulkley’s Schedule AEB-R-2, however, 25 shows that the range of authorized equity cost rate is 4.74 percent to 5.45 percent and, therefore 26 Staff’s proposed ROE and equity ratio results in an equity return which is well below the low end 27 of the range established by recent authorizations. Mr. Smith’s recommended equity ratio of 47.73 28 percent and ROE of 8.75 percent would provide an equity cost rate of just 4.17 percent, which is 29 even lower than Staff and are below the low end of the range established by recently authorized 30 ROES. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 5 of 9

1

2 Q. Do Staff’s and CAD’s ROE proposals concern you? 3 A. Yes, they do. As Ms. Bulkley testifies, “The investment community would find it very concerning 4 if the Commission were to adopt Ms. Shamblin’s proposal to cut the Company’s equity return by 5 48 basis points.” That concern would only be magnified by Mr. Smith’s recommendation to cut 6 that return by 100 basis points. Given this environment of rising interest rates and costs for the 7 Company, I find it difficult to understand the rationale behind recommendations to reduce the

8 Company’s rate of return on equity from that found reasonable by the Commission in the last 9 case. 10

11 Q. How do these equity cost rates stack up against the equity cost rates that are in place for your 12 sister utility companies in the American Water system? 13 A. Ms. Bulkley advised me that the median equity cost rate approved by regulators for the regulated 14 water utilities in the American Water system is 4.88%. Therefore, the rates of return and equity 15 ratios recommended by Staff and the CAD, respectively, would be significantly lower than the 16 median cost rate allowed for other American Water companies by regulators. 17

18 Q. Are there other factors that exacerbate the low equity cost rates recommended by the 19 witnesses for Staff and CAD? 20 A. Yes, most definitely. As I explained above, the Company has chronically experienced revenue 21 erosion and only partial rate recognition of expenses and plant additions. If this were to continue 22 with the rejection of Addendum Period cost recovery and our requested RSM, the effects of the 23 low equity cost rates recommended by Staff and the CAD would be compounded by the 24 knowledge that we would be unlikely even to achieve that substandard cost rate. 25

26 Q. What would the consequences be if the Commission were to accept Staff and CAD’s 27 recommendation regarding the Addendum Period and RSM, as well as such low equity cost 28 rates for WAW? 29 A. Such a decision would directly and adversely impact the Company’s ability to secure discretionary 30 funding. American Water owns the common stock of regulated water and wastewater utilities in West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 6 of 9

1 14 jurisdictions, including West Virginia. While American Water always ensures that each of its 2 water and wastewater utilities is afforded funding to provide safe, adequate, and reliable service, 3 investment funding is not limitless. American Water is competing with other companies and 4 industries in the marketplace for capital, and American Water’s subsidiaries are competing within 5 the American Water system for discretionary allocations of American Water’s investment and 6 financing capacity. 7 Discretionary allocations within American Water can be influenced by a company’s capital 8 requirements, as well as by market conditions and available funds. Like any rational investor, 9 American Water considers its opportunity to earn a return when considering investment 10 opportunities. All other things being equal, American Water, again, like any investor, generally 11 favors higher-return investments over lower ones, provided the investment risks are comparable. 12 If Staffs or CAD‘S unsupportive regulatory recommendations - compounded by low equity 13 returns and ratios - were adopted, the Company would have among the lowest authorized equity 14 cost rate of any of American Water’s regulated subsidiaries. As a consequence, American Water 15 would have much less incentive to allocate discretionary funding to WVAW than to its other 16 regulated subsidiaries. 17

18 Q. Is the concern about attracting money from investors, or what is best for customers? 19 A. Both; this is about aligning customer and shareholder interests. We have a multi-decade-long 20 investment need that is funded up-front by shareholders and lenders and recovered from 21 customers over a 40 plus-year time frame. American Water is acutely aware that utility statutes 22 and regulatory schemes vary from state to state; regulatory commissions have different policies, 23 administrative procedures, and precedents; and these differences affect American Water’s 24 investment decisions. investors have choices. The choices investors make must necessarily 25 consider the returns available on invested capital. When investors have an incentive to invest, 26 they wilt, and when they do not, they won’t. Imposing extraordinarily low shareholder returns 27 may have the temporary effect of lowering rates, but that practice ultimately imposes long-term 28 costs that cannot be measured in dollars alone. Discouraging discretionary funding that serves 29 the long-term interests of customers, in the name of “protecting” those customers, ultimately 30 harms the constituency the policy is meant to help. It is well-recognized that a reasonable ROE West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 7 of 9

1 and equity ratio is necessary to align both customer and investor interests. This results in a 2 stronger and more reliable water system for both current and future customers, reduces the need 3 for general rate cases, lessens the occurrence of customer “rate shock,” supports the 4 maintenance and improvement of essential infrastructure, ensures safety and reliability, and 5 allows for more efficient, streamlined regulation. 6 7 Q. What is your role in securing capital for WAW? 8 A. Part of my job involves making the case to American Water for investment in West Virginia. Every 9 affiliate employs someone in a capacity comparable to mine, and part of that person’s job is to 10 make the case for investment in their respective state. Because the collective need for capital 11 inevitably exceeds the resources available from American Water, the various states are effectively 12 competitors. This type of competition is healthy because it forces the utilities to identify and 13 develop projects that produce the greatest benefits at the least cost. 14 15 Q. Are you suggesting that American Water will cut-off investment to WAW if the Commission 16 adopts equity cost rate and other recommendations that place it among the worst performing 17 in the American Water system?

18 A. I am not saying that at all. As I said previously, the Company will fulfill its duty to provide safe, 19 adequate, and reliable service. West Virginia-American continues to make the necessary 20 investments in developing and maintaining adequate sources of supply, treatment, pumping, 21 transmission and distribution facilities, as well as to comply with applicable environmental laws 22 and regulations (Safe Drinking Water Act, the Clean Water Act, etc.). 23 24 Q. Where does the Commission’s approach make a difference? 25 A. When an investor is confronted with the choice of investing in West Virginia at the 4.23 or 4.17 26 equity cost rates recommended by Staff and CAD, respectively, or, for example, Iowa or Illinois, 27 at a 5.00% and 4.88% equity cost rate, respectively, the disparity in available returns will 28 necessarily steer the allocation of discretionary capital in a way that requires WVAW to manage 29 operations toward the “bare minimum” end of the acceptable range. By doing so, capital is freed- 30 up for jurisdictions with higher equity costs rates and a greater opportunity to earn those West Virginia American Water Company Case Nos, 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 8 of 9

1 authorized returns. The equity cost rates and ratemaking adjustments and policies advocated by 2 Staff and CAD would put WVAW in a subordinate position, resulting in capital funding at a level 3 necessary to maintain only adequate service, and certainly not optimal service. And it would 4 surely put me at a distinct competitive disadvantage against my colleagues as we vie for 5 investment by American Water. 6 7 Q. Have you quantified the range of what would be considered a “bare minimum” level of capital 8 investment versus a level that begins to suggest imprudence? 9 A. I have not, and I do not believe anyone else can, either. There are simply too many variables to 10 consider in deciding whether a utility is spending too much or too little. These judgments are 11 typically made in hindsight based on whether the decision was reasonable, given what was

12 known, or should have been known, at the time the decision was made. My point is not to

13 establish that there is a definite, quantifiable figure for what is “too much” or “too little.” 1 am 14 simply trying to stress the point that authorized ROES, equity ratios, and the resultant equity cost 15 rates have a very real influence in how capital allocation decisions are made in the real world.

16 And it is my firm belief that Staff’s and CAD’S respective ROE and equity ratio recommendations 17 in this case ignore that reality. When those recommendations are exacerbated by arguments to 18 reduce the equity return still further, to exclude rate base, expenses and revenue beyond the 19 historic test year, and to deny the RSM, the ability to make the case for investing in West Virginia 20 becomes infinitely more difficult. 21

22 Q. What type of capital projects would be at risk if the Commission authorized a 23 percent equity cost rate resulting from a low equity ratio? 24 A. I cannot provide a line-by-line description of every planned project and how it would be affected.

25 What I can say is that investment decisions would have to be re-evaluated. The internal 26 competition for capital with affiliates is difficult enough without being saddled with the lowest 27 equity cost rate in the American water system, which would simply render the Company 28 uncompetitive in relation to its affiliates. 29 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Brian K. Bruce Page 9 of 9

1 Q. If the Company cuts back capital investment to a level supported by the low equity cost rates 2 advocated by Staff and the CAD, who would be harmed? 3 A. Everyone: the Company, customers, and the West Virginia economy. The Company currently 4 invests between $55 and $70 million annually in system improvements and infrastructure. This 5 level of investment in the state has tremendous statewide impacts including jobs, spending on 6 goods and services, system reliability and improved customer service. The Commission is being 7 asked to authorize an equity cost rate that is among the lowest in our system and approve

8 unsupportive ratemaking adjustments and policies. Such a decision would have regrettable 9 consequences, starting with the unraveling of the benefits achieved through investment that got

10 the Company to where it is today. 11

12 Q. What do you want the Commission to do? 13 A. I obviously want the Commission to adopt the Company’s ROE recommendation and constructive 14 ratemaking policies. I also want the Commission to understand the very real consequences of 15 adopting a ROE, equity ratio and unsupportive ratemaking policies that together are so out of step 16 with reality and returns awarded in other jurisdictions. 17

18 Q. Does this conclude your rebuttal testimony? 19 A. Yes. PUBLIC SERVICE CO~~ISSION OF WEST VIRGINIA CHARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-S-42T

WEST VIRGINIA-AMERICAN WATER COMPANY

Rule 42T Tariff Filing to Increase Rates and Charges

REBUTTAL TESTIMONY OF ROD P. NEVIRAUSKAS

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-421 Rebuttal Testimony of Rod P. Nevirauskas Page 1of 13

1 Q* Please identify yourself. 2 A. My name is Rod P. Nevirauskas. I am employed by American Water Works Service Company, Inc.

3 (“Service Company”) as a Senior Director of Rates and Regulations, and I provided direct 4 testimony in this docket filed on April 30, 2018. In this rebuttal testimony, I will address capital 5 structure and short-term debt issues raised in the direct testimony filed by Staff’s witness Kaitlin

6 Shamblin and Consumer Advocate Division’s (“CAD”) witness, Ralph Smith. I will also address the 7 direct testimony of CAD’S witness Scott Rubin regarding the Company’s proposed revenue 8 stabilization mechanism (“RSM”). 9 10 Q. Do other witnesses also address Mr. Rubin’s RSM testimony? 11 A. Yes. Anne Bulkley addresses the issue of cost of equity and RSM. 12

13 REVENUE STABILIZATION MECHANISM 14 15 Q. Mr. Rubin argues that the Company’s proposed RSM should not be implemented without 16 aggressive conservation programs. Do you agree with Mr. Rubin’s assertion? 17 A. No, I do not. My response to Mr. Rubin is that he fails to recognize or acknowledge the undisputed 18 facts conclusively established in Mr. Roach’s testimony: (1) that water consumption per customer

19 for residential and commercial classes has been declining for years, (2) that it will continue to do 20 so for the foreseeable future, and (3) that the majority of the continuing decline in water 21 consumption will still come from sources other than WVAW, largely through mandated 22 government efficiency and conservation standards. In fact, the Commission found in the 23 Company’s last case that there was a clear trend in declining use per customer. Those facts, 24 established beyond doubt, warrant an RSM. While I agree with Mr. Rubin that approval of an RSM 25 will remove a disincentive for WVAW to more actively promote demand-side efficiency programs

26 (i.e., conservation), it is the existence of mandatory conservation by federal law that contributes 27 in large part to the clear and continuing pattern of existing usage decline. 28 29 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 2 of 13

1 9. Mr. Rubin cites a single 2010 Nebraska Public Service Commission decision that rejected a 2 decoupling proposal because a gas utility did not have a robust conservation program. Are you 3 aware of commission decisions that have tied approval of a water company’s RSM to an 4 aggressive water conservation program? 5 A. No, I am not. There are also numerous gas utilities around the country that have RSMs that are 6 also not tethered to conservation programs. As WVAW witness Roach has explained, several 7 federal statutes mandate significant water use reduction standards in new appliances, toilets and 8 other water using items such as shower heads which, when adopted by current homeowners and 9 introduced through replacements and home renovations, produce conservation increases. (Roach 10 Dir., pp. 12-16). Water conservation measures are a reality in West Virginia, and the Company has 11 demonstrated that fact in its testimony. No party has realistically disputed the Company’s 12 testimony on this issue because no party can do so. Federal standards are applicable nationwide, 13 and, as a result of those standards and other factors, Mr. Roach has documented a systemic 14 reduction in WVAW residential customers’ use of water by roughly 1.65 percent per year. (Roach 15 Dir., p. 3). 16 17 9. Did the Commission recognize the trend of declining usage per customer in the last case? 18 A. Yes, but the Commission at that time adopted what it admitted was a “cautious approach” and 19 only reflected the declining residential usage per customer adjustment up to the beginning of the 20 Rate Year. Although the Commission found that the trend of declining usage per customer 21 consumption “has been real, significant and damaging to WVAWC,” we do not believe that the 22 remedy chosen in the Company’s last rate case sufficiently addressed the adverse effects of the

23 trend on the Company, and we continue to believe that the RSM is the appropriate remedy for 24 that problem of continued revenue erosion. 25 26 Q. If conservation is not the primary reason for the RSM, as Mr. Rubin suggests, what is the main 27 reason?

28 A. As I stated in my direct testimony, the Company’s proposed RSM is designed to maintain the 29 Company’s revenues at the level the Commission approves in this case going forward. The 30 mechanism effectively removes the errors that are inherent in the process of forecasting the test West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-421 Rebuttal Testimony of Rod P. Nevirauskas Page 3 of 13

1 year level of sales. These forecasting errors are caused by the changes in volume of water sold 2 due to factors beyond the control of the Company or the Commission (i.e., the Commission has

3 no mechanism in traditional ratemaking to take this into account). The intent of this mechanism

4 is to better match the expected rate year revenues with actual revenues over time. As Mr. Roach’s 5 testimony demonstrates, and as the Commission has found, West Virginia has seen a continued 6 and persistent trend of declining usage per customer.1 Nevertheless, it appears Mr. Rubin believes 7 that utility revenues should be driven by the inability of the regulatory process to properly account

8 for declining customer usage or weather, rather than the utility’s efforts at providing safe,

9 adequate and reliable service to its customers. It is simply not in the best interests of customers 10 if the Company is deprived of sufficient revenue to cover its operational expenses or pay for 11 investment made to serve West Virginians. 12 13 4. Mr. Rubin asserts that an RSM is contrary to established regulatory principles and policies. Do 14 you agree? 15 A. No, I do not. The approval of alternative ratemaking approaches such as RSM and DSlC by 16 regulatory commissions is widespread in the utility industry. Moreover, the ratemaking process is 17 not a fixed formula or a set principle. The standard for setting rates is “just and reasonable,” which

18 has always been interpreted as a balancing of the interests of the parties. In effect, it is the end 19 result that is judged just and reasonable, not necessarily the process that leads to that end. At a 20 high level, the balance that must be struck is between the legitimate concern that customers be 21 provided safe and adequate service while being protected from paying excessive rates and the

In the Company’s last rate case, the Commission stated: “We are troubled by the loss of residential revenues from the decline in the average residential usage per customer. Clearly, over time, the decline in residential consumption has been real, significant and damaging to WVAWC.” The Commission went on to observe that the development of water and energy conserving plumbing fixtures and appliances has not stopped, and is likely to intensify.

West Virginia-American Water Company, Case Nos. 15-0676-W-42T and 15-0675-S-42T (Commission Order 17 dated February 24,2016) (“2015 Rate Case Order”) at 39. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 4 of 13

1 equally legitimate concern that the utility have a fair opportunity to recover its prudently incurred

2 costs of managing the system, including paying its expenses and recovering the cost of obtaining 3 capital to meet its service obligations. Since rates are set prospectively, traditionally, that balance 4 was achieved, in part, by using a sales normalization process, based on the assumption that any 5 random fluctuations in historical water sales are effectively smoothed out thereby providing the

6 utility with an opportunity, if it managed its system efficiently, to recover its prudent and

7 reasonable costs including its cost of capital. That system worked well for many years because the 8 assumption that the normalization process fairly represented, at least on average, the actual 9 results in the rate-effective period tended to hold (i.e., the errors were effectively smoothed out). 20 That assumption no longer holds because of factors that are beyond the control of the Company 11 as has been documented in this case. (See e.g., Roach Dir., pp. 11-20). 12 13 Q. Are you suggesting that WA should expect to be protected against any circumstances that 14 would cause future sales and revenues to be different from the levels established in this case 15 for the test year? 16 A. No. I am not suggesting that any utility expects some “perfect world” where test year revenues

17 exactly match actual results. That is not the assumption behind the test year concept, nor is it 18 particularly relevant to the issue in this case. The assumption behind the test year concept is that, 19 on balance, the estimates that are used to determine final rates are unbiased in the sense that 20 there is no systematic downward or upward bias in the rate effective period. For example, if the 21 Company were simply allowed to choose its level of expected sales there could be a concern that 22 would bias rates, and in turn revenue recovery, upward. To mitigate this potential bias the 23 Commission attempts to do the best job it can in deciding on sales normalization. Equally 24 important, however, if the normalization process cannot produce an unbiased result then another 25 method needs to be devised to maintain the proper balance. 26

27 Q. Does the RSM unfairly guarantee a level of revenue going forward? 28 A. No, it does not, just as straight fixed variable rate design does not unfairly guarantee a level of 29 revenue going forward. The RSM provides a way to re-establish the balance that was intended by 30 the normalization process. Utilities have a duty to manage their operations in an efficient way, West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 5 of 13

1 which includes both fiscal and physical management of the system. Neither an unbiased 2 normalization process nor the RSM relieve the utility of that duty in any way. If the normalization 3 process is unbiased then the utility can expect, on average over the long term, to recover the

4 revenues allowed by the Commission. I would not call that a guarantee, but I would call it a 5 reasonable expectation. When that expectation is no longer reasonable under the traditional

6 normalization process, an alternative method must be used to reestablish the balance. Page 6 of 7 my direct testimony includes a line graph that shows that actual consumption levels met or

8 exceeded authorized rate case usage levels only once in the last ten years. Over the past ten 9 years, the Company has sold over 3.9 billion gallons less than what the Commission orders 10 predicted, which is an average deficit of approximately 400 million gallons in sales per year. This 11 clearly shows that the forecast of consumption historically approved by the Commission has not 12 produced reasonable results. Simply put, under an unbiased normalization process and under the

13 RSM, the Company has a reasonable expectation that it will have a fair opportunity to recover 14 costs. In this way the RSM restores the balance intended by the traditional normalization 15 approach. 16

17 Q. Is the RSM the only possible approach to address the problem with normalization?

18 A. No. A substantial portion of the Company’s cost structure in the test year are fixed costs, Le., costs 19 that do not change as consumption changes. If the Company’s rates were set to recover fixed 20 costs through fixed charges and variable costs through volumetric rates, then sales forecasts, 21 while still important to set rates, would be less likely to bias revenue recovery one way or the 22 other. The Company proposed the RSM as an alternative approach to address the balancing of 23 the interests of the Company and its customers while at the same time not radically changing the 24 rate structure.

25

26 Q. Please address Mr. Rubin’s comments that the RSM will reduce the Company’s risk. 27 A. While RSMs, straight fixed variable rate design and other regulatory approaches may mitigate 28 some risk, there are other significant factors to consider that work in the reverse direction to 29 increase risk -for example, declining customer water use, declining customers, and the Company’s West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 6 of 13

1 dependence on a significant capital spending program requiring external financing. In other 2 words, these alternative ratemaking approaches constitute responses to other risks that have 3 arisen or been heightened in the water industry and are appropriate regulatory responses to

4 those risks. Company witness Ann Bulkley discusses the impact of an RSM on utility company risk 5 and the cost of capital. 6

7 Q. Do you believe that implementation of the RSM unduly shifts risks to customers and away from 8 WVAW (Rubin p. ll)?

9 A. No. This is an argument that is often leveled at RSMs and is completely without merit when 10 properly understood. It is important to remember that the RSM is effectively a surrogate for more 11 economic pricing methods by which fixed charges recover fixed costs and volumetric rates recover 12 only those costs that change with changing consumption. Under WVAW’s current and proposed 13 rate structure (absent the RSM), the entire cost of owning, operating, and maintaining the water 14 distribution system, which does not change based on increases or decreases in water 15 consumption, is largely paid for through revenues that are tied to increases or decreases in water 16 consumption. Under a more commercially responsive pricing scheme, fixed costs would be 17 included in fixed charges and variable costs would be included in variable charges, with all 18 customers paying an equal amount for contributions to the fixed costs of prudently managing the 19 system.

20 If rates were set in this fashion, then customers would pay a higher fixed rate and obtain a bill 21 reduction when their activity reduces the costs to serve the customer. In effect, the RSM recovers 22 fixed costs that otherwise should have been recovered in fixed charges but does not increase the 23 costs that are properly recovered through rates. The RSM addresses a specific malady in the 24 normalization process which mimics the process as it was intended to work. It is neither 25 inconsistent with nor contrary to regulatory policies or principles, nor does it shift of risk to 26 customers when properly understood.

27 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 7 of 13

1 Q. Mr. Rubin also argues that an RSM is particularly inappropriate for WAW due to affordability 2 concerns. Are his concerns justified?

3 A. No, they are not. Mr. Rubin appears to misunderstand or simply overlooks two fundamental facts 4 about the RSM: (1)it does not increase the costs that are properly recovered through rates and 5 (2) it operates in a perfectly symmetrical manner. If revenue exceeds the level forecasted in rates, 6 the RSM produces a credit to customers. If revenue falls short of the level forecasted in the rate 7 case, the RSM results in a surcharge.

8 As I explained in my direct testimony, the Company proposes that any credit be issued as soon as 9 administratively possible, with the amount of the credit determined based on the number of 10 customers. A one-time credit that is equal to all customers would benefit the low-usage 11 customers at a greater percentage, rewarding customers who conserve water at a higher 12 percentage than those that use more water.

13 The Company also is proposing that any surcharge be based on customers’ volumetric use. A 14 volumetric surcharge would ensure that the low-usage customers would continue to benefit from 15 their conservation because the volumetric rate would be equal for the entire Company. 16 Therefore, if a customer conserves water, he or she will save more money not only on the current 17 bill, but also on any adjustment applied the following year. No matter what happens with sales, 18 customers who use less will pay less. It provides incentive for customers to use water more 19 efficiently because reduced consumption translates into a reduced bill.

20

21 22 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 8 of 13

1 CAPITAL STRUCTURE 2 Q. Please describe CAD’s position on the Company’s capital structure for ratemaking purposes. 3 A. CAD recommends using the Company’s historical test year capital structure and ignores the 4 Company’s pro forma financings of debt and equity scheduled to be issued during the Addendum 5 Period.

6 Q. Do you agree with CAD’s recommendation? 7 A. No. As explained in my direct testimony, the Company included debt and equity financings 8 occurring during the Addendum Period. This capital structure reflects the capital that will be in 9 place to fund the Company’s rate base during the Addendum Period. This is consistent with the 10 2015 Rate Case Order, where (at p. 64) the Commission recognized debt and equity beyond the

11 historical test year for ratemaking purposes. 12 The Commission started with the WVAWC actual capital structure at 13 December 31,2014. The Commission adjusted the December 31,2014 actual 14 capital structure short-term balance by adding $28.457 million of short-term 15 debt that matches the additional net rate base for the Transition Period 16 included in the Commission determination of rate base at the beginning of the 17 Rate Year. These adjustments resulted in total capital that closely 18 approximates the level of rate base authorized for rate recovery in this Order. 19 The Commission made a final adjustment to reflect the $30 million of 20 additional equity financing and a corresponding reduction to the resulting 21 short-term debt balance. 22 23 CAD’S recommendation simply ignores all of the debt and equity financings conducted after the 24 end of the historic test year, including those converted from higher cost debt that is expiring after 25 the expected date of the rate decision in this case. 26

27 Q. Please comment on Staff’s position on the Company’s capital structure for ratemaking 28 purposes. 29 Staffs recommended capital structure is flawed, resulting in a mismatch of capital and rate base. 30 Staff selectively recognizes long-term debt financings and one common equity infusion that 31 occurs during the Addendum Period. By taking advantage of the lower-cost debt refinanced in 32 2017 without including the corresponding make-whole premiums, the LTD cost rate used for 33 ratemaking purposes is lower than it otherwise would be had there been an appropriate inclusion West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Rod P. Nevirauskas Page 9 of 13

1 of the premiums required to obtain the lower LTD cost rates. Further, by not recognizing an equity 2 infusion made in October 2018, Staffs recommended equity ratio is artificially low. Staffs

3 recommended adjustments represent a clear mismatch of rate base and capital and result in a 4 capital structure that is significantly in excess of their recommended rate base. 5

6 Q. You stated that the capital structure and cost of long-term debt proposed by Staff excludes the 7 make-whole premium the Company incurred to retire a portion of its long-term debt in 8 September 2017. Please elaborate. 9 A. On September 13, 2017, the Company refunded 55% of its 5.77% Note with AWCC to refinance

10 this debt at a lower interest rate. The dollar amount of the portion refunded was $39,050,000.

11 For AWCC to call and refund this debt, AWCC, and thus, WVAWC, was required to pay bond 12 holders a make-whole premium in the amount of $5,845,546. The make-whole premium was 13 added to the new replacement debt, resulting in a new Note in the amount of $44,895,546, which 14 was issued on September 13, 2017 with a ten-year term and an interest rate of 2.95%. Staff 15 witness Kaitlyn Shamblin excluded both the unamortized balance and the annual amortization of 16 the make-whole premium. This has resulted in understatement of the company’s embedded cost

17 of long-term debt shown on Appendix IUS-1, Schedule 1, Sheets 3 and 4 of 5, and also 18 understatement of the company’s overall cost of capital shown on Appendix IUS-1, Schedule 1, 19 Sheet 1of 5. Ms. Shamblin’s understatement of the overall cost of capital results not only from 20 the inappropriate exclusion of the make-whole premium from the cost of long-term calculation, 21 but also from understating the Company’s equity ratio by using a debt balance that was not 22 reduced by the unamortized make-whole premium balance. 23

24 Q. What is the basis for including the cost of the make-whole premium in the long-term debt cost 25 calculation, thereby allowing the Company to recover this cost from customers? 26 A. The make-whole premium is a cost the Company incurred to refinance a debt issuance at a lower 27 interest rate. if the make-whole premium was not paid, the lower interest rate could not have

28 been obtained because a debt issuance with a make-whole provision requires an upfront payment 29 be made to bondholders when calling and refunding the debt prior to maturity. Since the cost of 30 the make-whole premium is exceeded by the expected reduction in interest expense over the life West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Rod P. Nevirauskas Page 10 of 13

1 of the new debt, this transaction yields benefits to customers over time. It is unfair and 2 unreasonable, however, to reflect only the benefits while excluding the cost to obtain those 3 benefits, as Staff has done. Therefore, it is reasonable and appropriate to include this cost in the 4 Company’s embedded cost of long-term debt. 5

6 Q. What would the Staff-recommended capital structure and overall cost of capital look like if the 7 make-whole premium were properly included in the calculation of the embedded cost of long- 8 term debt? 9 A. Table 1 below shows the capital structure and cost of capital recommended by Staff witness 10 Shamblin: 11

Table 1

Cost Weighted Balance Ratio Rate -cost Short-Term Debt (Average) $32,051,176 5.03% 2.36% 0.12% Long-Term Debt 314,102,967 49.29% 5.23% 2.58% Preferred Stock 593,776 0.09% 9.00% 0.01% Common Equity 290,506,224 45.59% 9.27% 4.23% Total $637,254,143 100.00% 6.93% 12 13 14 Table 2 below shows the Staff recommended capital structure adjusted for the make-whole 15 premium. The capital structure in Table 1was adjusted to include the unamortized make-whole 16 premium balance at December 31, 2017, consistent with the time period used by Ms. Shamblin. 17 Specifically, I included the unamortized balance of the make-whole premium in the unamortized 18 debt expense total. I also included the annual amortization of the make-whole premium in the 19 annual cost total. This treatment is identical to that of a debt issuance expense. 20 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 11 of 13

1

Table 2

Cost Weighted Balance RatioRate-cost

Short-Term Debt (Average) $32,051,176 5.07% 2.36% 0.12% Long-Term Debt 308,452,273 48.84% 5.51% 2.69% Preferred Stock 593,776 0.09% 9.00% 0.01% Common Equity 290,506,224 46.00% 9.27% 4.26% Total $631,603,449 100.00% 7.08% 2 3 This demonstrates that, based on the capital structure proposed by Ms. Shamblin, excluding the 4 make-whole premium, which is a reasonable and prudent cost of the refinancing, would unfairly 5 reduce the Company’s weighted cost of capital from 7.08% to 6.93%. 6

7 Q. Please explain how Staff’s failure to recognize an equity infusion made in October 2018 results 8 in a recommended equity ratio that is artificially low.

9 A. While Staff recognized the Company’s first equity infusion during the Addendum Period

10 ($36,000,000), it simply ignores the second infusion that took place in October 2018 11 ($20,000,000). The funds from these equity infusions will be used to pay down short-term debt 12 that had been employed to temporarily fund additions to utility property. Adding this total 13 $56,000,000 infusion to paid-in capital produces a pro forma common equity balance of

14 $310,140,019 at February 28,2019, as shown on Statement C- Addendum, page 1. Table 3 below 15 shows the impact of including the $20,000,000 equity infusion to the cost of capital calculation 16 shown in Table 2: 17 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Rod P. Nevirauskas Page 12 of 13

Table 3 1 2 Cost Weighted Balance RatioRate-Cost 4 Short-Term Debt (Average) $12,051,176 1.91% 2.36% 0.05p Long-Term Debt $308,452,273 48.84% 5.51% 2.69% 6 Preferred Stock $593,776 0.09% 9.00% 0.01% 7 Common Equity 310,506,224 49.16% 9.27% 4.56% Total $63 1,603,449 100.00% 7.30% 8 9 10 The capital balances in Table 3 reflect the $20,000,000 addition to common equity and a

11 corresponding $20,000,000 decrease to short-term debt. Thus, if Staffs recommended capital 12 structure properly reflected the make-whole premium discussed above and the $20,000,000 13 equity infusion that occurred in October 2018, Staffs weighted cost of capital would be 7.30%. 14

15 Q. What is the Company’s recommended capital structure and overall cost of capital as updated 16 with actual data through September and the inclusion of the $20,000,000 October equity 17 infusion? 18 A. The Company’s updated recommended capital structure through September 2018 and including 19 the October 2018 equity infusion is shown in Table 4 below: 20 21 22 Cost Weighted 23 Balance Ratio -Rate -Cost 24 25 Short-Term Debt $14,5 12,486 2.25% 2.39% 0.05% Long-Term Debt 309,576,838 47.90% 5.06% 2.42% 26 Preferred Stock 394,162 0.06% 9.06% 0.01% 27 Common Equity 321,795,633 49.79% 10.80% 5.38% 28 Total $646,279,119 100.00% 7.86% West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Rod P. Nevirauskas Page 13 of 13

1 This capital structure includes the make-whole premium discussed above and updates to actual 2 data for short-term debt, long-term debt, and common equity through September 2018, as well 3 as the $20,000,000 equity infusion that occurred in October 2018.

4

5 Q: Do you have any additional comments regarding short-term debt? 6 A: Yes. Both CAD and Staff used a test year average of outstanding STD in their recommended capital 7 structures. Consistent with the Order in the Company’s last rate case, recognizing non-revenue 8 producing capital in rate base for the Addendum Period, the STD balance that should be included 9 for ratemaking purposes is the one the Company recommended, as a portion of the permanent 10 debt and equity financings that took place post-test year served to reduce the STD balance for 11 pro forma purposes. 12

13 Q. Does this conclude your rebuttal testimony? 14 A. Yes. PUBLIC SERVICE COMMISSION OF WEST VIRGINIA CHARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-S-42T

WEST VIRGINIA-AMERICAN WATER COMPANY

Rule 42T Tariff Filing to Increase Rates and Charges

REBUlTAL TESTIMONY OF CHRISTINA E. CHARD

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-421and 18-0576-S-42T Rebuttal Testimony of Christina E. Chard Page 1of 7

1 Q. Please identify yourself. 2 A. My name is Christina E. Chard. I am employed by West Virginia-American Water Company, Inc. 3 (“WVAW” or the “Company”) as the Manager of Rates and Regulatory Support, and I provided 4 direct testimony in this docket filed on April 30, 2018. 5

6 Q. What is the purpose of your rebuttal testimony? 7 A. The purpose of my rebuttal testimony is to address testimony filed by Staff witnesses John Fowler 8 and Tara Gilkey and CAD witness Ralph Smith regarding: 9 10 I. Staff’s and CAD’s Revenue Requirement Recommendations.

11 II. The Company’s Revised Revenue Requirement. 12 Ill. Adjustments to test year water sales for declining residential and commercial usage 13 and the Company’s request for recognition of residential and commercial customer 14 attrition. 15 IV. Staff’s adjustment for present rate revenues as a result of the 2018 Tax Cuts and Jobs 16 Act ( “TCJ A” ) settle ment . 17 V. Staff’s rejection of the Company’s revenue adjustment to unbilled revenue for water 18 and wastewater. 19 Vi. Staff’s rejection of the Company’s revenue adjustments for BPS Printing and the City 20 of Milton. 21 VII. CAD witness Ralph Smith’s statement regarding the Company‘s DSlC eligible 22 investment accounts. 23 24 I. StaWs and CAD’s Revenue Requirement Recommendations

25 Q. Please describe Staff’s and CAD’s revenue requirement recommendations. 26 A. Staff has recommended an overall increase in water revenues of $14,603,754, which includes a 27 reduction to going level revenues by $4,524,977 for the Tax Cuts and Jobs Act (General Order 28 236.1) which will be addressed later in this testimony. CAD recommends a base revenue increase 29 of no more than $13.4 million on an 8.75% return on equity, and a revenue increase of no more West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 2 of 7

1 than $17.5 million on a 9.75% ROE. Each of these recommendations fall short of a meaningful 2 base revenue increase that will allow the Company to recover its cost of service. 3

4 Q. Please explain why Staffs and CAD’s revenue requirement recommendations are inadequate. 5 A. As Company witness John Cox explains in his rebuttal testimony, Staffs and CAD’S continued use 6 of the average historical test year (“HTY”) as the base for establishing new rates is inconsistent 7 with the Commission’s order in the Company’s 2015 rate case, which recognized rate base 8 elements in the ‘transition period’ or Addendum Period, Case Nos 15-0676-W-42T and 15-0675- 9 S-42T (Feb. 24, 2016) (“2015 Rate Case Order). As a result, Staff and CAD’s revenue increase 10 recommendations completely ignores the impact of over approximately $42 million of utility plant

11 investments for non-transmission and distribution (“T&D”) plant that the Company will have 12 made by the time new rates are implemented. 13 14 The refusal to consider capital investments made through the Addendum Period and the rejection 15 of the Company’s proposed declining use and customer attrition adjustments will perpetuate 16 regulatory lag and likely reduce the time between general rate cases, something both the 17 Company and its customers want to avoid. 18

19 Q. Does the implementation of DSlC eliminate excessive regulatory lag? 20 A. No. As Mr. Cox explains, there is no rational basis to the argument that the implementation of 21 DSlC negates the need for terminal treatment of UPlS to reduce regulatory lag. While the DSIC 22 program reduces regulatory lag on recovery of a portion of transmission and distribution 23 investment, there is still another 60% or so of annual capital investment that would not be 24 recovered until the next base rate case. The Company established a rate base position for the 25 Addendum Period in its application for terminal rate base up to the time new rates go in to effect.

26 These expenditures will be in service at the time of the rate order and should be considered 27 known and measurable. 28 29 30 West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 3 of 7

1 II. The Company’s Revised Revenue Requirements

2 Q. Has the Company revised its revenue requirements as a result of the positions taken by Staff 3 and CAD? 4 A. Yes. The Company has updated data on certain cost of service elements that should be considered 5 in a revised revenue requirement. Taken together, these changes to certain cost of service 6 elements result in a revised revenue deficiency of $33M for water and $227K for wastewater. 7 This revision results in a proposed revenue increase that is $340K above the Company’s original 8 filing for both water and wastewater. Since this level is above the published notice in this case, 9 the Company will continue to request the as-filed amount of $32.9 million in total. A revised 10 revenue deficiency calculation and updated exhibits are attached as Exhibit CEC-1. 11 12 111. Usage and Attrition Adjustments

13 Q. Please summarize Staff’s and CAD’s proposed revisions to the Company’s declining use and 14 revenue adjustments. 15 A. Staff witness Jonathan Fowler conducted an analysis of residential and commercial declining 16 consumption per customer and recommended annual adjustments of 450 gallons and 3,750 17 gallons, respectively. Company witness Greg Roach addresses this in his rebuttal testimony. Staff 18 applied Mr. Fowler‘s adjustment over the fourteen months from the end of the historical test year 19 through March 1, 2019. Mr. Smith argues that one year of revenue loss for residential and 20 commercial declining usage should be eliminated from WVAWC’s revenues, not the two years 21 used by the Company. 22

23 Q. Do you agree with Staff’s and CAD’s recommended revenue forecasts? 24 A. No. The Company’s revenue forecast reflects the fact that actual historical test year usage 25 incorporates, on average, approximately one-half of the decline in usage occurring during the test 26 year. Additionally, as Mr. Roach demonstrates, the decline in consumption will continue through 27 2018 and into the rate year. Consequently, the Company calculated the revenue impact of 28 declining consumption to the mid-point of the rate year. The Company believes that this is a 29 reasonable estimate of revenues for the first year that new rates will be in effect. 30 West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 4 of 7

1 Q. Do you agree with Staff witness Fowler’s and CAD witness Smith’s recommendations to reject 2 the Company’s residential and commercial attrition forecast? 3 A. No. Both Fowler and Smith added back all the revenue associated with the Company’s Residential 4 and Commercial Customer Attrition Adjustments 17 (HN) and 18 (Going Level). The Company’s 5 Adjustment 17 for test year attrition for the residential and commercial classes was based on the 6 actual customer decline that occurred during the test year. As such, this adjustment should not 7 be removed. The Company’s Adjustment 18 was based on an actual three-year decline, of which 8 the Company expects to continue into the future. 9 10 While Staff and CAD may point to minor customer count increases for investment-heavy organic 11 growth primarily driven by public private partnership (“PPP”) main extensions and troubled 12 system acquisitions as a solution to the Company’s customer loss, they fail to acknowledge the 13 year over year decline that the Company has experienced to date. Figure CEC-1 below 14 demonstrates that while there are seasonal variances in customer counts, the month over month, 15 year over year decline continues.

~VA~Total Customer Counts 175,000

170.000

t 05.000

I60.000

1jS.t100

150.000

16 West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 5 of 7

1 This trend, coupled with the numerous news articles and U.S. Census Bureau data cited in my 2 direct testimony, support the Company’s position that the decline in customer base will continue 3 into the rate year. Therefore, the Company stands firm on its declining customer attrition 4 adjustment. 5 6 IV. TUA Rate Reduction

7 Q. Did Staff present an adjustment to reduce current rate revenues in response to the TUA 8 proceeding? 9 A. Yes. Staff included an adjustment to account for a reduction in current rate water revenues in 10 the amount of $4,524,977 ($44,384 in sewer revenues). Staff’s calculation of this adjustment is 11 said to be an application of the -3.48% tax reduction agreed to in the bench order for General 12 Order 236.1. 13

14 Q. Did the Company include a similar adjustment? 15 A. No, the Company believes that a reduction of present rate revenues for a temporary TUA order 16 artificially inflates the Company’s requested rate increase by the $4.5 million. The Company’s 17 Rule 42T filings include a federal tax rate of 21%. The temporary TUA negative surcharge was 18 intended to lower the effective tax rate from 35% to 21% from September 1,2018 to when new 19 base rates go into effect in February 2019. Staff’s proposed revenue increase of $14,603,754 is 20 $10,078,777 in comparison to the Company’s and CAD‘S revenue increases. The rate case should 21 be a comparison of the current authorized base rates to proposed base rates. This proposed 22 adjustment has no effect on total revenue requirement, similar to the DSlC surcharge which was 23 eliminated in the filing, and therefore the Commission should reject Staffs adjustment. 24 25 V. Unbilled Revenue

26 Q. Do you agree with Staff witness Gilkey’s adjustment to add back unbilled revenue? 27 A. No. On a going level basis, the Company decreased revenues by $717,591 to account for the 28 elimination of unbilled revenues. The unbilled revenue amount is the difference between the 29 amount of unbilled revenues booked in December 2016 versus the amount booked in December 30 2017; it is not a stand-alone amount booked only in December 2017. The entries for unbilled West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 6 of 7

1 revenues must be considered together when making this adjustment. All accounting accruals, 2 either revenues or expense, must have a corresponding reversal in the following period to 3 properly reflect the revenue or cost for the current accounting period. The Company, for book 4 purposes, used the accrual basis for revenues to account for sales for the twelve-month period of 5 January 1, through December 31, 2017, a 365-day period. If it did not book unbilled revenues in 6 December 2017 and reverse the December 2016 accrual, the measure of revenues would be on a 7 cash basis with the billing period approximating the period December 16,2016 through December 8 15, 2017. This period is based on the assumption that all customers are billed equally during the 9 month, with December 15th as the midpoint. This is obviously still a 365-day billing period, but 10 for financial reporting purposes the Company could not state that revenues are based on the 11 twelve-month period ending December 31,2017 when in actuality it ended December 15, 2017. 12 In developing pro forma revenues for ratemaking purpose, separate adjustments were made to 13 annualize the revenue effect of such factors as changes in the number of customers and increases 14 in rates (such as the DSlC surcharge increase from 0% to 1.09%) that became effective during the 15 historical test year. Therefore, to eliminate any duplication of revenue for ratemaking purposes, 16 unbilled revenue accrued per books must be removed. 17

18 Q. Has the Company consistently made an adjustment to eliminate per book unbilled revenue in 19 its rate cases?

20 A. Yes. The Company has been consistent in eliminating unbilled revenues either as a positive or 21 negative on a going level basis. In the last rate case the adjustment to eliminate unbilled revenues 22 proposed by the Company resulted in an increase to going level revenues of $1.3 million and the 23 Staff accepted that adjustment. 24

25 VI. BPS Printing and City of Milton Adjustments

26 Q. Do you agree with Staff witness Gilkey’s adjustments to add back revenue for BPS Printing and 27 the City of Milton? 28 A. The Company will concede to Ms. Gilkey‘s adjustments to add $421,953 to industrial revenues for 29 the Company’s adjustment that was made for plant closure of BPS Printing because another 30 industrial customer’s consumption has offset the consumption loss. However, the Company West Virginia American Water Company General Order No. 236.1 Rebuttal Testimony of Christina E. Chard Page 7 of 7

1 would like to highlight the fact that BPS printing represented approximately 8% of the Company’s 2 industrial revenues and is now permanently closed. The property is now being considered for 3 parking space for a local university. There is also no certainty that the industrial company whose 4 usage increased in recent months and provided an offset to the consumption loss will continue to 5 this rate of consumption into the future.

6 7 The Company also agrees to add back $91,732 to sale for resale revenue for the City of Milton 8 based on twelve months of actual usage. 9 10 11 VII. DSlC Eligible Investment Accounts

12 Q. Do you agree with CAD witness Smith’s statement on page 24 of his direct testimony regarding 13 plant accounts that are included in the Company‘s DSlC eligible investment adjustments? 14 15 A. No. Mr. Smith states on page 24, lines 10-14 that “Non-revenue producing plant additions in the 16 following accounts are includable in WVAWV’s DSIC.” He goes on to list plant accounts: (1) 17 Account 330.4-Standpipes, (2) Account 331.4-T&D Mains, (3) Account 333.4-Services, Account 18 334.4 Meters, and (4) Account 335.4- Hydrants with the source being CAD-5-B-32 response. The 19 meters account was not listed in the Company’s response to CAD-5-B-32 and has not been 20 considered DSlC eligible to date. Therefore, his assumption that the Company’s addendum period 21 investment in meters will be recovered in a DSlC surcharge is incorrect. In addition, only the 22 eligible investment in these accounts can be recovered in the DSlC surcharge 23

24 Q. Does this conclude your rebuttal testimony? 25 A. Yes. Exhibit CEC-1 Page 1 of 19

West Virginia-American Water Company Summary of Adjustments to Original Filing - Water Only October 12,2018

Adjusted Amount Filed Amount Revised HTY** HTY HTY Rate Base: Culloden Dam (329,845) Pa rkersburg (153,001) ClAC for negative acquisition adjustments (1,750) Cash Working Capital - 12 days for Support Service Costs (86,000) Total Revised Statement B (570,596) 587,312,717 586,742,121

Revenues: BPS Printing 421,953 City of Milton 91,732 Forfeited Discounts 10,736 Change in ProForma Revenue 568,964

Total Revised Statement A, Line 2 1,093,385 159,830,729 160,924,114

Operating Expenses: Compensation - Salaried Employees at 2080 hours (14,978) Compensation - Union Employees - Performance Pay 200,63 1 Compensation related (401K, DCP, ESPP, group insurance) 523 WV - OPEE (237,425) Service Company - OPEE (24,770) Amortization - Regulatory Assets (400,113) IOTG 1,557,869 Uncollecti bles 29,448 Total Revised Statement A, Line 5 1,111,185 60,632,095 61,743,280

**Note: These adjustments will flow through to the addendum period. Exhibit CEC-1 Page 2 oi 19

West Virginia-American Water Company - Water Revenue Requirement For the Twelve Months Ended December 31,2017

Test Year Period: For the Twelve Months Ended December 31,2017 Reference Rule 42 1 Rate Base Statement B $586,742,121 2 3 Rate of Return/Cost of Capital Statement C -7.93% 4 5 Required Utility Operating income (UOI) (Line 1x Line 3) 46,528,650 6 7 Present Rate UOI Statement A 29,598,575 8 9 Income Deficiency (Line 5 - Line 7) 16,930,075 10 11 Gross Revenue conversion factor see below 1.44303 12 13 Revenue Increase (Decrease) requested (Line 9 x Line 11) 24,430,664 14 15

Gross Revenue Conversion Factor 100.0000% Uncollectible% 1.8643% -1.8643% Subtotal 98.1357% State 0& 0 4.4000%

Subtotal 4.4000% 4.4000% -4.3180% 93.8177% State income Tax 6.5000% -6.0982% Subtotal 87.7196% Federal IncomeTax 21.0000% -18.4211% Subtotal 69.2985%

Gross up factor p_ 1.44303

PROOF increase Revenues 24,430,664 Less: Uncollectible @ 1.86% 455,461 Subtotal 23,975,203 Less: State 8&0 @ 4.40% 1,054,909 Subtotal 22,920,294 Less: State Tax @ 6.50% 1,489,819 Subtotal 21,430,475 Less: Federal Tax @ 21% 4,500,400 Increase in UOI 16,930,075 Exhibit CEC-1 Page 3 of 19

West Virginia-American Water Company - Water Statement A Statement of Net income For the Twelve Months Ended December 31,2017

Total Company Water Accounting Per Books Going Level Pro Forma Line Reference Per Books Per Books Adiustments Adjusted Adiustments Going Level Adiustments Pro Forma No. (1) (2) (3) (4) (5) (6) (7) (8) (9) 1 2 Operating Revenues Schedule A1 $144,154,164 $143,240,757 $23,383 $143,264,140 ($6,770,690) $136,493,450 $24,430,664 $160,924,114 3 4 Operating Revenue Deductions: 5 Operation and Maintenance Expenses Schedule A2 38,334,178 37,965,263 37,965,263 23,322,556 61,287,819 455,461 61,743,280 6 Depreciation and Amortization Schedule A3 22,205,753 22,064,308 22,064,308 416,474 22,480,782 22,480,782 7 Taxes Other Than Income Taxes Schedule A4 17,646,699 17,495,576 23,383 17,518,959 91,142 17,610,101 1,054,909 18,665,010 8 Provisions For Income Taxes Schedule A5 26,333.619 26,260,501 26,260,501 (20,744,328) 5,516,172 5,990,219 11,506,391 9 10 Total Deductions 104,520,249 103,785,648 23,383 103,809,031 3,085,844 106,894,875 7,500,589 114,395,464 11 12 Net Operating Income 39,633,915 39,455,109 39,455,109 (9,856,534) 29,598,575 16,930,075 46,528,650 13 14 Non Operating Income: 15 Other Income, net Schedule A6 (699,211) (693,552) (693,552) 693,552 16 Other income Deductions Schedule A7 15,006,378 14,925,349 14,925,349 1,327,408 16,252,757 16,252,757 17 Dividend Preferred Stock 67,850 67,484 67,484 (8,810) 58,674 58,674 18 19 Net income to Common Stock 25,258,898 25,155,828 25,155,828 (11,868,684) 13,287,144 16,930,075 30,217,219 20 21 Average Rate Base Statement B $627,284,026 $624,461,780 $624,461,780 $586,742,121 $586,742,121 22 23 Rate of Return 6.32% 6.32% 6.32% 5.04% 7.93% 24 25 Exhibit CEC-1 Page 4 of 19

West Virginia-American Water Company - Water Statement B Average Rate Base For the Thirteen Months Ended December 31,2017

12/31/2017 13 Months Avg Accounting Per Books Going Level Line Reference Water Adiustments Adjusted Adiustments Going Level No. (1) (2) (3) (4) (5) (6) (7) 1 2 Utility Plant In Service Schedule B1 $887,394,658 $843,601,034 $843,601,034 ($923,022) $842,678,012 3 Utility Plant Held For Future Use Schedule 82 4 Unclassified Plant In Service Schedule 83 5 Construction Work In Progress - Completed And In Service Schedule 84 6 Total Utility Plant 887,394,658 843,601,034 843,601,034 (923,022) 842,678,012 7 8 Accumulated Provision For Depreciation, 9 Depletion And Amortization Schedule 88 89,504,651 83,546,060 83,546,060 83,546,060 10 Retirement Work In Progress Schedule B9 (108,614) (91,514) (91,514) (91,514) 11 Contributions In Aid Of Construction Schedule 610 76,178,086 73,459,497 73,459,497 73,459,497 12 Customer Advances Schedule Bll 12,455,790 14,118,884 14,118,884 14,118,884 13 Total 178,029,914 171,032,927 171,032,927 171,032,927 14 15 Net Investment In Utility Plant 709,364,745 672,568,107 672,568,107 (923,022) 671,645,085 16 17 Working Capital Allowance: 18 Materials And Supplies Schedule B5 3,282,109 2,620,752 2,620,752 661,357 3,282,109 19 Prepayments Schedule B6 20 Working Cash Schedule 87 1,075,000 1,075,000 1,075,000 1,075,000 21 Total Working Capital Allowance 4,357,109 3,695.752 3.695.752 661,357 4.357.109 22 23 Other: 24 Customer Deposits Schedule 812 25 Accumulated Deferred Income Taxes Schedule 813 (93,648,373) (84,512,273) (84,512,273) (9,136,100) (93,648,373) 26 Accumulated Deferred Investment Tax Credit Schedule 814 (418,579) (438,280) (438,280) 19,701 (418,579) 27 Other Deferred Debits Schedule 815 4,806,878 4,842,722 4,842,722 (35,843) 4,806,878 28 Other Deferred Credits Schedule 816 29 Total Other Rate Base (89,260,073) (80,107,831) (80,107,831) (9,152,242) (89,260,073) 30 31 Total Average Rate Base $624,461,780 $596,156,028 $0 $596,156,028 ($9,413,907) $586,742,121 Exhibit CEC-I Page 5 Of 19

West Virginia-American Water Company - Water Revenue Requirement For the Period January 1,2018 through February 28,2019

Addendum Period: For the Period January 1,2018 through February 28,2019 Reference Addendum Period 1Rate Base Statement B $649,644,867 2 3 Rate of Return/Cost of Capital Statement C -7.86% 4 5 Required Utility Operating Income (UOI) (Line 1x Line 3) 51,062,087 6 7 Present Rate UOi Statement A 45,107,291 8 9 Income Deficiency (Line 5 - Line 7) 5,954,796 10 11 Gross Revenue conversion factor see below 1.44303 12 13 Revenue Increase (Decrease) requested (Line 9 x Line 11) 8,592,969 14 15

Gross Revenue Conversion factor 100.0000% Uncollectible% 1.8643% -1.8643% Subtotal 98.1357% State B& 0 4.4000%

Subtotal 4.4000% 4.4000% -4.3180% 93.8177% State Income Tax 6.5000% -6.0982% Subtotal 87.7196% Federal Income Tax 21.0000% -18.4211% Subtotal 69.2985%

Gross up factor 1.44303

PROOF increase Revenues 8,592,969 Less: Uncollectible @ 1.86% 160,199 Subtotal 8,432,770 Less: State B&O @ 4.40% 371,042 Subtotal 8,061,728

Less: State Tax @J 6.50% 524,012 Subtotal 7,537,716 Less: Federal Tax @ 21% 1,582,920 Increase in UOi 5,954,796 Exhibit CEC-1 Page 6 of 19

West Virginia-American Water Company - Water Statement A - Addendum Statement of Net Income For the Period January 1,2018 through February 28,2019

Pro Forma per Accounting Per Books Going Level Pro Forma Line Reference Rule 42 filing Adjustments Adiusted Adjustments Going Level Adjustments Pro Forma No. (1) (2) (3) (4) (5) (6) (7) (8) 1 2 Operating Revenues Schedule A1 $160,924,114 $0 $160,924,114 $0 $160,924,114 $8,592,969 $169,517,083 3 4 Operating Revenue Deductions: 5 Operation and Maintenance Expenses Schedule A2 61,743,280 61,743,280 61,743,280 160,199 61,903,479 6 Depreciation and Amortization Schedule A3 22,480,782 22,480,782 1,814,587 24,295,369 24,295,369 7 Taxes Other Than Income Taxes Schedule A4 18,665,010 18,665,010 18,665,010 371,042 19,036,052 8 Provisions For Income Taxes Schedule A5 11,506,391 11,506,391 (393,228) 11,113,163 2,106,933 13,220,096 9 10 Total Deductions 114,395,464 114,395,464 1,421,359 115,816,823 2,638,174 118,454,996 11 12 Net Operating income 46,528,650 46,528,650 (1,421,359) 45,107,291 5,954,795 51,062,087 13 14 Non Operating Income: 15 Other Income, net Schedule A6 16 Other Income Deductions Schedule A7 16,252,757 16,252,757 (206,529) 16,046,228 16,046,228 17 Dividend Preferred Stock 58,674 58,674 6,290 64,964 64,964 18 19 Net income to Common Stock 30,2 17,2 19 30,217,219 (1,221,120) 28,996,099 5,954,795 34,950,895 20

21 Rate Base Statement B ~ $586,742,121 $586,742,121 $649,644,867 $649,644,867 22 23 Rate of Return 7.93% 7.93% 6.94% 7.86% Exhibit CEC-1 Page 7 of 19

West Virginia-American Water Company - Water Statement B - Addendum Terminal Rate Base For the Period January 1,2018 through February 28,2019

Water Accounting Per Books Going Level Line Reference 2/28/2019 Adjustments Adjusted Adjustments Going Level No. (1) (2) (3) (4) (5) (6) 1 2 Utility Plant In Service Schedule B1 $943,289,721 $943,289,721 ($719,422) $942,570,299 3 Utility Plant Held For Future Use Schedule 82 4 Unclassified Plant In Service Schedule 83 5 Construction Work In Progress - Completed And In Service Schedule 84 6 Total Utility Plant 943,289,72 1 943,289.721 (719.422) 942.570.299 7 8 Accumulated Provision For Depreciation, 9 Depletion And Amortization Schedule 88 107,867,868 107,867,868 107,867,868 10 Retirement Work In Progress Schedule 89 (108,614) (108,614) (108,614) 11 Contributions In Aid Of Construction Schedule B10 76,178,086 76,178,086 76,178,086 12 Customer Advances Schedule Bll 12,455,790 12,455,790 12,455,790 13 Total 196,393,131 196,393,131 196.393.131 14 15 Net Investment In Utility Plant 746,896,591 746,896,591 (719,422) 746,177,169 16 17 Working Capital Allowance: 18 Materials And Supplies Schedule B5 3,282,109 3,282,109 3,282,109 19 Prepayments Schedule 86 20 Working Cash Schedule 87 1,586,000 1,586,000 1,586,000 21 Total Working Capital Allowance 4,868,109 4,868,109 4,868,109 22 23 Other: 24 Customer Deposits Schedule B12 25 Accumulated Deferred Income Taxes Schedule B13 (105,702,704) (105,702,704) (105,702,704) 26 Accumulated Deferred investment Tax Credit Schedule 814 (418,579) (418,579) (418,579) 27 Other Deferred Debits Schedule 815 4,720,872 4,720,872 4,720,872 28 Other Deferred Credits Schedule 816 29 Total Other Rate Base (101,400,410) (101,400,410) (101,400,410) 30 31 Total Rate Base $650,364,289 $0 $650,364,289 ($719,422) $649,644,867 Dihibif CEC-l Page 8 of 19

West Virginia-AmericanWater Company Statement C - Addendum Detail of Equity and/or Debt Capital For the PeriodJanuary 1,2018 through February 28,2019

- Total Company Total Company Unamortized Amounts Percent of Effective Cost Weighted Line Reference perBooks Adiustments Issue Costs Totai tocompany -costs No. Class of Capital (11 (2) (31 (41 (5) (6) (71 (8) 1 2 Shart-Term Debt $14,512,486 $0 $0 $14,512,486 2.250% 2.387% 0.050% 3 4 Long-Term Debt Page 2 318,006,704 (8,021) (8,421,845) 309,576.838 47.900% 5.059% 2.420% 5 6 Preferred Stock Page 3 600,000 (200,000) (5,838) 394,162 0.060% 9.065% 0.010% 7 8 Common Equity 321,795,633 - 321,795,633 49.790% 10.800% 5.380% 9 10 Totalcapital $654,914,824 ($208,021) ($8,427,684) $646,279,119 100.000% 7.860% West Virginia-AmericanWater Company Statement C - Addendum Detail of Equity and/or Debt Capital For the Period January 1,2018 through February28.2019

Debt lssueiype Interest Date Maturity Face Amount Outstanding Long-Term Sinking Unamortized Annual Carrying Annual Interest Annual Arnortz Annual

Line Cowon Rate ___Rate ___IDate Prlncioaim * DebtExpense -cost of Issue Ex@ -cost No. (1) (21 (31 (4) (5) (61 (71 (8) (9) (10) (11) (12) (131 (14) 1 General Mortgage Bonds 2 3 9.06 % SERIES 9.06% 11/21/199l 11/1/2031 13,000,000 13,000,000 $13,000.000 $30,703 $12,969,297 $1,177,800 52,423 $1,180,223 4 6.87 % SERIES 6.87% 12/28/1993 12/1/2033 11,800,000 11,500,oM) 11,500,000 28.913 11,471,087 790.050 1,960 792,010 5 8.19 % SERIES 8.19% 5/12/1995 4/1/2025 7,500,000 7,500,000 7,500,000 14,252 7.485.748 614,250 2,342 616,592 6 7.54 % SERIES 7.54% 4/12/1996 4/1/2026 11.000,000 11,000,000 11,000,000 18,796 10,981,204 829,400 2,652 832,052 7 7.19 % SERIES 7.19% 2/28/1997 2/28/2027 7,000,000 7,000,000 7,000,000 16.590 6,983,410 503,300 2,075 505,375 8 9 Unsecured Debt With AWCC 10 11 5.77 % SERIES 5.77% 12/21/2QoE 12/21/2021 31,950,000 31,950,000 31,950,000 19,876 31,930,124 1,843,515 7,077 1,850,892 12 5.62 % SERIES 8.62% 3/29/2007 3/29/2019 13 5.77 % SERIES 8.77% 3/29/2007 3/29/2022 14 7.21 O/o SERIES 7 21% 5fl¶/Z009 5/19/2019 15 4.30 % SERIES 4.30% 3/22/2013 12/1/2042 19,705,000 19,705,000 19,705,000 189,715 19,515,288 847,315 7,919 855,234 16 385 % SERIES 3.85% 11/20/2013 3/1/2024 s8.000.000 58,000,000 s8.000.000 333,477 57,666,523 2,233,000 66.658 2,299,658 17 2.95 %SERIES 2.95% 9/13/2017 9/1/2027 44,895.546 44,895,546 44,895,546 5,435,998 39,459,548 1,324,419 639,626 1,964,045 18 4.20 %SERIES 4 20% 8/9/2018 9/1/2048 60,000,000 60,000,000 60,000,000 647,629 59,352,371 2,520,000 21,774 2,541.774 19 3.75 %SERIES 3.75% 9/11/2018 9/1/2028 27,745,027 27,745,027 27,745,021 236,939 27,508,088 1,040,439 29,234 1,069,672 20 4.06 %SERIES 4.06% 1/31/2019 1/31/2049 24,800,000 24,500,000 24,500.000 488,639 24,011,361 993,475 16,333 1,009,808 21 22 Cariy Over Unamortized Debt Expense 23 24 5.90 % SERIES 0.00% 10/1/2004 9/30/2034 360,267 (360,267) 23,119 23,119 25 10.00 % SERIES 0 00% 12/1/2013 3/1/2024 600,052 (600,052) 119,944 119,944 26 27 Unsecured Debt with WVDWTRF 78 0% Loan - Grant 0.00% 1/29/2010 9/1/2031 1,211,131 1,211,131 (8,021) 1,203,110 i,203,110 29 30 Total Long-Term Debt $318,006,704 $318,006,704 so ($8,021) $317,998,683 $8,421,845 $309,576,838 $14,716,962 $943,137 $15,660,099 31 32 Embedded Cost of Long-Term Debt (14)/(11) = - 5.06% 33 34 Note 1 The 5.90% Series on Line 20 was called eariv due to the RWE sale of AWW common stock. This transaction resulted in a gain on that early tali which the 35 Company is netting against the unamoritzeddebt expense and amortizing over the original term of the loan. 36 37 Note 2: The $1.925 miilion, 0% loan was obtained from the West Virginia DWTRF Program. Lxhlbit CLC-1 Page 10 Of 19

West Virginia-American Water Company Statement C -Addendum Detail of Equity andfor Debt Capital For the Period January 1,2018 through February28,2019

Dividend Rate Dividend Number of Shares Date Sinking Outstanding Unamortized Net Annual Annual Amortz Annual Line Type. Par Value -Rate Authorized bvcharter -Fund Adiustments 1ssueExPense - oivideods of IssueExil -cost No 11) (21 (31 (41 (51 (61 (71 (8) (91 (101 (111 (12) (131 1 2 8.85 %SERIES 8.85% 20,000 ii/2i/i99i ($200,000) %nn,ono $400.000 $5,838 $394,162 $35.400 $330 $35,730 3 4 5 Total 20,000 $0 $4oo,mn 6 7 8 Embedded Cost of Preferred Stock (13) /(lo) = 9.06% Exhibit CEC-I Page 11 oi 19

West Virginia American Water Company - Wastewater Revenue Requirement For the Twelve Months Ended December 31,2017

Test Year Period: For the Twelve Months Ended December 31,2017 Reference Rule 42 1 Rate Base Statement 8 $2,679,553 2 3 Rate of Return/Cost of Capital Statement C -7.93% 4 5 Required Utility Operating Income (UOI) (Line 1x Line 3) 212,489 6 7 Present Rate UOI Statement A 116,050 8 9 Income Deficiency (Line 5 -Line 7) 96,439 10 11 Gross Revenue conversion factor see below 1.42016 12 13 Revenue Increase (Decrease) requested (Line 9 x Line 11) 136,958 14 15

Gross Revenue Conversion Factor 100.0000% Uncollectible% 1.8643% -1.8643% Subtotal 98.1357% State B& 0 2.8600%

Subtotal 2.8600% 2.8600% -2.8067% 95.3290% State Income Tax 6.5000% -6.1964% Subtotal 89.1326% Federal Income Tax 21.0000% -18.7179% Subtotal 70.4148%

Gross up factor 1.42016

PROOF Increase Revenues 136,958 Less: Uncollectible @ 1.86% 2,553 Subtotal 134,405 Less: State B&O @ 2.86% 3,844 Subtotal 130,561 Less: State Tax @ 6.50% 8,486 Subtotal 122,075 Less: Federal Tax @ 35% 25,636 Increase in UOI 96,439 Exhibit CEC-I Page 12 of 19

West Virginia American Water Company - Wastewater Statement A Statement of Net Income For the Twelve Months Ended December 31,2017

Total Company Wastewater Accounting Per Books Going Level Pro Forma Line Reference Per Books Per Books Adjustments Adiusted Adiustments Going Level Adiustments Pro Forma No. (1) (2) (3) (4) (5) (6) (7) (8) (9) 1 2 Operating Revenues Schedule A1 $144,154,164 $913,407 $0 $913,407 ($2,365) $911,042 $136,958 $1,048,000 3 4 Operating Revenue Deductions: 5 Operation and Maintenance Expenses Schedule A2 38,334,178 368,915 368,915 117,975 486,890 2,553 489,443 6 Depreciation and Amortization Schedule A3 22,205,753 141,445 141,445 5,125 146,570 146,570 7 Taxes Other Than Income Taxes Schedule A4 17,646,699 151,123 151,123 (3,842) 147,281 3,844 151,125 8 Provisions For Income Taxes Schedule A5 26,333,619 73,270 73,270 (59,019) 14,251 34,122 48,373 9 10 Total Deductions 104,520,249 734,753 734,753 60,239 794,992 40,519 835,511 11 12 Net Operating Income 39,633,915 178,654 178,654 (62,604) 116,050 96,439 212,489 13 14 Non Operating Income: 15 Other Income, net Schedule A6 (699,211) (5,659) (5,659) 5,659 16 Other Income Deductions Schedule A7 15,006,378 81,029 81,029 (5,257) 75,772 75,772 17 Dividend Preferred Stock 67.850 366 366 (98) 268 268 18 19 Net Income to Common Stock $25,258,898 $102,918 $0 $102,918 ($62,908) $40,010 $96,439 $136,449 20 21 Average Rate Base Statement B $627,284,026 $2,822,245 $2,822,245 $2,679,553 $2,679,553 22 23 Rate of Return 6.32% 6.33% 6.33% 4.33% 7.93% Exhibit CEC-I Page 13 of 19

West Virginia American Water Company - Wastewater Statement B Average Rate Base For the Thirteen Months Ended December 31,2017

12/31/2017 13 Months Avg Accounting Per Books Going Level Line Reference Wastewater Wastewater Adjustments Adiusted Adiustments Goinn Level No. (1) (2) (3) (4) (5) (6) (7) 1 2 Utility Plant In Service Schedule 81 $7,750,168 $7,550,001 $0 $7,550,001 $0 $7,550,001 3 Utility Plant Held For Future Use Schedule 82 4 Unclassified Plant In Service Schedule 63 5 Construction Work In Progress - Completed And In Service Schedule 84 6 Total Utility Plant 7,750,168 7,550,001 7,550,001 7,550,001 7 8 Accumulated Provision For Depreciation, 9 Depletion And Amortization Schedule B8 3,127,442 3,070,860 3,070,860 3,070,860 10 Retirement Work In Progress Schedule 89 36,424 35,532 35,532 35,532 11 Contributions In Aid Of Construction Schedule 610 1,716,065 1,716,065 1,716,065 1,716,065 12 Customer Advances Schedule Bll 13 Total 4,879,931 4,822,457 4,822,457 4,822,457 14 15 Net Investment In Utility Plant 2,870,237 2,727,544 2,727,544 2,727,544 16 17 Working Capital Allowance: 18 Materials And Supplies Schedule 85 5,008 3,978 3,978 1,030 5,008 19 Prepayments Schedule B6 20 Working Cash Schedule 67 (53,000) (53,000) (53,000) (53,000) 21 Total Working Capital Allowance (47.9921 (49.0221 (49 0221 1030 147 9971 22 23 Other: 24 Customer Deposits Schedule 812 25 Accumulated Deferred Income Taxes Schedule 813 26 Accumulated Deferred Investment Tax Credit Schedule 814 27 Other Deferred Debits Schedule 815 28 Other Deferred Credits Schedule 816 29 Total Other Rate Base 30 31 Total Average Rate Base $2,822,245 $2,678,523 $0 $2,678,523 $1,030 $2,679,553 32 33 34 35 Exhibit CEC-1 Page 14 of 19

West Virginia American Water Company - Wastewater Revenue Requirement For the Period January 1,2018 through February 28,2019

Addendum Period: For the Period January 1,2018 through February 28,2019 Reference Addendum Period 1Rate Base Statement 8 $3,380,469 2 3 Rate of Return/Cost of Capital Statement C 7.86% 4 5 Required Utility Operating Income (UOI) (Line 1x Line 3) 265,705 6 7 Present Rate UOI Statement A 202,186 8 9 Income Deficiency (Line 5 - Line 7) 63,519 10 11 Gross Revenue conversion factor see below 1.42016 12 13 Revenue Increase (Decrease) requested (Line 9 x Line 11) 90,206 14 15

Gross Revenue Conversion Factor 100.0000% Uncollectible% 1.864’3% -1 8fid?U Subtotal 98.1357% State B& 0 2.8600%

Subtotal 2.8600% 2.8600% -2.8067% 95.3290% State Income Tax 6.5000% -6.1964% Subtotal 89.1326% Federal Income Tax 21.0000% -18.7179% Subtotal 70.4148%

Gross up factor 1.42016 P~

PROOF Increase Revenues 90.206 Less: Uncollectible @ 1.86% 1,682 Subtotal 88,524 Less: State 8&0 @ 2.86% 2,532 Subtotal 85,992 Less: State Tax @ 6.50% 5,590 Subtotal 80,403 Less: Federal Tax @ 21% 16.885 I- Increase in UOI 63.518 Exhibit CEC-1 Page 15 of 19

West Virginia American Water Company - Wastewater Statement A - Addendum Statement of Net Income For the Period January 1,2018 through February 28,2019

Pro forma per Accounting Per Books Going Level Pro Forma Cine Reference Rule 42 filing Adiustments Adiusted Adiustments Going Level Adiustments Pro Forma

1 2 Operating Revenues Schedule A1 $1,048,000 $0 $1,048,000 $0 $1,048,000 $90,206 $1,138,206 3 4 Operating Revenue Deductions: 5 Operation and Maintenance Expenses Schedule A2 489,443 489,443 489,443 1,682 491,124 6 Depreciation and Amortization Schedule A3 146,570 146,570 17,255 163,825 163,825 7 Taxes Other Than Income Taxes Schedule A4 151,125 151,125 151,125 2,532 153,657 8 Provisions For Income Taxes Schedule A5 48,373 48,373 (6,952) 41,421 22,474 63,895 9 10 Total Deductions 835,511 835,511 10,303 845,814 26,688 872,501 11 12 Net Operating Income 212,489 212,489 (10,303) 202,186 63,518 265,705 13 14 Non Operating Income: 15 Other Income, net Schedule A6 16 Other income Deductions Schedule A7 75,772 75,772 9,348 85,120 85,120 17 Dividend Preferred Stock 268 268 70 338 338 18 19 Net Income to Common Stock $136,449 $0 $136,449 ($19,721) $116,728 $63,518 $180,247 20 21 Rate Base Statement B $2,679,553 $2,679,553 $3,380,469 $3,380,469 22 23 Rate of Return 7.93% 7.93% 5.98% 7.86% Exhibit CEC-1 Page 16 of 19

West Virginia American Water Company - Wastewater Statement B -Addendum Terminal Rate Base For the Period January 1,2018 through February 28,2019

Wastewater Accounting Per Books Going Level Line Reference 2/28/2019 Adjustments Adjusted Adjustments Going Level No. (1) (2) (3) (4) (5) (6) 1 2 Utility Plant In Service Schedule 81 $8,357,727 $8,357,727 $0 $8,357,727 3 Utility Plant Held For Future Use Schedule 82 4 Unclassified Plant In Service Schedule 83 5 Construction Work In Progress - Completed And In Service Schedule 84 6 Total Utility Plant 8,357,727 8,357,727 8,357,727 7 8 Accumulated Provision For Depreciation, 9 Depletion And Amortization Schedule 88 3,216,202 3,216,202 3,216,202 10 Retirement Work In Progress Schedule B9 (0) (0) (0) 11 Contributions In Aid Of Construction Schedule 810 1,716,065 1,716,065 1,716,065 12 Customer Advances Schedule 811 13 Total 4,932,267 4,932,267 4,932,267 14 15 Net Investment In Utility Plant 3,425,460 3,425,460 3,425,460 16 17 Working Capital Allowance: 18 Materials And Supplies Schedule 85 5,008 5,008 5,008 19 Prepayments Schedule 86 20 Working Cash Schedule 87 f50.0001 f50.0001 f50.0001 21 Total Working Capital Allowance (44.9921 (44.9921 f44.9921 22 23 Other: 24 Customer Deposits Schedule 812 25 Accumulated Deferred Income Taxes Schedule 813 26 Accumulated Deferred Investment Tax Credit Schedule 814 27 Other Deferred Debits Schedule 815 28 Other Deferred Credits Schedule 816 29 Total Other Rate Base 30 31 Total Average Rate Base $3,380,469 $0 $3,380,469 so $3.380.469 32 West Virginia-lmerican Water Company Statement C - Addendum Detail of Equity andfor Debt Capital For the Period January 1,2018 through February 28,2019

Total Company rota1 company Unamortized Amounts Percent of Effective Cost Weighted

Line Reference Adjustments Issue Costs ITotal toCompany __costs Class of Capital No. (1) (2) (31 (4) (5) (61 (7) (81 1 2 Shoe-Term Oebt $14,512,486 SO SO $14,512,486 2.250% 2.387% 0.050% 3 4 Long-Term Debt Page 2 318,006,704 (8,0211 (8,421,845) 309,576,838 47.900% 5.059% 2.420% 5 6 Preferred Stock Page 3 600,000 (200,000) (5,838) 394.162 0.060% 9.065% 0.010% 7 8 Common Equity 321,795,633 - 321,795.633 49.790% 10.800% 5.380% 9 10 Total Capital $654,914,824 ($208.021) ($8,427,684) $646,279,119 lW.000% 7.860% Exhibit CFC-l vase 18 Of 19

West Virginia-AmericanWater Company Statement C .Addendum Detail of Equity and/or Debt Capital For the Period January 1.2018 through February 20,2019

Debt IssueType Interest Date Maturity Face Amount Outstanding Long-Term Sinking Unamortized Annual Carrying Annual Interest Annual Amortz Annual Line Coupon Rate Rate __Date mpeieaoks && Fundn Debt Expense ValUe -cost of Issue Exg -cost NO. (11 I21 (31 (41 61 (61 (7) (81 191 110) (111 (121 1131 1141 1 General Mortgage Bonds 2 3 9.06 % SERIES 9 06% 11/21/1991 11/1/2031 $13,000,000 $13,000,000 $0 $13,000,000 $30,703 $12,969,297 $1,177,800 $2,423 $1,180,223 4 6.87 % SERIES 6.87% 12/28/1993 12/1/2033 11,800,000 11,soo,o00 11,5w,ooo 28,913 11,471,087 790,080 1,960 792,010 8 8.19 % SERIES 8.19% 511211995 4/1/2025 7,800,000 7,SOO,WO 7,500,000 14,282 7,485,748 614,250 2,342 616,592 6 7.54 % SERIES 7.84% 4/12/1996 4/1/2026 11,000,000 11,000,000 11,000,000 18,796 10,981,204 829,400 2,652 832,052 7 7.19 % SERIES 7.19% 2/28/1997 2/28/2027 7,000,000 7,000,000 7,000,000 16.890 6,983,410 503,300 2,075 808,375 8 9 Unsecured Debt With AWCC 10 11 5.77 % SERIES 5.77% 12/21/2006 12/21/2021 31,980,000 31,950,000 31,980,000 19,876 31,930,124 1,843,818 7,077 1,880,592 12 8.62 % SERIES 5.62% 3/29/2007 3/29/2019 13 577 % SERIES 5.77% 3/29/2007 3/29/2022 14 7.21 % SERIES 7.21% 5/19/2009 8/19/2019 18 4.30 % SERIES 4.30% 3/22/2013 12/1/2042 19,708,000 19,708,000 19,7OS,WO 189,715 19,818,288 847,318 7,919 888,234 16 3.85 % SERIES 3.85% 11/20/2013 3/1/2024 88,000,000 88,000,000 88,000,000 333,477 57,666,823 2,233,000 66,688 2,299,658 17 2.98 %SERiES 2.95% 9/13/2017 9/1/2027 44,895,546 44,895,846 44,895,846 5,438,998 39,459,548 1,324,419 639,626 1,964,045 18 4.20 %SERIES 4.20% 8/9/2018 9/1/2048 60,000,000 60,000,000 60.d00,WO 647,629 89,352,371 2,520,000 21,774 2,841,774 19 3.78 %SERIES 3.75% 9/11/2018 9/1/2028 27,745,027 27,748,027 27,745,027 236,939 27,508,088 1,040,439 29,234 1,069,672 20 4.06 %SERIES 4.06% 1/31/2019 1/31/2049 24,800,000 24,500,000 24,500,000 488,639 24,011,361 993,478 16,333 1,009,808 21 22 Carry Over Unamortized Debt Expense 23 24 5.90 % SERIES 0.00% 10/1/2004 9/30/2034 360,267 (360,267) 23,119 23,119 28 10.00 % SERIES 0.00% 12/1/2013 3/1/2024 600.052 (600.0521 119,944 119,944 26 27 Unsecured Debt with WVDWTRF 28 0% Loan -Grant 0.00% 1/29/2010 9/1/2031 1,211,131 1,211,131 18,0211 1,203,110 1,203,110 29 30 Total Long-Term Debt $318,006,704 $318,006,704 so ($8,021) $317,998,683 $8,421,845 $309,876,838 $14,716,962 $943,137 $15,660,099 31 32 Embedded Cost of Long-Term Debt (14)/(11) = 5.06% = 33 34 Note 1: The 5.90% Series on line 20 was called early due to the RWE_rale of AWW common stock. This transaction resulted In a galn on that early call which the 38 Company is netting against the unamoritzeddebt expense and amortizingover the originalterm of the ban. 36 37 Note 2 The $1.925 million, 0% loan was obtained from the West Virginia DWTRF Program. Exhibit CR.1 Pn$e19of 19

West Virginia-American Water Company Statement C - Addendum Detail of Equity andfor Debt Capital For the Period January 1.2018 through February 28,2019

Dividend Rate Dividend Number of Shares Date Sinking Outstanding Unamortized Net Annual Annual Arnortz Annual

Line Tyoe. Par Value __Rate Authorized by Charter lrrued _I_Fund Adiustmentr AmDunt issueExDense - - ofissue Exp cost I_ NO. (1) (2) (3) (4) (5) (6) (71 (8) (9) (10) (111 (121 (13) 1 2 8.85 %SERIES 835% 20,000 11/21/1991 ($200.0001 $600,000 $400,000 $5.838 $394~62 $35,400 $330 $3 5,7 3 0 3 4 5 Total 20,000 ($200,000) $600,000 $0 $400,000 $35,400 $330 $35.730 6 7 8 Embedded Cost of Preferred Stock (13)/(10)= 9.06% 9 10 PUBLIC SERVICE COMMISSION OF WEST VIRGINIA CHARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-S-42T

WEST VIRGINIA-AMERICAN WATER COMPANY

Rule 42T Tariff Filing to Increase Rates and Charges

REBU?TAL TESTIMONY OF JAMIE D. HAWN

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 1of 12

1 Q. Please state your name and business address. 2 A. My name is Jamie D. Hawn. My business address is 1025 Laurel Oak Road, Voorhees, NJ 08043. I

3 am employed by American Water Works Service Company, Inc. (“AWWSC” or the “Service 4 Company”) as a Senior Manager of Regulatory Services. 5 6 Q. Have you provided direct testimony in the case? 7 A. Yes. I provided direct testimony in Case No. 18-0573-W-42T and 18-0576-S-42Tfiled on April 30,

8 2018. Additionally, I will be adopting the direct testimony of Daniel P. Hunnell 11 relating to 9 production expenses (i.e. Power, Chemicals, Purchased Water and Waste Disposal). 10 11 Q. What is the purpose of your rebuttal testimony in the case? 12 A. In this rebuttal testimony, I will address the expense adjustments to Rule 42 and Addendum 13 Period, for water and wastewater, raised in the direct testimony filed by Staffs witness Tara L. 14 Gilkey (“Staff”), and CAD’S witness Ralph Smith (“CAD”). 15 16 Compensation and Compensation-Related Expenses 17 Q. Please describe Staff’s and CAD’s adjustment to total compensation expense? 18 A. Both Staff and CAD have proposed to reduce compensation expense by two factors: (1)capturing 19 only the 2017 historical test year (“HTY) hours worked and (2) removing the going level pay rate 20 through the end of the rate year for the non-union hourly and salary employees. The overall 21 impact of these adjustments to the Company’s compensation expense is a reduction of $656,527 22 and $934,588 by Staff and CAD, respectively. The total impact of the adjustments includes Staff’s 23 and CAD’s rejection of the Company’s use of 2,088 hours to determine annual pay. 24 25 Q. Please explain why Staff’s and CAD‘s use of 2,080 hours to calculate base wage expense for 26 hourly employees is inappropriate. 27 A. The number of work hours in a twelve-month period can vary from 2,080 to 2,096 work hours. 28 Any twelve-month period that begins on a Saturday or Sunday will have 2,080 work hours, with 29 the exception of a leap year, which will have 2,088 hours. Any twelve-month period that begins West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-421 Rebuttal Testimony of Jamie D. Hawn Page 2 of 12

1 on a Monday through Friday will have 2,088 work hours, except the leap year, which will have 2 2,096. The table below provides an analysis of the work days and hours for 2017-2021.

2017 Work Days/Hours Month January February March April May June July August September October November December Total WorkDays 22 20 23 20 23 22 21 23 21 22 22 21 260 HOUK 176 160 184 160 184 176 168 184 168 176 176 168 2080

2018 Work Days/Houn Month January February March April May June July August September October November December Total WorkDays 23 20 22 21 23 21 22 23 20 23 22 21 261 Hours 184 160 176 168 184 168 176 184 160 184 176 168 2,088

2019 Work Days/Hours Month January February March April May June July August September October November December Total WorkDays 23 20 21 22 23 20 23 22 21 23 21 22 261 HOUK 184 160 168 176 184 160 184 176 168 184 168 176 2,088

2020 Work Days/Hours Month January February March April May June July August September October November December Total WorkDays 23 20 22 22 21 22 23 21 22 22 21 23 262 Hours 184 160 176 176 168 176 184 168 176 176 164 184 2,096

2021 Work Davs/Hours Month January February March April May June July August September October November December Total WorkDays 21 20 23 22 21 22 22 22 22 21 22 23 261 3 Hours 168 160 184 176 168 176 176 176 176 168 176 184 2,088 4 The table shows the number of work hours in a calendar year, averaged over the five-year period 5 shown, is 2,088. This analysis demonstrates why the Company used 2,088 hours to calculate base 6 wages for hourly employees and supports the calculation used by the Company in developing 7 compensation and compensation related expenses.

8

9 Q. Do you agree with the adjustment made by Staff and CAD for total compensation expense? 10 A. No, with one exception. While the Company continues to propose the use of 2,088 hours worked

11 for union and non-union hourly employees, it is willing to adjust salaried employee hours to be 12 2,080, since the pay of that employee category is fixed on a 2,080 hour annual cycle. The Company 13 does not agree with limiting recovery based on 2017 HTY hours or current pay rates for non-union 14 hourly and salary employees. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Jamie 0. Hawn Page 3 of 12

1 The going level of compensation expense should be based on the number of employees proposed 2 in this case (317) multiplied by the annual hours in the pay year cycle. As Company witness Mount

3 explains in her direct testimony (p. 2), the Company has added positions since its last rate case

4 to better serve our customers. Employees that are providing high quality service to our

5 customers should not be excluded from cost recovery. Using HTY hours worked does not reflect

6 the full employee going level costs. Additionally, whereas Staff included terminated employees in 7 using the HTY, the Company excluded them and replaced them with the respective new hires or 8 open positions for its adjustment, a forward-looking approach designed to reflect the Company’s 9 projected experience during the rate year. As noted above, CAD incorrectly used 2,080 for hourly 10 new hires and open positions.

11 In addition to underestimating the working hours, Staff and CAD also failed to incorporate 12 the proper going level wage rates for non-union hourly and salary employees. Based on the

13 Company’s long experience with wage increases for non-union hourly and salary employees 14 (which includes wage increases each year since the last rate case), the Company has adjusted non- 15 union hourly and salary wage rates in the rate year (2/26/19 to 2/25/20). The Company’s going 16 level adjustment is based on a three-year average of pay rate increases from 2015 through 2017. 17 To ensure that Company wage and salary rates for these categories of employees remain

18 competitive, it must provide for periodic increases, and it has done so on an annual basis; neither 19 Staff nor CAD has suggested otherwise. The Company contends that these rate-year pay 20 increases are known and measurable wage and salary calculations for ratemaking purposes. 21 22 Thus, the Company asks the Commission to reject the Staff’s and CAD’S proposed 23 adjustments to reduce compensation by the HTY hours worked and no rate year increase. The 24 Company is willing, however, to modify its adjustment to include the full compensation and 25 associated benefits for all classification of employees but to limit the projected compensation 26 expense for salary employees to 2,080 hours. 27 28 29 30 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 4 of 12

1 Q. Did Staff and CAD make an adjustment to total compensation expense for performance pay? 2 A. Yes. Since both parties made an adjustment to reduce the expense by underestimating hours 3 worked and adjusted non-union hourly and salary employees to include only the March 2018 rate, 4 the amount of performance pay has been proportionately reduced as well. This resulted in an 5 additional reduction in compensation expense of $196,826 and $104,159 by Staff and CAD,

6 respectively. As discussed above, the Company strongly believes that the Company’s full 7 compensation expense should be accepted, and therefore the corresponding performance pay 8 should be accepted as well. In addition, the Company advised the parties in its response to CAD-

9 2-E-19 (attached here as Exhibit JDH-1) that performance pay has been extended to union 10 employees as part of negotiated benefits with American Water‘s collective bargaining units across 11 the country, and consequently the Company has an additional expense for performance pay 12 related to union employees that neither Staff nor CAD included in their recommendations. The 13 Company has further updated the adjustment and calculated the additional expense to be 14 $338,447. The Company requests the Commission reject Staffs and CAD’S adjustments. 15 16 17 Q. Did Staff and CAD make adjustments to other compensation-related line items? 18 A. Yes. In addition to adjusting the total compensation as discussed above, Staff and CAD adjusted 19 the corresponding 401k, Defined Contribution Plan (“DCP”), group insurance and payroll taxes 20 associated with the employee’s gross pay. 401K and DCP is based on the employee’s gross pay; 21 however, group insurance is a set cost by employee and not based on hours worked. Therefore, 22 when Staff pro- rated the group insurance using the HTY hours, it did so inaccurately. The result 23 was a reduction of $86,719 to the Company’s going level request. 24

25 Q. Did Staff or CAD make any adjustments to OPEB expense? 26 A. No. However, the Company has received an updated actuarial report on the OPEB costs which 27 have been reduced. A copy of this report, which was issued on 9/24/2018 is attached here as 28 Exhibit JDH-2-CONFIDENTIAL, and a redacted version as Exhibit JDH-2-Redacted. Thus, the 29 Company will update Statement G, Adjustment No. 38, reducing OPEB expense by $237,525 to West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Jamie D. Hawn Pane 5 of 12

1 $469,751 from $707,276. This reduction partially offsets the increase in performance pay for 2 union employees explained above. 3 4 Production Costs

5 Q. Do you agree with Staff and CAD on the adjustment for change in consumption to Power, 6 Chemicals, and Waste Disposal? 7 A. No. Staff proposed smaller reductions to Power, Chemicals, and Waste Disposal on the basis of 8 Mr. Fowler‘s declining consumption projections, which were lower than Mr. Roach’s calculated 9 declining consumption figures. CAD also proposed a reduction for such calculation by reflecting

10 only one year of the Company’s decline based on the 2015 Rate Case Order. The Company stands 11 by its declining consumption adjustment, which the Company’s witness Mr. Roach will confirm in 12 his rebuttal testimony to Mr. Fowler’s and CAD’S positions. Accordingly, the Company’s 13 Adjustment Nos. 25’28, and 30 continue to be accurate. 14

15 Q. Did Staff and CAD make an adjustment for the Company’s increase in power expense? 16 A. Yes. Both Staff and CAD removed the Company’s adjustment for a 5% increase in the Appalachian 17 Power Company (‘‘APCo”) costs due to a pending rate filing. The Company anticipates an increase 18 based on APCo’s pending case and the respective Tariff pages filed that are applicable to the 19 Company’s power bills. Therefore, the Company maintains its adjustment is valid. The Company 20 is willing to update the adjustment to reflect the proper amount if an outcome is reached prior to 21 the Commission’s Order in this case. 22 Staff and CAD also applied this same reduction to the Company’s Power expense for 23 admin and general. As discussed above the Company maintains its filed position. 24

25 Q. Staff made an adjustment to remove all going level waste disposal costs projected by the 26 Company. Do you agree with the adjustment? 27 A. No. In addition to the change in consumption adjustment made by Staff to the waste disposal 28 expense discussed above, they also removed the Company’s normalization of waste disposal

29 adjustment. It is appropriate to use an average of the last three years of actual data to normalize 30 the waste disposal expense levels because costs vary due to weather conditions (which differ from West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 6 of 12

1 year to year). The Company believes this is a more accurate method to determine variable waste 2 disposal expense.

3 4 Service Companv Costs 5 6 Q. Did Staff and CAD make an adjustment to remove the going level adjustment to the Service 7 Company costs? 8 A. Yes. Staff removes a total amount of $1,390,169 and CAD removes a total amount of $1,490,686 9 in Service Company expense. The parties agree on the removal of $102,358 in costs, but Staff and

10 CAD assert that the Company did not adequately support the $1,193,898 going level adjustment. 11 CAD further asserts that the Company has not demonstrated that the Service Company expense 12 is reasonable. 13 14 Q. Did you agree with the Staff and CAD adjustments? 15 A. No. First, the Company provided sufficient information to substantiate its going-level adjustment. 16 The Company’s workpapers for Statement G, Adjustment No. 39, pages 2 and 3 demonstrate the 17 Company’s calculation for the going level Service Company expense. Also, in the Company’s 18 response to CAD-2-1-12 (provided at page 15 of Mr. Smith’s Exhibit LA-2), the HTY and the going 19 level expenses for Service Company were provided by the general ledger account level, which 20 identified where the adjustments were applicable. 21 22 Second, Company witness Baryenbruch concludes that overall Service Company costs are 23 reasonable. In his direct testimony at p.3, he concludes that “[ilf all the managerial and 24 professional services now provided by the Service Company had been outsourced during 2017, 25 [WVAW] and its ratepayers would have incurred $3.9 million in additional expenses.” That is more 26 than three times the level of Service Company cost increases the Company is seeking in this case, 27 and does not take into account potential rate increases for outside consultants during the rate 28 year. 29 30 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 7 of 12

1 Q. Please provide more detail on the Company’s proposed going level adjustment for Service 2 Company compensation expense. 3 A. The employee costs for the historical test year were annualized for the increase to base pay 4 effective March 2017 to include an average base pay increase of 2.92% for non-union and 2.25% 5 for union employees. Also, as provided in CAD-1-J-07, the Company removed from the base year, 6 expenses that were non-rate-making costs or were one-time costs. Then Service Company 7 employee base year expenses were adjusted to annualize a base pay increase effective March 12, 8 2018 of 2.79% for non-union and 2.25% for union employees. A three-year average base pay 9 increase of 2.69% for non-union, and 2.25% for union employees was then applied to derive the 10 going level labor expense through the rate year. Exhibit JDH-3 reflects another view for the 11 summary of the as filed adjustments.

12 Q. Please provide more detail on the Company’s proposed going level adjustment for other Service 13 Company expenses. 14 A. Additional adjustments were made for increased customer service support, pension and OPEB 15 expense, depreciation, and capital lease interest. The increase in customer service organization 16 costs includes an increase in overflow agent calls resulting from enhancements to the Customer 17 Service Center (“CSC”) phone technology. These enhancements allow for a greater volume of calls 18 to be received by our CSC. In addition, our customers are now better able to self-serve through a 19 new Interactive Voice Response (“IVR”) system. This allows our CSC agents to focus on more 20 complex issues that increase call handling time, and overflow agents to help support the calls 21 coming through the CSC to minimize customer hold times. 22 The adjustment for pension and OPEB expense reflects the ASC 715 accounting in a similar fashion 23 as WVAW and will be updated for the lower OPEB amounts per the actuarial report. The additional 24 reduction for OPEB of $24,770 updates the overall going level adjustment to $1,169,128 for the 25 Service Company. In addition, adjustments were made to depreciation and capital lease interest 26 expense associated with the existing and additional assets retained by Service Company for use 27 in providing services to WVAW and its affiliates. Therefore, the Company requests the 28 Commission reject Staffs and CAD’S proposed adjustment for the removal of going level Service 29 Company expense. 30 West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 8 of 12

1

2 Q. Does CAD make additional adjustments to Service Company expense? 3 A. Yes. CAD proposes removing the following miscellaneous expenses in their entirety, claiming 4 they do not benefit WVAW customers: DC SERP expense, 401(k) restoration expense, Employee 5 Stock Purchase Plan (“ESPP”) expense, advertising expense and Company dues and membership 6 fees. As CAD mentions in the Exhibit LA-1, Schedule C-16, the Company did not include costs 7 associated with DC SERP expenses, however the remaining expenses, which the Company did not 8 remove as non-rate making costs is equal to a going level amount of $60,405, as shown below.

Account Account Name Going level 504250 401k Restoration Expense $2,954 5wuooO Employee Stock Purchase Plan Expense 14,302 52524000 Co Dues/Membership Deductible 43,149 9 Total $60,405 10 The Company does not agree with CAD to remove these additional costs.

11 12 The expenses associated with dues and memberships relate to professional development 13 for more efficient and effective service through knowledge of the latest trends and techniques, 14 and joint efforts between business and industry to improve local economy. 15

16 Q. Does Staff and CAD make an adjustment to Service Company expense for LTPP? 17 A. Yes. Staff and CAD proposes to remove 50% of the Service Company LTPP. The proposed 18 adjustment reduces the Company’s going level expense by $149,391, which is included in the 19 overall proposed Service Company amount. For the reasons discussed in Ms. Mount’s rebuttal 20 testimony, the Commission should reject the Staffs and CAD’S recommendation. 21 22 West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Pane 9 of 12

1 Other Operating and Maintenance Costs 2 3 Q. Staff made an adjustment to Admin and General postage expense. Do you agree with this 4 adjustment? 5 A. No. Staff eliminates $733 in postage expense without any explanation or support. The Company 6 maintains that its proposed postage expense, based on the rate increases for UPS and USPS 7 discussed in my direct testimony, is appropriate for setting rates in this proceeding. 8 9 9. Staff made an adjustment to the amortization of tank painting expense. Do you agree with this 10 adjustment? 11 A. No. Staff eliminates $143 in tank painting expense without any explanation or support. The 12 Company maintains that its proposed adjustment, based on the Gross Domestic Product: Implicit

13 Price Deflator (GDPDEF) Index, is appropriate for setting the rates in this proceeding. The 14 Company provided to Staff, as an informal audit request the table for the calculation of the 15 inflation factors used in its calculation, provided here as Exhibit JDH-4. 16 17 Q. Please describe Staffs and CAD’s adjustment to regulatory expense. 18 A. Staff and CAD propose to reduce the Company’s rate case expense by $224,572 and $116,363,

19 respectively. Staffs adjustment reflects rate case expense at the 2015 rate case levels, however 20 CAD’s adjustment reflects a reduction to the costs associated with Service Company, legal fees 21 and the cost of capital witness. 22 23 9. Do you agree with Staff’s and CAD’s proposed adjustments to regulatory expense? 24 A. No. Not only does Staff propose a reduction to 2015 levels, the adjustment also excludes the 25 unamortized portion of the 2015 rate case costs authorized in Case No. 15-0676-W-42T, which 26 reflects three years of the five-year amortization period. In addition, Staff failed to recognize the 27 amount of the 2016 DSlC initial filing costs which was deferred for recovery to this current case. 28 Staff indicated that these costs were excluded since the amortization would expire by the time 29 new rates are set to begin, however this is incorrect. The Company should be allowed to recover 30 the remaining unamortized portion of these costs at the time new rates become effective, which 31 are the amount reflected in the Company’s adjustment. West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Jamie D. Mawn Page 10 of 12

1 2 CAD reduced a portion of the Company’s requested regulatory expense, asserting it is

3 unreasonable. Regarding the Service Company portion of regulatory expense, CAD removes 4 $201,150 of expense reducing it to 2015 levels. CAD’S proposed adjustment is inappropriate 5 because Revenue Analytics and costs associated with a fully litigated proceeding were not 6 included in the Service Company portion of the regulatory expense in the last WVAW case. 7 8 For this case, the services projected include both Regulatory Services and Revenue 9 Analytics hours to prepare the rate case filing, file the case, respond to discovery and participate 10 in hearings. It is appropriate to include Regulatory Services and Revenue Analytics costs in 11 regulatory expense for WVAW. Regulatory Services and Revenue Analytics provide support to all 12 American Water regulated utility companies and is experienced in providing support in all aspects 13 of the preparation, filing and support of rate and regulatory filings. Service Company charges 14 WVAW’s rate case expense account for Regulatory Services and Revenue Analytics only when 15 those teams provide services to prepare, file, and execute a rate case. The other charges billed by 16 Service Company to WVAWC do not include costs from Regulatory Services and Revenue Analytics 17 to support the rate case and these costs are not duplicative. 18 In addition to the Service Company costs, CAD also makes adjustments to the legal fees 19 and cost of capital consultant expense. The legal fees were limited to $200,000 (reducing the 20 expense by $100,586) based on CAD’S assumption that WVAW could hire an additional in-house 21 counsel to support rate case proceedings in place of external counsel. The Legal department 22 balances its costs between in-house, Service Company, and outside counsel and the decision to 23 add an in-house counsel solely to address rate case expense (or even “recurring regulatory 24 proceedings”) must be balanced against the ability to leverage decades of matter-specific 25 experience available at prudent cost from outside counsel and the ability to reduce expenses in 26 non-rate case years. CAD’s blanket assumption does not capture these considerations. CAD’S 27 proposed reduction in the cost of capital consultant expense of $49,000 does not reflect current 28 costs. CAD averaged WVAW’s cost of capital consultant witness expense from the 2010,2012 and 29 2015 rate cases. Its proposed adjustment based on expenses stretching back eight years is 30 unreasonable and inappropriate. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-05764-421 Rebuttal Testimony of Jamie D. Hawn Page 11 of 12

1 2 The Company will update rate case expense for all work performed up to and through the 3 briefing of this case to properly reflect its actual expenses for recovety. 4 5 6 Q. Do you have any additional items to update? 7 A. Yes, the Company has an update to insurance other than group (“IOTG”) expense. In my direct 8 testimony at page 12, lines 22 and 23, I state that the “[gloing level expense is based on current 9 policies in effect through 2018, and includes a retrospective adjustment, based on a three-year

10 average.’’ On Statement A2, page 4 of 4, line 152 (Insurance Expense) in the Rule 42 filing, the 11 Company has found an error in its Statement GI Adjustment 49 - IOTG. The going level expense 12 should not include a retrospective adjustment because the amounts historically captured through 13 such adjustments are now captured in premium amounts. In addition, the three-year average to 14 normalize the costs for administration fees was calculated incorrectly. The calculation captured 15 liability balances not the actual invoices (activity). When the Company calculates this line item 16 using the actual invoices from 2015,2016 and 2017, the proper adjustment to the IOTG reflects a 17 positive cost of $14,762. Therefore, the overall adjustment to IOTG would result in an increase 18 to expense of $117,669, not the decrease to expense in the amount of ($1,440,200). Thus, the 19 Company under estimated its revenue requirement by approximately $1.5 million for water. 20 21 Additionally, the impact for the portion of expense allocated to wastewater (Statement 22 GI Adjustment 19) resulted in a $.009 million under statement of the Company’s going level 23 expense. 24 25 Exhibit JDH-5-CONFIDENTIAL and a redacted version as Exhibit JDH-5-Redacted is 26 attached to supplement the Company’s response to CAD-1-A-02. It contains the actual invoice 27 support for the adjustment to administrative fees with an updated Statement G, Adjustment 49 28 and Statement G, Adjustment 19 for water and wastewater, respectively. The Company 29 acknowledges that this is not ideal timing for uncovering the error but wants to raise it for all 30 parties’ review and consideration. The Company respectfully requests that, to the extent West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Jamie D. Hawn Page 12 of 12

1 allowance of this modified expense does not result in exceeding the Company’s noticed rate 2 increase, it be considered for cost recovery in this case.

3

4 Q. Did Staff and CAD make similar adjustments to the wastewater expenses in their direct 5 testimony? 6 A. Yes. Staff made similar adjustments to water and wastewater for the expenses of compensation, 7 including LTPP, and other compensation-related expenses, as well as power, chemicals, waste 8 disposal, and Service Company. 9 As for CAD, I was not able to determine where wastewater adjustments were included in their 10 direct testimony. 11

12 Q. Do you agree with the adjustments made by Staff to the wastewater expenses? 13 A. No. Based on my discussions above, the Company reflects the same positions for both the water 14 and wastewater expense adjustments. 15 16 17 Q. Does this conclude your rebuttal testimony? 18 A. Yes. Exhibit JDH-1 Page 1 of 2 WEST VIRGINIA-AMERICAN WATER COMPANY CASE NOS. 18-0573-W-42T and 18-0576-S-42T WV PUBLIC SERVICE COMMISSION - CONSUMER ADVOCATE DIVISION SECOND SET OF DISCOVERY - RECEIVED September 4,2018

Prepared by: Jamie Hawn, Senior Manager, Regulatory Services; Carolyn Mount, Director HR Business Partner Witness: Jamie Hawn, Senior Manager, Regulatory Services; Carolyn Mount, Director HR Business Partner Date Prepared: September 25,2018

CAD-2-E-19

EXPENSES Incentive Compensation. a. Identify the total amount of direct charged incentive compensation that the Company has included in its test year revenue requirement, by amount and account. b. Identify the total amount of incentive compensation allocated from the Service Company that the Company has included in its test year revenue requirement, by amount and account. c. Identify the amount of direct charged incentive compensation that relates to meeting the Company’s financial performance goals, for the test year ending December 31, 2017, in total and by account.

d. if different than part ‘IC“, identify the amount of incentive compensation allocated from the Service Company that relates to meeting the Company’s financial performance goals, for the test year ending December 31,2017, in total and by account.

RESPONSE:

a. The total amount of direct charged and allocated performance based compensation included in the Company’s Rule 42 (test year actuals and adjustments) and Addendum filing are outlined in the table below.

Since the Company filed its 42T application on April 30, 2018, it has negotiated benefits with collective bargaining units across the country. The negotiated benefits to date include the extension of performance based compensation to collective bargaining employees (the “union members”) at the rate of 3% of their pay, beginning in 2019. As a result, the Company estimates an additional $334K of going level Annual Performance Plan Expense.

Page 1 of 2 Exhibit JDH-1 Page 2 of 2 WEST VIRGINIA-AMERICAN WATER COMPANY CASE NOS. 18-0573-W-42T and 18-0576-S-42T WV PUBLIC SERVICE COMMISSION - CONSUMER ADVOCATE DIVISION SECOND SET OF DISCOVERY - RECEIVED September 4,2018

Prepared by: Jamie Hawn, Senior Manager, Regulatory Services; Carolyn Mount, Director HR Business Partner Witness: Jamie Hawn, Senior Manager, Regulatory Services; Carolyn Mount, Director HR Business Partner Date Prepared: September 25,2018

Water Tota I \W Billed through Performance Based Account Account Description Amount Senrice Co Compensation 50171005 Annual Performance Plan Expense $575,072 $750,079 $1,325,151 50171600 &50171800 Long Term Performance Plan Expense %W 298,860 389,84& $666,060 $1,048,940 St7l.5,

Wastewater Tota I WV Billed through Performance Based Account Account Description Amount Service Co Compensation 50171008 Performance Plan Expense 7,300 $225 $7,525 50171600 & 50171&00 Long Term Performance Plan Expense 11 11 $7,300 $235 $7,536

b. Refer to part a for this response.

c. The Company’s performance compensation plans align the interests of our customers, employees and shareholders. All of the Company’s goals relate to the Company‘s financial performance. The operational components measure performance that can most directly influence customer satisfaction, health and safety, environmental performance, and operational efficiency, which affect the Company’s financial performance (e.g., long-term cost savings or avoided costs). Importantly, to achieve performance pay financial goals, such as targeted earnings per share (“EPS’’) performance, demands attention to operating efficiency. That is, unless the utility controls or reduces its operating costs, it cannot achieve a targeted EPS. Well- grounded financial measures keep the organization focused on improved performance at all levels of the organization, particularly in increasing efficiency, decreasing waste, and boosting overall productivity.

d. Refer to part c for this response.

Page 2 of 2 Exhibit JDH-2-Redacted Page 1 of 2

West Virginia American Water Company Workpaper-Statement G Adjustment 38 Pension and Other Post-Retirement Benefit Plan

ASC 715 Accrual Costs 2018 - Updated AW Pension & Postretirement WVAWC Pension & Postretirement Welfare Cost Welfare Cost American Water American Water Qualified Pension Total Retiree Qualified Pension Total Retiree Pension and Postretirement Welfare Cost Plan Welfare Plans Plan Welfare Plans Service cost (OR) $1,783,668 $292,365 interest cost (INT) 4,000,812 1,082,220 Expected return on assets (INT) (5,189,391) (1,884,157) Amortization Transition obligation (asset) I I Prior service cost (credit) (OR) 32,509 (2,445,632) Net loss (gain) (OR) 1,619,862 84,973 Pension Cost -- $2,247,460 ($2,870,2301 Exhibit JDH-2-Redacted Page 2 of 2 9/24/2018

American Water 2018 Posteretinnent Beneift Plan Accounting Cost ASC 715-60 (formerly FAS 106)without reflecting purchase accounting

(i) (ii) (iii) (iv) (VI Based on Janualy Projected to FY2018 Prorated 1,2018 Actuarial 8/31/2018 Prior to Effect of Plan 8/31/2018 After Plan [8/12 of (i) + 4/12 of Valuation Plan changes Changes Changes (iv)] Reconciliation of Funded Status

1. Unfunded Benefit Obligation (a) Accumulated Postretirement Benefit Obligation (b) Fair Value of Assets (c) Funded Status

2. Accumulated Other Comprehensive Income (a) Unrecognized transition obiigation (asset) (b) Unrecognized prior service cost (credit) (c) Unrecognized net actuarial loss (gain) (d) Total Accumulated Other Comprehensive Income

3. Prepaid (accrued) pension cost $ 0 I II I

FY2018 4. Net Periodic Postretirement Benefit Cost Annual Annual Annual Prorated (a) Servicecost (b) Interest cost (c) Expected return on assets (d) Amor ization (e) Transition obligation (asset) (9 Pnor service cost (credit) (9) Net loss (gain) (h) Total Postretirement Benefit Cost (ANNUAL) 1I. 1 Assumptions Discount rate Expected Return on Assets Healthcare trend

Mortality

Expected contribu ions in he year after measurement Expected benefit payments Average future working lifetime to full eligibility age

Under ASC 715-60, the net amount recognized is equal to the funded status; this accrued pension cost is shown for cost determination purposes only. West Virginia American Water Company - Water Exhibit JDHJ West Virginia American Water Company - Wastewater Page 1of 1 For the Twelve Months Ended December 31,2017

Adjustments Normalize base year for merit (average Depreciation 4/1/19 Merit 12 months Removed one- increase) Labor and 4/1/18 Merit Increase. and LOP increase. Labor Total 12 months ended Function ended 12/31/17 time items Labor related Labor and Related CSC Costs PensionlOPEB lnterco Rent Interest and Related Adjustments 2/29/20

Business Development $270,966 ($1,932) $1,650 $6,491 so ($1,525) ($3931 $0 $6,441 $10,732 $281,698 Central Lab 58,220 0 222 872 0 ($279) 0 0 865 1,680 59,900 Corp Admin 1,240,121 (1) 0 0 0 ($4.6891 0 438,713 0 434,023 1,674,144 Engineering 77,322 1 461 1,813 0 ($400) 0 0 1,799 3,674 80.997 External Affairs & Public Poiicy 301,972 (5,325) 1,425 5,605 0 ($8431 0 0 5,562 6,424 308.396 Finance 1,535,402 (1,890) 8,975 35,304 0 ($5.202) (72) 0 35,032 72,147 1,607,550 Regulated Ops 468,665 (15,620) 2,358 9,276 0 (53,745) 0 0 9,205 1.474 470,139 Human Resources 931,887 (5,709) 4,840 19,039 0 ($2,1791 (881 0 18,892 34,795 966,682 Investor Relations 39,485 0 194 763 0 WI 0 0 757 1,630 41,115 Legal 515,764 4,086 2,861 11,255 0 ($2.9161 (381 0 11,169 26,417 542,181 Supply Chain 124,226 (6261 637 2,506 0 ($786) 0 0 2,487 4,217 128,443 Corporate Security 164,959 59 611 2,405 0 ($228) 0 0 2,386 5,234 170,192 Customer Service Organization (GO) 3,105,152 (3651 13,899 56.177 503,608 ($29,345) 0 0 57.443 601,417 3,706,569 Facilities 238,127 0 35 139 0 $0 0 0 138 312 238,439 Safety & EnvironmentalCompliance 108,180 (9.606) 534 2,100 0 ($700) 0 0 2.084 15.589).. 102.592 Technologyand Innovation (T&l) 2,313,731 (65,430) 7,547 29,686 0 ($5,880) (701 0 29,457 (4,690) 2,3W,M1 Total $11,494,179 ($102.358) $46,250 $183,432 $503,608 l$58,801) ($661) $438,713 $183,716 $1,193,898 $12,688,077 Exhibit JDH-4 Page 1 of 3 WEST VIRGINIA-AMERICAN WATER COMPANY CASE NOS. 18-0573-W-42T and 18-OS76-S-42T WV PUBLIC SERVICE COMMISSION -STAFF AUDIT EIGHTH SET OF DISCOVERY Date Prepared:

Staff Audit 08-03

Water Operations

Please submit the Company’s Workpaper detailing how the present value inflation factor % was developed for tank painting expense, including the index used.

Response:

The Company used the Excel download of the Gross Domestic Product: Implicit Price Deflator (GDPDEF) Index located at HTTPS://fred.stlouisfed.oralserieslGDPDEFfor the calculation of the present-value inflation factor percentage.

Please refer to Staff Audit 08-03-Attachment for the calculation of the present-value inflation factor percentage. Exhibit JDH-4 Page 2 of 3

West Virginia-American Water Company Staff Audit 08-03 Case No. 18-0573-W-42T

GDPDEF - Gross Domestic Product: Implicit Price Deflator, Index 2009=100, Quarterly, Seasonally Adjusted Frequency: Quarterly INDEX CALCULATION Year end GDPDEF used in 10/1/17 GDPDEF Inflation Factor % observation-date GDPDEF Calculation Adjustment Year YEARLY OCT GDPDEF Used 2004-01-01 88.108 2004-04-01 88.875 2004-07-01 89.422 2004-10-01 90.049 90.05 2004 114.27 f 90.05= 1.269 2005-01-01 90.883 2005-04-01 91.543 2005-07-01 92.399 2005-10-01 93.100 93.10 2005 114.27 f 93.10= 1.227 2006-01-01 93.832 2006-04-01 94.587 2006-07-01 95.247 2006-10-01 95.580 95.58 2006 114.27 / 95.58= 1.196 2007-01-01 96.654 2007-04-01 97.194 2007-07-01 97.531 2007-10-01 97.956 97.96 2007 114.27 / 97.96= 1.167 2008-01-01 98.516 2008-04-01 98.995 2008-07-01 99.673 2008-10-01 99.815 99.82 2008 114.27 / 99.82= 1.145 2009-01-01 100.062 2009-04-01 99.895 2009-07-01 99.873 2009-10-01 100.169 100.17 2009 114.27 /100.17= 1.141 2010-01-01 100.522 2010-04-01 100.968 2010-07-01 101.429 2010-10-01 101.949 101.95 2010 114.27 f 101.95= 1.121 2011-01-01 102.399 2011-04-01 103.145 2011-07-01 103.768 2011-10-01 103.917 103.92 2011 114.27 f 103.92= 1.100 2012-01-01 104.466 2012-04-01 104.943 2012-07-01 105.508 2012-10-01 105.935 105.94 2012 114.27 f 105.94= 1.079 2013-01-01 106.349 2013-04-01 106.570 2013-07-01 107.084 2013-10-01 107.636 107.64 2013 114.27 /107.64= 1.062 2014-01-01 108.083 2014-04-01 108.692 2014-07-01 109.187 2014-10-01 109.345 109.35 2014 114.27 f 109.35= 1.045 2015-01-01 109.326 2015-04-01 109.916 2015-07-01 110.286 2015-10-01 110.513 110.51 2015 114.27 /110.51= 1.034 2016-01-01 110.582 Exhibit JDN-4 Page 3 of 3

West Virginia-American Water Company Staff Audit 08-03 Case No. 18-0573-W-42T

GDPDEF - Gross Domestic Product: Implicit Price Deflator, Index 2009=100, Quarterly, Seasonally Adjusted Frequency: Quarterly INDEX CALCULATION Year end GDPDEF used in 10/1/17 GDPDEF Inflation Factor % observation-date GDPDEF Calculation Adjustment Year YEARLY OCT GDPDEF Used 2016-04-01 111.249 2016-07-01 111.628 2016-10-01 112.190 112.19 2016 114.27 /112.19= 1.019 2017-01-01 112.746 2017-04-01 113.029 2017-07-01 113.614 2017-10-01 114.271 114.27 2017 114.27 / 114.27= 1.000 Exhibit JOH-5-RedaRed Page 1 of 11 West Virginia-American Water Company - Water Insurance Other than Group Statement G For the Twelve Months Ended December 31,2017 Adjustment 49 Page 1of 1

656.8 -659.8 Going Level Adiustment for Insurance Other than Group

Policy Date Amount of From To Policy Mths Current Policy Annual Policy Description 05/01/17 04/30/18 12 $1,267 $1,267 Crime 05/01/17 04/30/18 12 15,402 15,402 Directors and Officers 05/01/17 04/30/18 12 9,489 9,489 Employment Practices 05/01/17 04/30/18 12 3,494 3,494 Fiduciary 05/01/17 04/30/18 12 1,947 1,947 Employed Lawyers 01/01/18 03/31/18 3 17,387 69,549 Auto Liability 01/01/18 03/31/18 3 463,637 1,854,550 General Liability 01/01/18 03/31/18 3 95,152 380,610 Workmens Compensation (135,669) WC Capitalization 01/01/18 03/31/18 3 4,395 17,581 Consult Fee (AL,GL,WC) 01/01/18 01/01/19 12 200,536 200,536 Excess Liab#l 01/01/18 01/01/19 12 21,671 21,671 Excess Liab#2 01/01/18 01/01/19 12 249,742 249,742 Property 01/01/17 1213 1/17 12 12,799 12,799 Cyber Crime 01/01/17 12/31/17 12 14,762 14,762 Administrative Fees-3 yr avg2 1,111,681 2,717,730

Going Level Expense 2,717,730

Test Year Expense (19,399,939) Less: Removal of 2017 costs relating to Chemical Spill' 22,000,000 Total Adjusted Test Year Expense 2,600,061

Going Level Adjustment $117,669

'See Adjustment 23 'Three-year Average of Administrative Fees 2015 $42,254 2016 17,854 2017 (15,546) Average 14,854 Water % 99.38% Total $14.762 Exhibit JDH-5-Redacted West Virginia American Water Company - Wastewater Page 2 of 12 Insurance Other than Group Statement G For the Twelve Months Ended December 31,2017 Adjustment 19 Page 1of 656.8 -659.8 Goinn Level Adiustment for Insurance Other than Group 1

Policy Date

Policy Amount of From To Months Current Policy Annual Policy Description 05/01/17 04/30/18 12 $8 $8 Crime 05/01/17 04/30/18 12 96 96 Directors and Officers 05/01/17 04/30/18 12 59 59 Employment Practices 05/01/17 04/30/18 12 22 22 Fiduciary 05/01/17 04/30/18 12 12 12 Employed Lawyers 01/01/18 0313 1/18 3 108 434 Auto Liability 01/01/18 03/31/18 3 2,892 11,570 General Liability 01/01/18 0313 1/18 3 632 2,529 Workmen's Compensation (191) WC Capitalization 01/01/18 03/31/18 3 27 110 Consult Fee (AL,GL,WC) 01/01/18 01/01/19 12 1,251 1,251 Excess Liab#l 01/01/18 01/01/19 12 135 135 Excess Liab#2 01/01/18 01/01/19 12 1,558 1,558 Property 01/01/17 12/31/17 12 80 80 Cyber Crime 01/01/17 12/31/17 12 92 92 Administrative Fees-3 yr avgl 6,974 17,765

Per Book Expense (119,214) Going Level Adjustment $136,979

'Three-year Average of Administrative Fees 2015 $42,254 2016 17,854 2017 (15,546) Average 14,854 Water % 0.62% Total $92 Exhibit JDH-5-Redacted Page 3 of 12

05 NORTH EaETRO NATIONAL ACCOUNTS DEPARWNT ...... INVOICE NO DATE PREPARED SAX ** DATE DUE AM0VW.i' RUE ** ** 18 3 713 6 09/01/15 2996L1020 ** 10/31/15 $7 ** *c ** PRODUCER: NA026 MARSff USA INC ** VALUATION ADJUSTMENTms XMSURED: AMERICAN WATER WORKS CO., INC. ** VALUATION DATE; 06/30/15 **

PLGASE RETURN PAGE 1 WITH PAYMENT, XNTIRE PAYNEXT DUE ON OR BEFORE DUE DATE. TRAVELERS NATIONAL ACCOUNTS DEPARmNT INVOICE NO: 1837136 'DATE PRISPARET): OS/Ol/ZS AMOUNT DUE: $743,833.00 SAIr 2396L1020 AXERICAN WATER WORKS CO., INC. DATE DUE: 16/31/15 PRODUCER: NA026 MARSH USA INC BINDER N"XBER Mom DUE VALUATION ADJUSTMENT 804190 01/01/75 01/01/76 804191 01/01/76 0 1/0117 7 80.1192 01/01/77 01/01/78 802025 10/01/?8 10/01/79 804293 01/01/78 10/01/78 804195 10/01/7 9 10/01/80 802026 10/01/80 10/Ol/83. 802027 10/01/81 10/01/82 802028 10/01/a2 10/01/83 804194 10/01/83 10/01/34 802029 l0/01/84 10/01/85 802030 10/01/85 l0/01/86 302031 10/01/86 10/01/87 802032 10/01/87 10/02/88 802033 l0/01/83 lo/ol/as 802034 10/03/89 10/031/90 302035 10/01/90 16/ Ol/ 9 1 802036 10/01/91 lO/Ot/92 802039 10/61/92 ro/o3./9s 802040 10/01/93 zo/or/94

PAGE 1 OF 2 Exhibit JDH-5-Redacted Page 4 of 12 T LER 05 NORTH METRO NAT'TONAL ACCOUNTS DE~A~~~ ****************it******************** IWO'TClg NO DATE PREPARED SAX ** DATE DWE AMOUNT DUE ** 1837136 09/01/15 2996L1020 ** 10/31/15 s7-43.rgg-3.00 ** ** cOok,333 ;; PRODUCER: ~~026mRsn USA INC ** VALUATION ADJETSTBIEPST

PLEASE RETURN PAGE 1 WITH PAYMENT, ENTIRE PAYHENT DUE OH OR REFORE DUE DATE. BMDER MJMBER momDUE VALUATION ADJUSTMEW CONT . 801336 09/01/94 09/01/95 802041 10/01/94 10/0 I/ 9 5 801337 09/01/95 09/01/96 803651 lO/Ul/Ss 10/01/96 173951 10/01/96 10/01/97 186048 10/01/97 lO/Ol/98 188362 10/01/97 09/05/98 199188 10/01/98 10/01/99 204419 10/01/99 10/01/00 206170 10/01/00 10/01/01 207826 10/01/01 10/01/02 209244 lO/Ol/OZ 01/01/04 213032 01/01/04 01/01/05 216555 01/01/05 01/01/06 218906 01/01/06 01/01/07 221111 01/03/07 0%/01/08 223338 01/01/08 01/01/09 225388 01/01/09 01/01/10 227611 01/01/10 01/01/11 229548 02/01/12 01/01/12 231758 01/01/12 01/01/13 233978 01/01/13 01/01/14 236123 01/01/14 01/01/15 238293 02/0i/x5 01/01/16 BILLING REPRESENTATIVE

PAGE 2 OF 2 Exh.bit JDH-5- Redacted Display Transaction: OverVrrew Page 5 of 12

t-q 3Taxes * D~plaayCunency $4 Accompanyglg Doamnts

CrossSC no. 5000042567103316 Type I(R Vendor Invotce Postng Date 0 12' 14 / ZC 16 Exhibit JDH-5-Redacted Page 6 of 12 TRAVElER 05 NORTH METRO NATIONAL ACCOUNTS DEPARTMENT...... INVOICE NO DATE PREPARED SA1 ** DATE DUE AMOUNT DUE ** 1904504 08/25/16 2996211020 ** 10/24/16 $231,523.00 ** ** ** PRODUCER: NA026 MARSH USA INC ** VALUATION ADJUSTMENT ** INSURED: AMERICAN WATER WORKS CO., INC. ** VALUATION DATE: 06/30/16 ** ***************************e*********

PLEASE RETURN PAGE 1 WITH PAYMENT, ENTIRE PAYMENT DUE ON OR BEFORE DUE DATE. TRAVELERS NATIONAL ACCOUNTS DEPARTMENT INVOICE NO: 1904504 DATE PREPARED: 08/25/16 AMOUNT DUE: $231,523.00 SAI: 2996L1020 AMERICAN WATER WORKS CO., INC. DATE DUE: 10/24/16 PRODUCER: NA026 MARSH USA INC BINDER KLTMBER AMOUNT DUE VALUATION ADJUSTMENT 804190 01/01/75 01/01/16 804191 01/01/16 01/01/77 804192 01/01/71 01/01/18 802025 10/01/18 10/01/79 004193 01/01/18 10/01/78 io4Fis io joij79 10/01/80 802026 10/01/80 10/01/81 802021 10/01/81 10/01/82 802028 10/01/82 10/01/83 804194 10/01/83 10/01/84 802029 10/01/84 10/01/85 802030 10/01/85 iojoij86 802031 10/01/86 10/01/81 802032 10/01/87 10/01/88 802033 10/01/88 10/01/89 802034 10/01/89 ioioiiso 802035 10/01/90 iojoiigi 802036 10/01/91 10/01/92 802039 10/01/92 10/01/93 802040 10/01/93 10/01/94

BILLING REPRESENTATIVE Exhibit JDH-5-Redacted Page 7 of 12 TRAVELERS J 05 NORTH METRO NATIONAL ACCOUNTS DEPARTMENT...... INVOICE NO DATE PREPARED SA1 ** DATE DUE AMOUNT DUE ** ** ** 1904504 08/25/16 2996L1020 ** 10/24/16 $237,523.00 ** ** VALUATION ADJUSTMENT ** PRODUCER: NA026 MARSH USA INC ** ** INSURED: AMERICAN WATER WORKS CO., INC...... VALUATION DATE: 06/30/16

PLEASE RETURN PAGE 1 WITH PAYMENT, ENTIRE PAYMENT DUE ON OR BEFORE DUE DATE. BINDER NUMBER AMOUNT DUE VALUATION ADJUSTMENT corn. 801336 09/01/94 - 09/01/95 802041 10/01/94 - 10/01/95 801337 09/01/95 - 09/01/96 803651 10/01/95 - 10/01/96 173951 10/01/96 - 10/01/97 186048 10/01/97 - 10/01/98 188362 10/01/97 - 09/05/98 199188 10/01/98 - 10/01/99 204419 10/01/99 - 10/01/00 206170 10/01/00 - 10/01/01 207626 ioioiioi - 10/01/02 209244 10/01/02 - 01/01/04 213032 01/01/04 - 01/01/05 216555 01/01/05 - 01/01/06 218906 01/01/06 - 01/01/07 221111 01/01/07 - 01/01/08 223338 01/01/08 - 01/01/09 225388 01/01/09 - 01/01/10 227611 01/01/10 - 01/01/11 229548 01/01/11 - 01/01/12 231758 01/01/12 - 01/01/13 233978 01/01/13 - 01/01/14 236123 01/01/14 - 01/01/15 238293 01/01/15 - 01/01/16 240446 01/01/16 - 01/01/17 BILLING REPRESENTATIVE

PAGE 2 OF 2 lof 1

Exhibit 1DH-5-Redacted Page 8 Of 12

JOURNAL ENTRY REQUEST Shared Sewices center131 Woodcrest Rd, cherty HUI,NJ 08003 (666)777-8426

Batch # (SSC USE) Journal # (SSC USE Company We Oocument De Prepared By (Operating Unl) Prepared By (SSC) Requested By (Operating UnR) Approved By (SSC)

Quarter impacted This Is not a Reversing Entry11 Doc Header Text 25 Character Maximum JOURNAL ENTRY DESCRIPTION:

R~8aAnpKR500005(BM)~wasramdedinomac(lybSaviea~yRcpsid~~~~harrbc~to55711000tothcsrbsid~ErdryncQdedbmwee)rrrgsstoshs

Please add rows as necessary to the table below in order to complete your requested journal entry ICornpany) GL I Trading I Posting I I I I Credit I

I 1 1 I I I I i I I Totals Exhibit JDH-5-Redacted Page 9 of 12 TRAVELER

September 10,20 17

RE: Annual Valuation Adjustment American Water Works Co., Inc. Account Number: 2996L1020 Due Date: 10/27/17 Policy Period(s): 1/1/1975 - 1/1/2017 Amount Due: ($216,3 19.00) Valuation Date As Of: 06/30/17

Enclosed please find our billing package for the annual adjustment.

This adjustment indicates a return of ($2 16,319.00) due to American Water Works Co, Inc. Please note that all plan periods will remain open through the next valuation.

If you receive this package and you are not the proper recipient, please contact me, otherwise 1 will call you within 7-10 days to verify receipt of this package.

Please review this valuation adjustment, complete the form below indicating your agreement with our calculations and return this letter to my attention (via fax or regular mail). After the completed forni or email request has been received, a check will be issued to you by the due date, assuming there are no outstanding balances on your account.

Should you have any questions please contact me.

Sincerely,

Approved by: DBroker [I7 Insured Date: Signature: Print Name & Title: Make Check Payable to: Mail Check to (Company Name): , Attention to: Street Address:

~ city: State: Zip: Exhibit JDH-5-Redacted Page 10 of 12 Page 1 of 2 Account Number: 29~6~~~2~ Invoice ~~m~er: 1974048 Date Due: Amount Due:

Vaiuation A~~ustmen~as of June 30, 2017 Producer: MARSH USA INC Client: A~E~~CANWATER WORKS CO.,INC.

Remit Payment To: Your Receivable Account ~anager~ - -I

As of June 30.2017

Please return invoicewith ~ay~entby the due date. Exhibit JDH-5-Redacted Page 11 of 12 Page 2 of 2 Your Receivable A~mmtManager: Account Number: 29%iLIQ20 Invoice Number: 1974048 Date Due: 1012712017 Amount Due:

Valuation Adj~~~~~tas of June 30, 201 7 Producer: MARSH USA INC Client: A~E~I~ANWATER WQRKS CO.,ING.

As of duns 30,2017

01/01/1976 * OI/O’l/1977 1 Of 1

Exhibit JDH-1-Redacted Page 12 of 12

JOURNAL ENTRY REQUEST Shared Services Center131 Woedcrest Rd, Cherry Hill. NJ 08003 (866)777-8426

Batch # (SSC USE) Journal # fSSC USE Company Code Prepared By (Operating Unlt) Prepared By (SSC) Requested

Quarter Impacted Thls is not a Reversing Entry11 Doe Header Text 25 Character Maximum ---..._.--.In1 IRNAl FNTRY..... , ----nFRrRIPTlON...... -...

TmrnkrsRttrosp&bva. drdc )Sm&mdto Sevke company. Dqnuitrharld bcappHed b sS111OOOt0(he m.EnBy ne&edto mavectmges to subs

Please add rows as necessary to the table below in order to complete your requestedjournal entry. F WEST VIR~INIA CHARLESTON

0. 18-Q573-~-42T Case No. 18-0576-

ATER OMP PA^^

Rule 4ZT Tariff Filing to Increase ates and Charges

UTTAL TESTIMO ~RE~OR~P. ROACH West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 1of 12

1 Q. Please identify yourself.

2 A. My name is Gregory P. Roach. I am employed by American Water Works Service Company as a

3 Senior Manager of Revenue Analytics, and I provided direct testimony on behalf of West

4 Virginia American Water Company (“WVAWC” or “the Company”) in this docket filed on April 30, 5 2018. In this rebuttal testimony, I will address issues related to reductions in residential and 6 commercial customer usage raised in the direct testimony filed by Staff‘s witness Jonathan M., 7 Fowler.

8

9 Q. What issues in Mr. Fowler‘s testimony will you be addressing in your rebuttal testimony? 10 A. I will be addressing the impact of the following issues related to Mr. Fowler’s modeling results:

11 1. Mr. Fowler used data from the Company’s “Annual Reports” which are on file with the

12 Commission. That data was subsequently revised as part of on-going billing and audit 13 practices by the Company.

14 2. Mr. Fowler used a three-year moving average for his dependent variable term, average 15 Residential or Commercial usage per annum. The result of using the three-year moving

16 average term allowed Mr. Fowler very little basis to estimate the annual rate of water usage 17 reductions for either the Residential or Commercial customer groups.

18 3. Mr. Fowler’s statistical regression models suffer from autocorrelation of the error terms 19 which have a significant adverse impact on his results.

20

21 From my analysis of these three issues, I have concluded that Mr. Fowler’s recommended 22 declining usage adjustments of only 450 gallons for the residential class and not to exceed 3,750

23 for the commercial class are understated. For the reasons identified here and in my direct

24 testimony, I continue to recommend that the Company’s rates take into account a continuing 25 annual decline of 630 gallons per residential customer per year (an annual decrease of 1.65% per

26 year) and a continuing annual decline of 4,305 gallons per commercial customer per year (an 27 annual decrease of 2.01% per year). 28 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 2 of 12

1 USAGE AND REVENUEDATA REVISIONS 2 3 Q. Please describe the nature of the usage data in the Company’s Annual Reports filed 4 Commission that Mr. Fowler relied on for his regression modeling. 5 A. The data submitted to the Commission is based on the Company’s Hyperion data base which is

6 populated each successive month as part of the month-end close process. That data is static,

7 meaning that it is never adjusted, so that the Company has a complete set of data allowing for 8 auditing, verification and reference of externally reported data going forward, including the 9 Commission report. As such, this data never changes by design; it is static in nature. 10

11 Please describe the nature of the usage data employed b the Company as t 12 statistical ~odeiin~of esidential and Commercial usage. 13 A. The data employed by the Company comes from standardized reporting that is part of our SAP 14 Enterprise Resource System. 15 16 at types of mod~~icationsto usage data may occur in the S ata that are not necessaril 17 in the data su~~~ttedto the ~om~issionin its 18 A. The data bases employed by SAP are dynamic in nature. That is, they include modifications and 19 adjustments to usage and revenue data for customer leaks and certain other miscellaneous items. 20 Comparing the dynamic attributes of the SAP data base with the static nature of the Hyperion 21 data base, the SAP data base will typically reflect differences in both usage and revenue. 22 23 Do you rely on the SA data for your analysis of Residential and Commerc~alusage? 24 A. Yes. The Company’s usage data must be reconciled with Test Year revenue for purposes of 25 developing a revenue requirement and designing proposed rates. In order to be internally 26 consistent for purposes of the rate case, we employ the SAP data for purposes of modeling and 27 forecasting Residential and Commercial average use. 28 29 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 3 of 12

1 Q. Were there differences between the Hyperion and SAP data over the historical period that both 2 Mr. Fowler and you studied in your respective analysis? 3 A. Yes there were significant customer adjustments to usage, and hence revenues, for the years 4 2014, 2015 and 2016. Those differences in the two data streams would have significant impact

5 on the results of any statistical analysis employing data from those years. 6 7 Q. as the SAP usage data you employed in your analysis shared with Commission Staff? 8 A. Yes, that data was supplied to all parties as part of the Company’s filing of work papers in this

9 case.

10 11 Q. Would the differences in data employed by Mr. Fowler as compared to data you used have a 12 significant impact on the results of Mr. Fowler’s analysis? 13 A. Yes, they would and they do. As I will show in detail later in this rebuttal testimony, the 14 differences in the two data sets for 2014-2016 had a significant impact on the results of Mr. 15 Fowler’s statistical analysis employing data from those years - it significantly understates the level 16 of Residential and Commercial average usage reductions. 17

18 EGRESSION ANALYSISEMPLOYING THREE-YEAR MOVING AVERAGE

19 20 Q. re there problems with employing a three-year moving average as the dependent variable in 21 regression analysis?

22 A. Yes, there are. The two main reasons an analyst would not employ the use of a three-year moving

23 average as the dependent variable in regression modeling are: 1) using a moving average 24 dependent variable term violates the principal of using discrete observations for each time period

25 relating one observation to another, that is, usage from each time period corresponds with 26 weather in that time period and that all the usage being modeled occurred in the same time

27 period and 2) modeling and forecasting a three-year moving average does NOT provide ANY 28 visibility to the annual rate of reductions in usage.

29 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 4 of 12

1 Q. Why should the analysis of time series data via regression analysis be organized by discrete time 2 periods?

3 A. Time series regression analysis presumes that the analyst is associating dependent data

4 observations with observations from the same period or in some unique cases from lagged or 5 leading time-periods. 6

7 Q. What relationship has Mr. Fowler actually modeled and hence forecasted? 8 A. I appreciate Mr. Fowler’s thoughtful approach. However, Mr. Fowler has chosen to average three 9 successive time series observations and associate that observation with time from a single time- 10 period. In so doing, Mr. Fowler is no longer analyzing a relationship between usage and the time-

11 period in which that usage occurred. Rather, Mr. Fowler has modeled a relationship where a 12 three-year average usage term is dependent on time from the last time-period on the three-year 13 average. As such, Mr. Fowler’s model does not provide any insight into the relationship of average 14 usage over each successive time period. Rather, he has modeled a relationship between a three-

15 year average of usage and the last time period of that average. From that perspective, Mr. Fowler

16 cannot make any recommendations related to annual reductions in either Residential and 17 Commercial average use reductions as he has not modeled annual usage over time.

18 19 lease explain what yo mean when you state that r. Fowler’s approach oes not provide any

20 visibility to annual reductions in Residential an Commercial usage.

21 A. Let me use an example to illustrate what is the mathematical weakness of using a three-year

22 moving average as a dependent variable in regression analysis. In each of the time series 23 illustrated in Table GPR-lR, three-year average usage is 400 gals/cust/yr. Unfortunately, the 24 average use data point based on a three-year average, fails to give visibility to what are

25 significantly different trends in usage for each of the time series as illustrated by the annual trends

26 which are +200, -200, -300 and -400 gals/cust/year respectively. Hence, employing a successive 27 three year moving average term would completely ignore and mask the underlying annual trend

28 of the data. As a result, by associating a three-year average dependent variable term with time 29 in his regression modeling, Mr. Fowler has completely lost both the ability of the data to describe West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 5 of 12

1 the annual decline in average usage and subsequently cannot provide any reliable projection as 2 to what the annual decline in average usage may be.

3

4 STATISTICAL kSUES R. FOWLER’SREGRESSION MODELING

5 6 . Have you reviewe Mr. Fowler‘s testimony related to his estimation of a tren of average water 7 use reductions for bot the Residential and Commercial customer classes base

8 regression analysis of Residential and Commercial usage?

9 A. Yes. Mr. Fowler generally describes his analysis as beginning with annual usage and customer

10 data from WVAW “Annual Reports” (presented in his Table 1- Annual Average Month~yUsage

11 2008-2017), and his modification of this data set to create a trend of the three-year moving 12 average series (presented in his Figure No. 1- Residential Usage 2006-2017, and Figure No. 2 13 Commercial Usage 2007-2017). From that point, Mr. Fowler performs regression analyses of the

14 modified data sets to estimate a trend for “average usage”. Based on these analyses, Mr. Fowler 15 concludes that reductions in average usage are evident in West Virginia for both residential and

16 commercial users, but at a lower rate than estimated by WVAW. 17 18 . Are there any theoretical complications with r. Fowler’s metho 19 A. Yes. As discussed and illustrated above, Mr. Fowler begins his analysis with a data transformation 20 that violates the precepts of the linear regression technique whereby data from similar time series West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 6 of 12

1 periods are analyzed for their relationship to one another as the time series progresses over time. 2 Mr. Fowler has chosen to statistically relate data that are not from similar time periods. 3 4 Are there any other mathematical issues with Mr. Fowler‘s decision to transform the data 5 points into a moving average before preforming his regression analysis? 6 A. Yes. Mr. Fowler’s averaging technique does not give equal weight to each year of usage. Mr.

7 Fowler underweights the most recent years of activity and over weights older data. With a three- 8 year moving average, 2017 is counted once in the data set, 2016 is counted twice (in the average 9 for 2016 and 2017), and 2015 and past years are all counted three times in his moving average 10 data set. Hence, Mr. Fowler’s three-year averaging technique is not balanced nor consistent 11 across the data series. 12 13 e for purpose$ of il~ustrationthat owler‘s moving avera e ~echn~q~eis valid, 14 to interpret his regression results from his te$timony? 15 A. No. The only regression model validation Mr. Fowler provides in his testimony is the R2 value.

16 While this metric indicates the strength of a valid regression model, the metric is meaningless if 17 the regression model is not valid. Other key metrics are needed in order to determine if the model

18 is valid. Some standard key metrics are autocorrelation statistics, P-values, and F-statistics. 19 20 ere you able to estimate these key regression validation metrics for r. Fowler’s models?

21 A. Yes. I was able to recreate Mr. Fowler’s regressions to generate these metrics. However, I first 22 needed to estimate the data used by Mr. Fowler in his regression. Mr. Fowler did not include the 23 full data set he used for his regression analysis in his testimony. For example, Mr. Fowler includes 24 10 years of monthly usage in his Table 1(2008-2017), but presents his regression line of moving

25 average data in his Figure 1that span from 2006 to 2017. Mr. Fowler would have had to use data 26 from 2004-2017 to create the moving average data and regression. For example, the 2006 moving 27 average data point would be the average of 2004, 2005, and 2006. I was able to complete Mr. 28 Fowler’s presumed regression data set by obtaining additional years of data from the same source 29 used by Mr. Fowler, the WVAW Annual Reports. 30 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 7 of 12

1 9. Was there any additional information needed to recreate Mr. Fowler’s regression analysis that 2 was lacking from his testimony?

3 A. Yes. Mr. Fowler does not reveal which independent variables he uses in his regressions. Through 4 trial and error, I was able to determine that the independent variable he most likely used was 5 time. Based on this determination, and using the completed data set described in my previous 6 answer, I recreated Mr. Fowler’s regression analyses.

7 8 Q. Did you obtain the same results as Mr. Fowler with the recreated data set and the time 9 independent variable?

10 A. Yes. In both the commercial and residential regressions, I obtained the same R2 as Mr. Fowler

11 (91%) and the same approximate declining usage rates he cites in his testimony (451 gallons for 12 residential, and 3,739 for commercial). The recreated data set also visually aligns with Mr. 13 Fowler’s Figure No. 1and Figure No. 2 which gives me confidence that these models closely align 14 with Mr. Fowler’s.

15 16 Q. Based on your recreation o r. Fowler’s regressions, are there any issues that cause you to 17 question the regressions’ statistical validity?

18 A. Yes. Beginning with the residential model, although it does have the high R2 of 91% cited by Mr.

19 Fowler in his testimony, as well as an acceptable P-value for the independent variable and a strong

20 model F-statistic, the model is statistically unreliable due to the presence and effects of 21 autocorrelation of the error terms and should not be used to define or forecast a trend of for the

22 dependent variable, average use per customer. This is because the Durbin-Watson statistic’ of

In statistics, the Durbin-Watson statistic is a test statistic used to detect the presence of autocorrelation at the lag 1time period in the model residuals (prediction errors) from a regression analysis. It is named after James Durbin and Geoffrey Watson. The small sample distribution of this ratio was derived by John von Neumann (von Neumann, 1941). Durbin and Watson (1950, 1951) applied this statistic to the residuals from least squares regressions, and developed bounds tests for the null hypothesis that the errors are serially uncorrelated against the alternative that they follow a first order autoregressive process. The value of the statistic always lies between 0 and 4. If the Durbin- Watson statistic is substantially less than 2, there is evidence of positive serial correlation. As a rough rule of thumb, if the Durbin-Watson is less than 1.0, there may be cause to suspect autocorrelation of the error terms. Small values of the statistic indicate successive error terms are positively correlated. If the value of the statistic > 2, successive error terms are negatively correlated. In regressions, this can imply an underestimation of the level of statistical significance. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 8 of 12

1 .785 establishes that the error terms are not independent, violating the fundamental presumption 2 underlying a statistically valid linear regression analysis. This statistical issue is likely exacerbated 3 by Mr. Fowler’s decision to create the new moving average data set for his regression. The results

4 of our recreation and simulation of Mr. Fowler’s residential class regression are provided in Exhibit

5 GPR-1R. 6 7 . In regression modeling, hat is autocorrelation of the error terms and how does it irn 8 results from the regression analysis? 9 A. As explained in Appendix A attached to this rebuttal testimony, autocorrelation of the error terms

10 is a characteristic of data in which the correlation between the values of the same variables is

11 based on related objects. Informally, it is the similarity between observations as a function of the 12 time lag between them. In regression modeling, when autocorrelation is present, the estimate 13 errors follow a pattern, showing that something is wrong with the regression model. If the 14 presumption of unrelated error terms is violated and the error term observations are correlated, 15 autocorrelation is present. The presence of autocorrelation results in estimated variances of the

16 regression coefficients that will be biased and inconsistent, and therefore hypothesis testing is no 17 longer valid. In most instances, the R2 will be overestimated and the t-statistics will tend to be 18 higher. 19 20 owler’s cornmercial regression analysis suffer from the same effects o 21 correlation as his residentia~modeling does? 22 A. Yes. Despite reporting a high R2 of 91%, Mr. Fowler’s commercial regression also suffers from the 23 effects of autocorrelation of the error terms. Using the Durbin-Watson autocorrelation statistic 24 as an indicator of autocorrelation, the regression model reports a value of 1.218 which is slightly 25 below the upper threshold indicating the presence of autocorrelation for a model with 11 26 observations. The results of our recreation and simulation of Mr. Fowler’s commercial class 27 regression is provided in Exhibit GPR-2R. 28 29 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 9 of 12

1 Q. Due to the presence of autocorrelation of the error terms, does Mr. Fowler’s commercial 2 regression analysis suffer from the same statistical errors as his residential model?

3 A. Yes. As with the residential model, the presence of autocorrelation results in estimated variances 4 of the regression coefficients that will be biased and inconsistent, and therefore hypothesis

5 testing is no longer valid. Further it indicates that the R2 is likely overestimated and the t-statistics

6 will tend to be higher given a false measure of validity to the model. 7 8 Q. Mr. Roach, you testified earlier that Mr. Fowler used the data that had not been adjusted for 9 leaks and other miscellaneous adjustments in his regressions. Did you conduct an analysis with 10 the adjusted data using Mr. Fowler‘s moving averaging technique?

11 A. Yes. For purposes of illustration, setting aside the mathematical and time series analysis issues 12 introduced in Mr. Fowler’s modeling by use of a three-year moving average, in order to provide 13 some visibility to impact of unadjusted data set on Mr. Fowler’s regression analysis results I 14 completed regressions in his style with the adjusted usage data set. Both the residential and

15 commercial regressions resulted in much higher usage declines than reported by Mr. Fowler, 613 16 gallons per annum for residential, and 4,196 gallons per annum for commercial (versus the 451

17 gallons for residential, and 3,739 for commercial developed by Mr. Fowler). As with Mr. Fowler’s 18 models based on the unadjusted usage data, both models suffered from similar autocorrelation 19 issues experienced with the unadjusted times series data. The results of these regression

20 simulations are provided in Exhibit GPR-3, Pages 1and 2 of 2. 21

22 CONCLUSIONS 23 24 Q. r. Roach, would you please summarize your conclusions related to Mr. Fowler’s Residential 25 Commercial average use trend analysis.

26 A. Yes.

27 1. By employing data from the WVAW Annual Commission Report, Mr. Fowler failed to 28 employ usage and revenue data that had been revised for customer leak and

29 miscellaneous adjustments creating significant variances in the years of 2014-2016 in 30 pa rti c u I a r. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Gregory P. Roach Page 10 of 12

1 2. Mr. Fowler, by employing a three-year moving average of the dependent variable,

2 average use per customer, loses all fidelity and visibility to the annual reduction in average

3 use as illustrated in Table GPR-1R above. Due simply to his mathematics, Mr. Fowler‘s

4 analysis cannot provide any insight into the annual decline in average use for either

5 residential or commercial customers. Through application of a moving average 6 dependent variable term in his regression analysis, Mr. Fowler violates one of the basic

7 presumptions of statistical regression time series analysis. That is, that dependent

8 variables are associated with and forecasted using independent variables such as time or

9 climatic conditions from similar time periods, not those which occurred three years prior. 10 By employing a three-year moving average, Mr. Fowler lost all visibility to the actual 11 annual trend of residential and commercial reductions in usage per customer. The results 12 of his regression analysis do not address nor provide insight into the annual declines for

13 either residential or commercial customers. 14 3. Laying aside Mr. Fowler’s fundamental time series data issue created by using a three-

15 year moving average, both his residential and commercial regression models suffer from 16 autocorrelation of the error terms resulting in estimated variances of the regression 17 coefficients that will be biased and inconsistent, and therefore hypothesis testing is no

18 longer valid. Further it indicates that the R2 is likely overestimated and the t-statistics will

19 tend to be higher given a false measure of validity to the model. 20 4. Regardless of the time horizon employed and whether a regression of discrete 21 independent data points or a moving average is used, the rate of decline of residential

22 usage is 613-630 gallons per annum and the decline of commercial usage is 4,196-4,305

23 gallons per annum. Discounting Mr. Fowler’s moving average regression method due to 24 autocorrelation and mathematical issues, the correct rate of decline for WVAM

25 residential usage is 630 gallons per year and WVA~commercial usage decline is 4,305 26 gallons per year as per my direct testimony. 27

28

29 A. Yes. 30 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 11 of 12

1 Appendix A

2 Glossary of Technical and Statistical Terms

3 Autocorrelation - Autocorrelation is a characteristic of data in which the correlation 4 between the values of the same variables is based on related objects. Informally, it is the 5 similarity between observations as a function of the time lag between them. In regression 6 modeling, the estimate errors follow a pattern, showing that something is wrong with the 7 regression model. ... If this assumption is violated and the error term observations are 8 correlated, autocorrelation is present.

9 Cooling Degree Day - (“CDD”) A cooling degree day (CDD) is a measurement 10 designed to quantify the demand for energy needed to cool a building. It is the number of 11 degrees that a day’s average temperature is above 65” Fahrenheit (1 8” Celsius), which is 12 the temperature above which buildings need to be cooled. Annual CDD would be the sum 13 of all CDD occurring in a calendar year.

14 atson Stati$tic - The Durbin Watson statistic is a number that tests for 15 autocorrelation in the residuals fkom a statistical regression analysis. The Durbin-Watson 16 statistic is always between 0 and 4. A value of 2 means that there is no autocorrelation in 17 the sample.

18 F-Statistic - The F value is the ratio of the mean regression sum of squares divided by 19 the mean error sum of squares. Its value will range from zero to an arbitrarily large 20 number. The value of Probability (F) is the probability that the null hypothesis for the full 21 model is true (i.e., that all of the regression coefficients are zero). The higher the F value, 22 the greatest confidence that the null hypothesis can be rejected.

23 ay - (“HDD”) A heating degree day (HDD) is a measurement 24 designed to quantify the demand for energy needed to heat a building. It is the number of 25 degrees that a day’s average temperature is below 65 O Fahrenheit (1 8 O Celsius), which is 26 the temperature below which buildings need to be heated. Annual HDD would be the 27 sum of all HDD occurring in a calendar year.

28 - In statistics, the coefficient of determination, denoted R2 or r2 and 29 pronounced “R squared”, is the proportion of the variance in the dependent variable that 30 is predictable &om the independent variable(s).

31 tatistic - The t statistic is the coefficient divided by its standard error. The standard 32 error is an estimate of the standard deviation of the coefficient, the amount it varies 33 across cases. It can be thought of as a measure of the precision with which the regression West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Gregory P. Roach Page 12 of 12

1 coefficient is measured. The higher the t statistic, the greater probability is that the 2 regression coefficient has been estimated precisely. Exhibit GPR-1R Page 1of 1

West Virigina American Water Company Residential Average Use Model Employed by Fowler As Simulated with Annual Report Time Series Data §et

39 4a 42 03

1 3 5 Page I of I

West Virigina American Water Company Commercial Average Use ode1 Employed by Fowler As Simulated with Annual Report Time Series Data Set

10 11,807,176 f Exhibit GPR-3R Page 1 of 2

West Virigina American Water Company Residential Average Use Model Employed by Fowler As Simulated with Leak and Miscelaneous Adjustments to Usage

E Exhibit GPR-3R Page 2 of 2

West Vi~iginaAmerican Water Company Commercial Average Use Model Employed by Fowler As Simulated with Leak and iscelaneous Adjustments to Usage

Meen Enor MAE Minimum Maximum hiAPE AD”rls( MdSEIwr1 T 402151 627971 14 45 PUBLIC SERVICE COMMISSION OF WEST VIRGINIA CHARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-S-42T

WEST VIRGINIA-AMERICAN WATER COMPANY

Rule 42T Tariff Filing to Increase Rates and Charges

REBUTTAL TESTIMONY OF CAROLYN MOUNT

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Carolyn Mount Page 1of 5

1 Q* Please state your name and business address. 2 A. My name is Carolyn Mount. My office address is 1600 Pennsylvania Ave., Charleston, WV

3 25302.

4 5 Q. Did you provide direct testimony in this case? 6 A. Yes. On April 30, 2018, I filed direct testimony in this case. 7 8 Q. Please describe the purpose of your rebuttal testimony. 9 A. I will address the positions taken by Tara L. Gilkey on behalf of the Public Service Commission of

10 West Virginia (“Commission”) staff (“Staff”) and Ralph C. Smith on behalf of the Consumer 11 Advocate Division (“CAD”) regarding Long-term Performance Plan (“LTPP”) costs. West Virginia- 12 American Water Company, Inc. (“WVAW” or the “Company”) witness Jamie Hawn addresses the 13 Staff’s and CAD‘S payroll expense adjustments to base pay, including corresponding adjustments 14 to performance pay. 15 16 LTPP Costs - Companv and Service Company 17 18 Q. Please describe the adjustments proposed by Staff and CAD to the Company’s and American 19 Water Works Service Company, Inc.’s (“Service Company”) LTPP costs. 20 A. Staff and CAD propose to remove half of the LTPP costs for both WVAW and Service Company 21 employees based on the Commission’s February 24, 2016 Order in Case No. 15-0676-W-42T 22 (“2015 Rate Case Order”), where the Commission concluded that LTPP should be shared since

23 plan goals benefit both customers and shareholders. 24 25 Q. Do you agree with Staff’s and CAD’S proposed adjustments? 26 A. No. In the 2015 RC Order at page 49, the Commission directed that: 27 WVAWC in future rate cases should provide analysis that demonstrates 28 that the total compensation to its employees (both direct and AWWSC 29 employees) is in line with the market salary for each type of job 30 classification. The analysis should address how the market value for 31 each job classification is determined and provide examples that show West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Carolyn Mount Page 2 of 5

1 how the actual salaries for various job classifications compare to the 2 market-determined salaries. 3 4 The direct testimony and supporting benchmarking study of Company witness Robert V. Mustich 5 explained that, when assessed against the market, WVAW employees' target total 6 remuneration-base pay plus short-term and long-term variable pay at target levels (including 7 LTPP), plus benefits-is reasonable. In fact, Mr. Mustich's extensive compensation analysis shows 8 that WVAW employees' target total remuneration is actually lower than both Mid-Atlantic 9 regional market and national market median levels for comparable positions. (Mustich Dir., p, 6-

10 7). So, even if WVAW employees receive their total target performance payout, their total 11 compensation is still less than their market peers. And, as Mr. Mustich explained, if WVAW 12 employees don't receive any performance pay, their base salaries alone would put them 13 significantly below the market median. (Mustich Dir., p. 7-8). Neither Staff nor CAD acknowledge 14 this testimony, much less dispute it 15 16 Mr. Mustich's study demonstrates that LTPP is a reasonable and integral part of our total 17 compensation plan and should not be treated differently than overall compensation costs. As 18 part of overall compensation, LTPP is a reasonable cost of providing utility service. Therefore, it 19 should be assessed under the same lens as other operating costs: if it is prudently incurred and 20 reasonable in amount, relative to what the industry pays for the same services, it should be 21 recoverable through rates. Therefore, the Commission should concern itself when employee 22 compensation is too high, which may unreasonably increase rates, and when employee 23 compensation is too low, which may adversely impact service to customers through a number of 24 factors including the failure to attract and retain motivated, competent employees. It is an 25 undeniable fact that employees judge their compensation on its totality; including base wages, 26 performance-based wages and benefits. Management should have the flexibility to design the 27 compensation package that is most properly structured to compensate employees properly and 28 to motivate efficiency, safety, courtesy and other valuable employee traits. It is undisputed that 29 WVAW's employees are not overcompensated. It is not appropriate, therefore, to disallow any West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Carolyn Mount Page 3 of 5

1 portion of their compensation. Again, performance pay is not in addition to WVAW employees’ 2 reasonable compensation; it makes WVAW employees’ compensation reasonable.

3 4 Q. Why should the Commission not overlook Mr. Mustich’s study and testimony? 5 A. First, Mr. Mustich’s study and testimony addresses the exact areas that the Commission in 2016 6 directed the Company to address in its next rate case, and his findings relate directly to the 7 inquiries the Commission identified in the 2015 Rate Case Order. Second, employee 8 compensation is the key tool that WVAW uses to attract and retain the talented employees it 9 needs to meet its service obligations to West Virginia customers. Without performance pay, 10 including LTPP, WVAW employee compensation would be insufficient to attract and sustain a 11 qualified workforce. 12 13 Q. Do long-term performance pay programs such as the LTPP assist the Company in reducing 14 employee attrition? 15 A. Yes. Long-term financial-goal performance pay programs, like the LTPP, are particularly intended 16 to reduce attrition at the higher ranks of the organization. Excessive instability at that level may 17 have significant negative financial effects on the organization, such as on EPS, which ultimately 18 impact customer rates. So, as WVAW witness Mustich explains, these types of performance pay 19 programs are well-accepted in the industry. Importantly, the American Water LTPP achieves its 20 goals of reducing leadership attrition at a lower cost to customers than simply increasing 21 leadership’s base pay, because performance pay under the LTPP is stock-based. Because stock- 22 based compensation vests on a phased basis in three installments over a prospective three-year 23 period, employees must remain with the organization to realize the vesting of their awards. 24 25 Q. Please explain why it is particularly inappropriate for Staff and CAD to disallow Service Company 26 charges related to performance pay. 27 A. As Company witness Patrick Baryenbruchexplains in his direct and rebuttal testimony, the Service 28 Company provides services to American Water’s affiliates at cost and at prices that are more 29 advantageous than could be obtained in the market place (see e.g. Baryenbruch Direct at 3:9-12).

30 The Service Company, for example, provides legal, finance, accounting, engineering, design, West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Carolyn Mount Page 4 of 5

1 environmental, and customer services to WVAW and its regulated utility affiliates. The overall 2 question that a regulator should ask regarding these services is whether they are reasonable when

3 compared with services that the Company can obtain in the market. If, for example, WVAW were 4 to obtain operating services from the market, like an outside engineering firm, you would not 5 expect an adjustment for that firm’s performance-based compensation plan, even if the plan 6 included financial goals. Rather, the Commission would assess the reasonableness of the 7 engineering firm’s costs relative to the market, as it did Service Company costs.

8 Q. Are WVAW’s Service Company charges reasonable? 9 A. Yes. Mr. Baryenbruch testifies on the value of Service Company costs and demonstrates that they 10 are equal to or less than the costs we would have to pay for equivalent services. The Service

11 Company is providing WVAW-and its customers-enhanced value, at a reasonable cost. It is 12 inappropriate to disallow a component of that cost simply because it doesn’t comport with 13 historical views of employee compensation.

14 15 Q. Do you believe the fact that both customers and shareholders benefit from LTPP goals should 16 be a basis for disallowing half of the LTPP expense? 17 A. No. All of the financial and operational performance metrics provide benefits to our customers 18 and shareholders. Overall employee compensation should be assessed like every other prudently 19 incurred operating expense. The question is “are WVAW’s total salaries and benefits 20 reasonable?” Mr. Mustich has demonstrated unquestionably that they are. Therefore, this 21 reasonable operating expense should not be disallowed. 22 23 Correction of Direct Testimony 24 25 Q. Was there a misstatement in your direct testimony? 26 A. Yes, there was, and I would like to correct it. At page 4, lines 1-4 of my direct testimony, I stated 27 that the Company’s medical, prescription, dental, and vision insurance costs have remained flat 28 or slightly decreased since the last case. In our review of data in preparation of a response to West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of Carolyn Mount Page 5 of 5

1 CAD-3-E-29, we realized that this was a misstatement. The Company’s supplemental response to 2 CAD-3-E-29, attached here as Exhibit CM-1, addresses this issue.

3

4 Q. Does this conclude your rebuttal testimony? 5 A. Yes Exhibit CM-1 (1 of 4) WEST VIR~INIA-AMERICANWATER COMPANY CASE NOS. 18-0573-W-42T and 18-0576-S-42T W PUBLIC SERVICE COMMISSION - CONSUMER ADVOCATE DIVISION THIRD SET OF DISCOVERY - RECEIVED SEPTEMBER 5,2018

Prepared by: Carolyn Mount, Director HR Business Partner and Jamie Hawn, Senior Manager, Regulatory Services Witness: Carolyn Mount, Director HR Business Partner and Jamie Hawn, Senior Manager, Regulatory Services Date Prepared: September 25,2018

Date of Supplement: October 12,2018

CAD-3-E-29

Group Insurance Programs. Refer to the direct testimony of Company witness Mount and Statement G, Adjustment No. 36 from the HTY Rule 42 filing. On page 4 (lines 1-4) of her testimony, Ms. Mount states that the Company’s medical, prescription, dental and vision insurance costs have remained flat or slightly decreased since the last rate case. In addition, Ms. Mount states that the Company’s life insurance, short-term and long-term disability costs have remained flat or slightly increased. a. Referring to Statement G, Adjustment No. 36, for the Company’s union, nonunion hourly and non- union salaried employees, please provide a breakout of the test year group insurance costs of $2,593,588 by the categories shown for the going level amounts which total $4,638,958 prior to application of the O&M expense percentage. b. Pursuant to part “a”, please provide similar data for each year since the Company’s last rate case (Case No. 15-0576-W-42T)in the format shown on Statement G, Adjustment No. 36.

RESPONSE: a. Please see CAD-3-E-29-Attachment, tab named “Subpart A - 2017” for a breakout of the test year group insurance costs for 12 months ended December 31,2017. b. Please see CAD-3-E-29-Attachrnent’ tab named “Subpart B - 2016 for a breakout of the group insurance costs for 12 months ended December 31,2016. Also, see tab named “Subpart B - 2015” for a breakout of the group insurance costs for 12 months ended December 31,2015.

The Company is currently reviewing the difference between the amount reflected in the last case versus the current filing.

RESPONSE: Supplemental 10/12/2018

Upon review, the Company has determined that the statement at page 4, lines 1-4 of Ms. Mount’s direct testimony that the Company’s medical, prescription, dental and vision insurance costs have remained flat or slightly decreased since the last rate case is incorrect, as is borne out by the group insurance cost information in CAD-3-E-29-Attachment. The Company intends to correct this error in Ms. Mount’s rebuttal testimony. Exhibit CM-1 (2 of 4)

West Virginia American Water Company - Water CAD-3-E-29-Attachment CAD-3-E-29-Atta chme n t Case No. 18-0573-W-42-T Group Insurance Expense For the Twelve Months Ended December 31,2017

Medical Total Gross Dental & Employee Net Medical, Long Term Short Term Group Vision Contribution Dental, Vision Basic Life AD&D Disability Disability Insurance Union $3,020,251 ($478,145) $2,542,106 $26,319 $392 $0 $11,660 $2,580,477 Non Union Hourly 1,105,618 (178,781) 926,836 13,478 51 9,656 4,641 954,662 Non Union Salaried 886,498 (138,478) 748,020 16,789 61 12,501 3,703 781,075 Total Group insurance Expense $5,012,367 , ($795,404) $4,216,963 $56,586 $504 $22,157 $20,004 $4,316,213

(1,730,834)Capitalized Portion (5,121)Retiree Contribution 13,299 Miscellaneous Entries/Accruals $2,593,558 2017 Test Year Per Book Expense

Page 1 of 3 Exhibit CM-1 (3 of 4)

West Virginia American Water Company - Water CAD-3-E-29-Attachment CAD-3-E-29-Attachment Case No. 18-0573-W-42-T Group Insurance Expense For the Twelve Months Ended December 31,2016

Medical Total Gross Dental & Employee Net Medical, Long Term Short Term Group Vision Contribution Dental, Vision Basic Life AD&D Disability Disability Insurance Union $2,588,126 ($442,855) $2,145,271 $26,675 $400 $0 $3,263 $2,175,609 Non Union Hourly 948,367 (160,494) 787,873 12,146 1,389 9,968 1,253 812,629 Non Union Salaried 731,141 (123,759) 607,382 14,449 1,640 12,064 941 636,477 Total Group Insurance Expense $4,267,633 ($727,108) $3,540,526 $53,271 $3,429 $22,032 $5,457 $3,624,715

(1,648,295)Capitalized Portion (9,490)Retiree Contribution 669,584 Miscellaneous Entries/Accruals $2,636,514 2016 Test Year Per Book Expense

Page 2 of 3 Exhibit CM-1 (4 of 4)

West Virginia American Water Company - Water CAD-3-E-29-Attachment CAD-3-E-29-Attachment Case No. 18-0573-W-42-T Group Insurance Expense For the Twelve Months Ended December 31,2015

Medical Total Gross Dental & Employee Net Medical, Long Term Short Term Group Vision Contribution Dental, Vision Basic Life AD&D Disability Disability Insurance Union $2,468,163 ($421,438) $2,046,725 $25,990 $419 $0 $2,074 $2,075,208 Non Union Hourly 948,904 (165,521) 783,383 12,255 1,392 6,615 815 804,459 Non Union Salaried 700,754 (122,287) 578,467 13,620 1,547 7,501 553 601,689 Total Group insurance Expense $4,117,822 ($709,247) $3,408,575 $51,865 $3,358 $14,116 $3,442 $3,481,356

(1,298,407) Capitalized Portion (41,191) Retiree Contribution 42,914 Miscellaneous Entries/Accruals $2,184,672 2015 Test Year Per Book Expense

Page 3 of 3 Rule riff Filing to increase s and Charges

F West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 1of 10

1 Please state your name an business address. 2 A. My name is John R. Cox. I am employed by American Water Works Service Company, Inc.

3 (“AWWSC”) as Director of Rates and Regulations. 4

5 Did YOU provide direct testimony in this case?

6 A. Yes. On April 30, 2018 I filed direct testimony in this case on behalf of West Virginia American 7 Water Company (“WVAWC” or the “Company”). In addition, I will be adopting the rate base

8 portion of the direct testimony of Company witness Daniel P. Hunnell II who is no longer employed

9 at AWWSC. 10 11 Please describe the purpose of your rebuttal testimony.

12 A. I will address rate base, including cash working capital, adjustments proposed by Staff witness

13 Tara L. Gilkey and Consumer Advocate Division (“CAD”) witness Ralph C. Smith in their direct

14 testimony. My rebuttal testimony is composed of the following sections,

15 I. Comments on the Staff and CAD testimony recommending the Commission

16 disallow all the Company’s plant additions and associated adjustments during the

17 Addendum Period of January 1,2017 to February 25,2018.

18 II. Comments on CAD’S other proposed adjustments to rate base.

19 Ill. Comments on Staff‘s other proposed adjustments to rate base.

20

21 1. A$$endum Per~o$Plant Add~tions

22 in Ms. Gilkey’s an r. Smith’s recomm treatment of the C

23

24 A. For water rate base, both Ms. Gilkey and Mr. Smith recommend terminal treatment for test year

25 Transmission and Distribution Plant, and the test year thirteen-month average balance for all West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 2 of 10

1 other utility plant in service (“UPIS”). For wastewater rate base, Ms. Gilkey recommends a test

2 year thirteen-month average balance for all UPIS. (Mr. Smith did not address the Company’s

3 proposed wastewater rate base.) This treatment results in no recognition in base rates of any

4 utility plant additions made by the Company after December 31,2017 and only average rate base

5 for Non-Transmission and Distribution Plant put in service during 2017.

6

7 Q. Please describe mith’s rationaie for not including A

8

9 A. Both Ms. Gilkey and Mr. Smith rely on the premise that the Company now has a DSlC program

10 that provides an immediate return on its DSIC-eligible plant investment, which they claim negates

11 the need for inclusion of UPIS beyond the historic test year.

12

13

14 een- averag~~istor~~a~ rate base n-Tra~s~issionan

15

16 A. No. First, this issue was fully litigated in the Company’s last base rate cases at Case Nos. 15-0676-

17 W-42T and 15-0675-5-42T (“2015 Rate Case”). In that case, the Commission exhaustively

18 addressed the question of whether to modify the standard historic test year terminal rate base

19 treatment for the Company. The Commission found that HN-based rate base recognition had

20 failed to provide a reasonable matching of revenues, expenses, and rate base present in the year

21 rates would go into effect (“Rate Year”). 2015 Rate Case Order at 26. Consequently, the West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 3 of 10

1 Commission decided to provide full recognition of terminal rate base treatment for all non-

2 revenue producing additions at the beginning of the Rate Year (Addendum Period):

3 It is just and reasonable to establish rate base at the beginning of the Rate Year 4 because inclusion of additional investment in rate base elements for the 5 Transition Period (i) will provide a reasonable level of known and measurable rate 6 base that will be used and useful and in service at the time the new rates 7 authorized in this proceeding become effective, (ii)will provide a better matching 8 of revenues, expenses and rate base present in the Rate Year than would 9 adherence to a non-representative HTY approach, and (iii)will better mitigate the 10 impact of regulatory lag than would AFFAC. 11 12 2015 Rate Case Order at 96 (Conclusion of Law 9). In reaching this determination, the Commission

13 said that it was “at a crossroads” in the Company’s rate regulation, and that the Company had

14 met its burden of proof that thirteen-month average HTY rate base regulation is inadequate given

15 the Company’s unique combination of declining residential usage, little to no growth, and the

16 need for costly system repairs - a unique combination that still exists today. 2015 Rate Case Order

17 at 26.

18 Second, if the Commission approves the Staff and CAD recommendation of HTY treatment

19 for rate base, then the Company would experience extreme regulatory lag on all of its non-DSIC

20 rate base investment - a lag that could be as long as three to four years, depending on when the

21 next base rate case is filed. In other words, Staff and CAD recommend a return to a “non-

22 representative HTY approach” that the Commission found not to match Rate Year rate base with

23 expected revenues and investment in that year.

24

25 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 4 of 10

1 PIS investment that was the asis for the rates set by the

2 ~o~missionin the 2

3 A. Yes. As shown below, the Company exceeded the UPlS investment levels utilized by the

4 Commission for terminal rate base recovery by $13 million.

5

UtRtry Plpnt fn SLnrlcc e8 of Addendum Year Ended 02/2016

02fZOX6 02/N)16 Addemdurn Period Addendum Period

$7~1,~7~,2~0$794,198,915 $13,125,705

7379,962 7.735,675 ($144,287)

53,172 $801,9~4,5~$12,981,41% 6

7

8

9 A. No, not with respect to rate base investment the Company has already made - investment that

10 would be stranded without any recovery until rates from the next base rate case go into effect.

11 The DSlC is a valuable tool, but by its design, it primarily addresses ~~o~ec~e~infrastructure

12 investment. Moreover, between the 2015 Rate Case and the effective date of new rates in this

13 rate case, the Company will have invested over $142 million in plant additions that have not been

14 included in DSlC rate recovery. Under the Staff/CAD recommendation, over $42 million of this

15 investment would not be reflected in base rates in this case or a 2019 DSlC as proposed by Staff

16 or CAD.

17 Although these observations on the usefulness of a DSIC are accurate, from the

18 Commission's perspective, they miss the larger point: there is no question that when the West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-421 Rebuttal Testimony of John R. Cox Page 5 of 10

1 Commission approved the Addendum Period rate base recognition concept, it did so knowing that

2 it had simultaneously authorized the Company to seek Commission approval of a Staff-

3 recommended surcharge mechanism very much like the DSIC. The Commission expressly found

4 that such a mechanism “would also be effective in reducing the regulatory lag . . .,” 2015 Rate

5 Case Order at 26. It even appears that the Commission may have considered approval of the Staff

6 recommendation if only the Company had proposed and publicly noticed it, noting that

7 “unfortunately,” the Company had not. jcJ.

8

9 If all of the rate base in service throu ddendu~Period is recognize in base rates, would

10

11 A. No. The DSIC only applies to certain items of rate base that are not being recovered in base rates.

12 If those DSIC eligible items are included in base rate, they are not recovered thorough the DSIC.

13 The argument raised by Staff and CAD - that recognition of rate base through the Addendum

14 Period is made unnecessary by the DSIC - is simply insupportable and irrelevant to the facts.

15

16 entation as a basis to oppose the

17 Co~~ission’sauthorization of the

18

19 A. No, neither of them did. In my view, neither witness has provided cogent evidence that the

20 Company’s combination of declining residential usage, little to no customer growth, and

21 increasing need for infrastructure replacement no longer exists. Because these conditions that

22 the Commission cited to in approving the Addendum Period still exist there is no justification for West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of John R. Cox Page 6 of 10

1 Ms. Gilkey and Mr. Smith requesting that the Commission unravel the rate base approach it

2 approved in the 2015 Rate Case.

3

4 . What is the Company’s reco~mendationfor the Comm~ssion?

5 A. The Company recommends the Commission establish terminal rate base at the beginning of the

6 rate year and reject Ms. Gilkey’s and Mr. Smith’s attempts to revert to a thirteen-month average

7 plant balance for the historic test year and eliminate the Addendum Period investmentsfrom rate

8 base. The Staff and CAD recommendations would result in an abrupt about-face to the

9 constructive rate policies recently adopted by the Commission in our most recently concluded

10 case and reintroduce the regulatory lag that the Commission sought to address in the 2015 Rate

11 Case, to the detriment of the Company and its customers.

12

13 II.

14 Please res~on itness ith’s rec ation e Cull m

15 regulato~asset from rate

16 A. Based on the Company’s Joint Stipulation and Agreement for Settlement filed in the General

17 Order No. 236.1 proceeding (In the Matter ofthe Effects on Utilities of the 201 7 Tax Cuts and Jobs

18 Act), Mr. Smith recommends removing the regulatory asset associated with the Culloden Dam

19 from rate base. The CAD agreed that the remaining portion of the Culloden Dam regulatory asset

20 should be amortized over three years as provided in the General Order 236.1 Stipulation. The

21 Company agrees with these recommendations.

22 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 7 of 10

1 Q. Please respond to CAD witness Mr. Smith’s recommendation on the payment lag for AWWSC

2 charges in the Company’s proposed cash working capital (CWC) calculation.

3 A. Mr. Smith recommends that a 12.0-day lag for AWWSC charges to WVAWC, which approximates

4 a payroll lag, be used when calculating the CWC for service company charges. Mr. Smith

5 references a Commission-ordered adjustment in Case No. 08-0900-W-42T to reflect a payroll-type

6 lag for AWWSC charges. The Company does not dispute Mr. Smith’s recommendation on this

7 point.

8 r. Smith address the issue of non-cash items in the CWC calculation?

9 A. Mr. Smith said that although he does not agree that non-cash items such as depreciation and

10 amortization expense and deferred income taxes should be included in the CWC calculation, he

11 acknowledges that the Commission permitted the inclusion of a revenue lag for these items in the

12 2015 Rate Case, and consequently he has chosen not to remove them.

13 Ill.

14 toric test year thirteen-

15 ~onth awe rage balanc ork in Progress,

16 Investment Tax Cre

17 A. No. In the 2015 Rate Case Order, the Commission approved the use of terminal rate base at the

18 time new rates go into effect. Accordingly, in this case, the Company has requested rates based

19 on a terminal rate base at February 28, 2019. The Company started with the balances of all rate

20 base components at December 31,2017. The Company then updated the major components of

21 rate base through the Addendum Period to arrive at a terminal rate base at February 28, 2019.

22 The use of historic test year terminal rate base balances (at December 31, 2017), with major rate West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 8 of 10

1 base component balances updated through the Addendum Period (at February 29, 2019),

2 provides for a better matching of rate base components for the Addendum Period than starting

3 with a historic test year thirteen-month average as recommended by Ms. Gilkey.

4

5 Q. Please discuss Ms. Gilkey’s adjustments to

6 . Ms. Gilkey also applied thirteen-month average treatment to these two Company adjustments.

7 Adjustment 72 adjusts acquisition costs from an HTY thirteen-month average to a year-end

8 balance, and Adjustment 73 makes the same adjustment for the unamortized equity portion of

9 the AFFAC after its discontinuation on February 26,2016. For the same reasons that apply to the

10 recognition of the Addendum Period in rate base, the Company believes that these two Staff

11 adjustments are not warranted.

12

13 lease discuss ilkey’s adjust~e~tto CIAC.

14 A. In Staff Adjustment No. 73, Ms. Gilkey treated negative acquisition adjustments of $450,804 as a

15 credit to contributions in aid of construction (CIAC), pursuant to the ruling in WVAWC Case No.

16 03-0353-W-42T. The Company accepts this recommendation but disputes the amount of the

17 adjustment recommended by Ms. Gilkey. In developing the amount of the asset book value in

18 excess of the amount paid that should be credited to CIAC, one must consider the offset of the

19 accumulated amortization of the negative utility plant acquisition adjustment (UPAA) to utility

20 plant. The chart below shows the asset book value of the negative UPAA at December 31, 2017.

21 The Company agrees to credit CIAC for $2,299. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Pane 9 of 10

1

20 Year Balance Amortkation End 12/31/2017 AUB-16 16 17 MOV-17 D&S 17 Big Sandy -19 42,2991

2

3

4

5 A. Based on a stipulated resolution of the Company’s 2001 rate case, Ms. Gilkey recommended that

6 the unamortized balance of the acquisition costs of the Parkersburg municipal water system not

7 be included in rate base. The Company accepts this adjustment.

8

9 s. Gilkey’s adjust men^ to property tax lag days.

10 A. Ms. Gilkey recommends 517 lag days, as ordered by the Commission in the 2015 rates case, be

11 utilized in the calculation of CWC instead of the Company’s 509 lag days. The only difference

12 between the Company’s lag days of 509 and Ms. Gilkey’s lag days of 517 is that the Company

13 utilized updated information based on the same concept as Commission authorized. The

14 Company recommends that the Commission adopt the Company’s 509 lag days for property

15 taxes.

16 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Cox Page 10 of 10

1 Q.

2 A. Ms. Gilkey made an adjustment to rate base based on the over capitalization of OPEB expense, as

3 ordered in Case No. 95-0228-W-42T’ which is similar to the adjustment the Company is

4 requesting. The main difference between Ms. Gilkey’s recommended adjustment and the

5 Company’s recommended adjustment is that Ms. Gilkey’s adjustment is based on a thirteen-

6 month HTY balance while the Company adjustment is based on the Addendum Period. In

7 addition, Ms. Gilkey’s and the Company’s capitalization ratios (which are based on actual

8 experience) differ slightly for certain years. The Company recommends that the Commission

9 adopt the Company’s adjustment to rate base for overcapitalization of OPEBs as filed in the case.

10 u na sts en

11 asset from rate base.

12 A. As discussed above, pursuant to the provisions of the Joint Stipulation and Agreement for

13 Settlement filed in the General Order No. 236.1 proceeding, the Company is removing the

14 Culloden Dam from rate base.

15 ase adjust~entsto Sewer.

16 A. Ms. Gilkey is recommending adjustments to the Company’s sewer rate base similar what she

17 recommended for water, including use of a thirteen-month average balance for ail utility plant in

18 service and other rate base components. For the reasons set forth above, the Company

19 recommends the Commission reject s. Gilkey’s proposed adjustments and adopt the Company’s

20 rate base, included in the addendum, which is based on terminal rate base treatment at the time

21 new rates will go into effect.

22 ? 23 A. Yes. PUBLIC SERVICE OMM MISSION OF WEST VIRGINIA CHARLESTON

Case No. 18-0573-W-42T Case No. 18-0576-S-42T

WEST VIRGINIA-AMERICAN WATER COMPANY

Rule 42T Tariff Filing to Increase Rates and Charges

REBUTFA1 TESTIMONY OF JOHN R. WILDE

October 12,2018 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of John R. Wilde Page 1of 7

1 Q. Please identify yourself.

2 A. My name is John R. Wilde. I am employed by American Water Works Service Company, Inc. as 3 Vice President -Tax, and I provided direct testimony on behalf of West Virginia-American Water 4 Company ("WVAW" or the "Company") in this docket filed on April 30,2018. 5

6 Q. What is the purpose of your rebuttal testimony? 7 A. The purpose of my rebuttal testimony is to address testimony filed by Consumer Advocate 8 Division ("CAD") witness Ralph C. Smith relating to the Company's treatment of excess 9 accumulated deferred income taxes ("EADIT") resulting from the enactment of the Tax Cuts and

10 Jobs Act of 2017 (the "TCJA"). Specifically, I address the following EADIT-related 11 recommendations by Mr. Smith: 12

13 1. WVAW's should be required to justify and support its classification of EADIT between 14 "protected" (required to meet IRS normalization requirements that pertain to the use of 15 accelerated tax depreciation) and "unprotected" (for which IRS normalization 16 requirements to not apply and hence for which disposition is up to the Commission's 17 discretion). 18 19 2. WVAW's appropriately classified "protected" EADIT should be amortized according to the 20 Average Rate Assumption Method ("ARAM"). Compliance with IRS normalization 21 requirements is necessary to preserve the utility's ability to utilize accelerated tax 22 depreciation. 23 24 3. WVAW's EADlT should be appropriately categorized between "protected" and 25 "unprotected" and an appropriate amortization period for the "unprotected" EADIT 26 should also be addressed and applied in the current proceeding. 27 28 4. The EADIT related to tax deductions for repairs should be classified as "unprotected." The 29 amortization of "unprotected" EADIT is subject to the discretion of the Commission. Once 30 the amounts of WVAW's "unprotected" EADIT are known, an amortization period should 31 be determined to rapidly return those amounts to customers in the form of reductions to 32 WVAW's revenue requirement in the current base rate case, or as customer bill credits, if 33 the quantification occurs after base rates are set in the current rate case. 34 35 5. Due to WVAW's delays in quantifying EADIT and properly classifying it between 36 "protected" and "unprotected," and providing the related EADIT amortizations for 2018 37 through the effective date of rates in the current WVAW base rate case, interest 38 calculated at WVAW's most recently authorized weighted average cost of capital West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Wilde Page 2 of 7

1 (”WACC”) should be applied on the EADIT balances from January 1,2018 through the rate 2 effective date when such EADIT amortization through the date that EADIT amortization 3 begins to flow back to WVAW rate payers. 4 5 Q. Has the Company entered into a Joint Stipulation and Agreement for Settlement with CAD, 6 Staff, and other parties in the Commission’s proceeding, In the Matter ofthe Effects ofthe 2017 7 Tux Cuts undlobs Act, General Order No. 236.1 (the “TUA Stipulation”)? 8 A. Yes, it has. 9 10 9. Does the TUA Stipulation provide for the treatment and disposition of the Company’s net 11 EADITs? 12 A. Yes, it does. Paragraph 14 provides as follows: 13 14. Net Excess ADITS. 14 The Parties agree and recommend that after the Company has determined (i)its 15 protected and unprotected accumulated deferred income taxes balances 16 (“Excess ADITS”) and (ii) the appropriate amortization periods through the 17 application of the Average Rate Assumption Method (“ARAM”), the Company 18 should implement a credit to customers to the extent consistent with federal 19 normalization requirements through the mechanism described in subparagraph 20 (b) below in an amount equal to the combined annual base rate impact of 21 22 i. the flow back of the protected portion of the Excess ADITS as determined 23 under ARAM; and 24 25 ii. the amortization of the unprotected portion of the Excess ADITS over a 26 period of years to be determined at that time; and 27 28 iii. a corresponding adjustment to rate base to avoid an inconsistency between 29 the calculation of deferred tax expense and the rate base established in the 30 2018 Rate Case, 31 32 together and expressed as an annual revenue requirement, the “Excess ADIT 33 Credit.” 34 35 The Parties agree and recommend that in the Commission’s final order in the 36 2018 Rate Case, the Commission should establish a second rate change through 37 which the Company will implement the Excess ADIT Credit to take effect on a 38 target date of July 1, 2019 (or as soon as is reasonably possible after the West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-S-42T Rebuttal Testimony of John R. Wilde Page 3 of 7

1 determination of the Excess ADIT Credit) and in accordance with the procedure 2 described in subparagraph (c) below. 3 4 c) As soon as is reasonably possible, the Company will file schedules showing the 5 amount, components, and calculation of the Excess ADIT Credit as a closed entry 6 in the 2018 Rate Case docket. Staff and CAD may file any response within 20 days 7 of the Company’s filing, and the Company may file any reply within 5 days of 8 receiving such response. Based on these filings and any hearing it may wish to 9 hold, the Commission will issue an order establishing the Excess ADIT Credit 10 effective for service rendered on and after July 1, 2019 (or such later date to be 11 set by the Commission as circumstances may require). 12 13 d) The Parties agree and recommend that the rate reduction associated with the 14 Excess ADIT Credit should remain in effect until new rates from the Company’s 15 next base rate case go into effect, and that the continuation of the flow back of 16 protected Excess ADITS and the amortization of the unprotected Excess ADITS 17 should be incorporated into the revenue requirement approved in that base rate 18 case. 19

20 Q. Does the TUA Stipulation address any of Mr. Smith’s recommendations? 21 A. Yes, it does. Specifically, it addresses his first three EADIT-related recommendations listed above.

22 It provides a process and timeline for the Company’s determination of its protected and

23 unprotected EADIT balances (recommendation no. l),the determination of the appropriate 24 amortization periods through the use of ARAM (recommendation no.2), and the amortization of 25 unprotected EADIT (recommendation no. 3). I am advised that the TCJA Stipulation has been 26 approved by the Commission. Therefore, to the extent Mr. Smith’s recommendations are 27 inconsistent with its terms, those recommendations should be rejected. 28

29 Q. In his fourth EADIT-related recommendation, Mr. Smith asserts that the Company’s tax 30 deduction for repairs should be treated as “unprotected,” and therefore not subject to IRS 31 normalization requirements. Do you agree? 32 A. No. While the repairs deduction is not “protected” under IRS rules, the Company, as I stated in 33 my direct testimony (p.7), is required to normalize repairs as a condition of a consent agreement 34 with the IRS. WVAW’s parent company, American Water Works Company, Inc. (“American 35 Water”) qualified for the repairs deduction through a Federal Form 3115 “Application for Change 36 in Accounting Methods”, which was filed for the taxable year ended December 31, 2008. That West Virginia American Water Company Case Nos. 18-0573-W-42Tand 18-0576-5-42T Rebuttal Testimony of John R. Wilde Page 4 of 7

1 application resulted in a consent agreement that was signed by the IRS on July 30, 2010 and by 2 American Water on September 10,2010. (A copy of the consent agreement is attached as Exhibit 3 JRW-1.) The consent agreement requires the Company to use a normalized method of accounting 4 to account for those repair deductions. The consent agreement thus dictates how the EADIT 5 associated with the repairs must be addressed. Absent clear direction or guidance from the IRS 6 to the contrary, WVAW is required to comply with the IRS consent agreement, or else risk the loss 7 of all or part of the benefits it has achieved on behalf of customers in accelerating tax deductions 8 by applying its tax repairs method of accounting. 9

10 Q. Is Mr. Smith’s recommendation to treat the EADIT resulting from repairs deduction in a manner 11 inconsistent with a normalization method of accounting prudent?

12 A. No. Mr. Smith does not address WVAW‘s obligations to abide by the consent agreement, or the 13 consequences that might accrue to WVAW and its customers if the Company violates the terms 14 and conditions set forth the consent agreement it signed that secured the permission necessary 15 to change its method of accounting for tax repairs. Violating the consent decree, and the 16 consequences of an IRS finding that WVAWC is using an impermissible method of accounting, 17 could have a significant impact on customers, including a loss of the entire ADIT balance related 18 to those repair deductions that would include the EADIT piece. If Mr. Smith’s recommendation 19 were adopted, his goal of forcing swift payback of EADIT would be realized -the problem would 20 be that the payback could flow to the IRS, not to customers. Also, Mr. Smith proposes to bifurcate 21 and flow through a tax benefit to customers at a rate that is different from the rate at which the 22 underlying book expense will be recovered from customers. This will cause intergenerational 23 issues with respect to the use of the WVAWC distribution system and the cost to use that system. 24 Therefore, Mr. Smith’s recommendation should be rejected as contrary to customers’ interests. 25 26 West Virginia American Water Company Case Nos. 18-0573-W-421 and 18-0576-S-42T Rebuttal Testimony of John R. Wilde Page 5 of 7

1 Q. In his fifth EADIT-related recommendation, Mr. Smith recommends that the Company be 2 required to accumulate interest on its EADIT balances from January 1, 2018 through the date 3 the amortization of EADIT begins. Do you agree? 4 A. No. There is no reason to accumulate interest on EADIT since the customer is not being harmed 5 and the Company is not benefiting from the lower ADIT amount in rate base. The split of a 6 subsidiary utility’s EADIT between what is to be considered “protected” or “unprotected” is a 7 complex analytic, as is the application of the method to be used to normalize the balance classified 8 as protected. The time that a utility has to complete those tasks is subject to the current state

9 of the utility’s data structure and system configuration at the time the law is enacted, and the 10 ultimate goal is deriving a result that best serves customers interests, which includes not violating 11 the tax normalization rules. Moreover, if the parties and the Commission had intended that 12 interest be applied to EADIT balances, that provision would have appeared in the TCJA Stipulation 13 or in the Commission‘s approval of it. Applying interest now would contravene the TCJA 14 Stipulation.

15

16 Q. Mr. Smith suggests that interest is appropriate because the Company’s management “caused” 17 the delay in classifying protected and unprotected EADIT and determining the related 18 amortization periods (pp. 79-80,82). Do you agree? 19 A. No. Prior to the passage of the TCIA, the Company was not required to use ARAM. In the absence 20 of such a requirement, there was no reason to expend the resources necessary to upgrade and 21 configure its systems to do so. 22

23 Q. When does the Company anticipate completing the process of classifying protected and 24 unprotected EADIT and determining the related amortization periods?

25 A. American Water as a whole, including WVAW, is already in the process of working to break down 26 its records necessary to calculate protected and unprotected EADIT. It is implementing software 27 to manage this in an efficient and accurate manner. It is anticipated that it will take until the end 28 of Q1 2019 to complete the process and at that time WVAW will be able to split EADIT between 29 “protected” and “unprotected” and to calculate the exact return of the EADlT to customers using 30 ARAM. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Wilde Pane 6 of 7

1

2 Q. Why will it take so long to update the records to determine a precise breakdown between the 3 “protected” and “unprotected” amounts and calculate ARAM? 4 A. The Company has not kept its deferred tax inventory in the detail needed for ARAM because it 5 was not using ARAM. In addition, Repairs deductions are normalized, and as such there is no 6 need to separate repairs from any other plant-related differences. For example, Repairs has its 7 own line in the Company’s deferred tax inventory. This line is mostly the tax return deductions 8 taken, but also included in the balance are net adjustments of prior deductions taken (IRC Section 9 481(a) adjustments). So, in order to calculate true deferred taxes related to the Repairs 10 deductions, the Company first needs to break down those 481(a) adjustments between basis 11 adjustments and accumulated depreciation adjustments. Then it needs to relate book 12 depreciation and book retirement information that occurred over life that should be associated 13 with those tax repair deductions. Repairs is a temporary difference, and the book information is 14 what will drive the reversal of the deductions. Currently, the book depreciation information, 15 whether related to repair book property or non-repair book property, ends up on other line items 16 within the deferred tax inventory. The Repairs line in the deferred tax inventory is mostly an 17 accumulation of tax return deductions. It is not a completely self-containedtemporary difference. 18 Therefore, one cannot calculate the excess deferred taxes on that one line item and know that it 19 is correctly classified as “protected” or “unprotected”. As this information is determined and 20 developed, the Company will refine how much is truly in the “protected” bucket for method/life 21 differences. This is also the case for other basis differences and is one of the reasons for the time 22 line required to complete the project and be comfortable that the Company can sustain its 23 numbers under IRS review. 24

25 Q. Does the TUA Stipulation provide the Company with time to complete the process of classifying 26 protected and unprotected EADlT and determining the related amortization periods?

27 A. Yes, it does. The parties, including CAD, agreed that in its final order in this rate case, “the 28 Commission should establish a second rate change through which the Company will implement 29 the Excess ADiT Credit to take effect on a target date of July 1, 2019 (or as soon as is reasonably 30 possible after the determination of the Excess ADIT Credit).” The target implementation date of West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of John R. Wilde Pane 7 of 7

July 1, 2019 for the EADIT Credit is calculated to allow the Company to implement and complete

L its ARAM calculations as described above. 3 4 Q. Does this conclude your rebuttal testimony? 5 A. Yes. IExhibit JRW-3 I

AMERICAN \WATER September 10,2010

Courier's Desk Internal Revenue Service Attn: CC:ITA:BOI- Innessa Glazman 11 11 Constitution Avenue, N.W., Room 5336 Washington, DC 20224

RE: American Water Works Company, Inc. & Subs. EN: 5 1-0063696 CAM- 10842 1-09 CONSENT AGREEMENT

Dear Ms. Glazman:

This letter relates to a Form 3 115, Application for Change in Accounting Method, filed by the above-mentioned Taxpayer on behalf of itself and various subsidiaries, requesting permission to change their method of accounting for (1) costs to repair and maintain tangible property, and (2) .dispositions of certain tangible depreciable property, for the taxable year that ended December 31,2008.

Please find enclosed a Consent Agreement dated July 30, 2010, and signed by the Taxpayer on September 10, 2010. However, we note that the EINs for two of the entities subject to the Form 3 115 and enciosed Consent Agreement, American Water Engineering, Inc., and United Water Virginia, Inc., were incorrectly reflected in Appendix A to the Consent Agreement. In its information response to the IRS, by letter dated July 1, 2009, the Taxpayer provided the correct Ems of the two entities, American Water Engineering, Inc. (Ern: 76-0654501), and United Water Virginia, Inc. @IN: 54-1016694). The Taxpayer will be effecting the change permitted in the Consent Agreement.

If you have any questions, please call the Taxpayer's authorized representative, Robert Weiss, at 202-414-1421.

Sincerely,

Mark Chesla Vice President and Controller

Enclosures Executed Consent Agreement CONSENT AGREEMENT

Internal Revenue Service Department of the Treasury Washington, DC 20224

American Water Works Company, Inc. Person to Contact: and Subs. lnnessa Glazrnan P.O. Box 5600 Telephone Number: Cherry Hill, NJ 08003 (202) 622-7327 Refer Reply to: Attn: Mark N. Chesla CC:1TA:BO 1 CAM-I 0842 1-09 VP and Controller Employee Identification Number: 52-08393 EIN: 51-0063696 JUL 3 0 2010 In re: Application for Change of Accounting Method Form 31 15 - See Appendix A

Dear Mr. Chesla:

This letter refers to a Form 31 15, Application for Change in Accounting Method, filed by American Water Works Company, Inc. & Subs., ElN:51-0063696, on behalf of thirty applicants (see Appendix A) (collectively "the taxpayer"), requesting permission to change the taxpayer's method of accounting for: (I)costs to repair and maintain tangible property, and (2) dispositions of certain tangible depreciable property. The change is requested for the taxable period beginning January 1,2008 and ending December 31 , 2008 ("year of change").

The Department of the Treasury has published proposed regulations that clarify the application of §§ 162 and 263 of the lntemal Revenue Code to expenditures paid or incurred to repair, improve, or rehabilitate tangible property. Guidance Regarding Deduction and Capitalization of Expenditures Related to Tangible Property, 73 FR 12838- 01 (March 10,2008),2008-1 C.B. 871, A threshold issue in applying the rules under $j§162 and 263 is determining the appropriate unit of property to which the rules should be applied. The proposed regulations reserve the rules for determining the appropriate unit of property for network assets, which are defined as railroad track, oil and gas pipelines, water and sewage pipelines, power transmission and distribution lines, and telephone and cable lines. See Q 1263(a)-3(dX2)(iii)(C)(2) of the proposed regulations, 73 FR 12857. The preamble to the proposed regulations states that the unit of property for network assets should be addressed on an industry-by-industry basis in future Internal Revenue Bulletin guidance. See preamble discussion at 73 FR 12843.

Section 6.09 of Rev. Proc. 2010-1, 2010-1 I.R.B. I,16, provides that the Internal Revenue Service generally will not issue a letter ruling if the request presents an issue that cannot be readily resolved before a regulation or any other published guidance is issued, A letter ruling includes an Associate Office's response granting or denying a 2

American Water Works Company, Inc. & Subsidiaries CAM-108421 -09

request for a change in a taxpayer's accounting method. Section 2.01 of Rev. Proc. 2010-1. The unit of property determination for network assets is an issue that cannot be readily resolved before a regulation or other published guidance is issued. Further, because the taxpayer's proposed method of accounting is based on the unit of property determination, the propriety of the taxpayer's proposed method of accounting is also an issue that cannot be readily resotved. Thus, the Service declines to rule on whether the taxpayer's unit of property determination for its network asset is correct, and accordingly, whether its proposed method of accounting is a proper method of accounting.

Further, pursuant to section 4.02(1) of Rev. Proc. 2010-3, 2010-1 I.R.B. 110, 118, the Service will not ordinarily issue a letter ruling or determination letter on any matter in which the determination requested is primarily one of fact. The determination of the unit of property for dispositions of tangible depreciable property is a factual one. Thus, the Service declines to rule on whether the taxpayer is using the appropriate unit of property for determining dispositions of tangible depreciable property subject to its Form 31 15 and, accordingly, whether its proposed method of accounting for determining dispositions of such property is a proper method of accounting.

FACTS

The taxpayer is a corporation that is in the business of operating as public water and wastewater utility company that pumps, treats, and distributes water to and from residential, commercial, and industrial customers in the United States. The taxpayer uses an overall accrual method of accounting. Its principal business activity code is 221 300. The taxpayer is requesting permission to: (I) change its method of accounting for costs associated with the routine repair and maintenance of all of the taxpayer's network assets; and (2) change its units of property for determining dispositions of certain tangible depreciable property .

Routine repair and maintenance costs

The costs included in this request consist of costs associated with the routine repair and maintenance of taxpayer's tangible property. The taxpayer represents that these costs are incurred to keep the taxpayer's property in ordinarily efficient operating condition, and that they do not materially increase the value or substantially prolong the useful life of any unit of property compared to the value or useful life of the property before the general decline or event that led to the repairs or maintenance. The taxpayer represents that the repair and maintenance costs do not adapt any unit of property to a new or different use. The taxpayer represents that the repair and maintenance costs do not include costs to replace any unit of property or any major components or substantial 3

American Water Works Company, Inc. & Subsidiaries CAM-I 08421-09

structural parts of any unit of property. The taxpayer represents that the repair and maintenance costs are not incurred as part of a plan of rehabilitation, modernization, or improvement to any unit of property. The taxpayer represents that the repair and maintenance costs do not result from any prior owner's use of any unit of property.

Section 162 allows a deduction for all the ordinary and necessary expenses paid g during the taxable year in carrying on any trade or business.

Section 1.162-4 of the Income Tax Regulations allows a deduction for the cost of incidental repairs that neither materially add to the value of property nor appreciably prolong its useful life, but .keep it in an ordinarily efficient operating condition.

Under the taxpayer's present method of accounting for repair and maintenance costs, the taxpayer capitalizes the repair and maintenance costs described above and recovers these costs using the appropriate method over the applicable recovery period and the applicable convention as prescribed by §I68(a).

Under the taxpayer's proposed method of accounting for repair and maintenance costs, the taxpayer will treat the repair and maintenance costs as ordinary and necessary business expenses pursuant to §§ 162 and I.162-4.

Disposition of certain tangible depreciable property

The items of tangible depreciable property subject to the taxpayer's request to change its units of property for determining dispositions are described as network assets. Such property is depreciated by the taxpayer under § 168.

The taxpayer represents that:

I.None of the assets that are the subject of the taxpayer's Form 31 15 are leasehold improvements.

2. None of the assets subject to the taxpayer's Form 31 15 is subject to a general asset account election under § 168(i)(4) and the regulations thereunder.

3. None of the assets subject to the taxpayer's Form 31 15 is subject to a mass asset account election under former $j168(d)(Z)(A).

4. Depreciation for all of the assets subject to the taxpayer's Form 31 15 is not determined in accordance with § l.l67(a)-l I(regarding the Class Life Asset Depreciation Range System (ADR)). 4

American Water Works Company, Inc. & Subsidiaries CAM-108421-09

5. None of the assets subject to the taxpayer's Form 31 15 is subject to the repair allowance under 3 1.167(a)-I l(d)(2) (including expenditures incurred after December 31 , 1980, that were for the repair, maintenance, rehabilitation, or improvement of property placed in service by the taxpayer before January 1, 1981).

6. None of the assets subject to the taxpayer's Form 31 15 were disposed of in a transaction to which a nonrecognition section of the Code applies (for example, § 1031, transactions subject to 3 168(i)(7)).

7. There is no building (and its structural components) that is the subject of the taxpayer's Form 31 15.

Under the taxpayer's present method of accounting, the taxpayer uses a method other than the functional interdependence test to identify the unit of property for purposes of determining when a depreciable network asset is disposed of.

Under the taxpayer's proposed method of accounting, the taxpayer will use the functional interdependence test to identify the unit of property for purposes of determining when a depreciable network asset is disposed of. The taxpayer will use the same unit of property for purposes of determining when a depreciable network asset is placed in service (and when depreciation begins) and when the depreciable network asset is disposed of (and when depreciation ends).

The taxpayer has represented that, on the date the Form 31 15 was filed, it was not under examination and it was not before an appeals office or a federal court with respect to any income tax issue. See sections 3.07, 3.08(2) and 3.08(3) of Rev. Proc. 97-27, 1997-1 C.B. 680, as modified by Rev. Proc. 2002-19, 2002-1 C.B. 696.

SECTION 481 (a) ADJUSTMENT

The information provided indicates that, as of the beginning of the year of change, the required aggregate adjustment under 3 481(a) (the 3 481(a) adjustment) for the year of change is ($461,238,422). This amount represents a netting of the net negative 9 481(a) adjustment for maintenance and repairs with the net positive €j 481(a) adjustment for dispositions. The netting represents a one-time exception allowed the taxpayer for the year of change based on its particular situation. As a rule, the netting of the § 481(a) adjustment for maintenance and repairs with the § 481(a) adjustment for dispositions is not allowed under the provisions of Rev. Proc. 97-27. The €j481(a) adjustment for each applicant is shown in Appendix A. The net amount represents a decrease in computing taxable income. 5

American Water Works Company, Inc. & Subsidiaries CAM-I 08421 -09

CONSENTnERMS AND CONDITIONS OF CONSENT

Based solely on the facts presented and representations made, permission is hereby granted the taxpayer to change its method of accounting from the present method to the proposed method, beginning with the year of change, provided that:

The taxpayer takes the entire net 5 481(a) adjustment into account in computing taxable income in the year of change. See section 2.02(1) of Rev. Proc. 2002-19, 2002-1 C.B. 696, as amplified and clarified by Rev. Proc. 2002-54, 2002-2 C.B. 432.

The taxpayer keeps its books and records for the year of change and for subsequent taxable years (provided they are not closed on the date it receives this letter) on the method of accounting granted in this letter. This condition is considered satisfied if the taxpayer reconciles the results obtained under the method used in keeping its books and records and the method used for federal income tax purposes and maintains sufficient records to support such reconciliation; and

No portion of any net operating loss that is attributable to a negative § 481 (a) adjustment may be carried back to a taxable year prior to the year of change that is the subject of any pending or future criminal investigation or proceeding concerning (a) directly or indirectly, any issue relating to the taxpayer's federal tax liability, or (b) the possibility of false or fraudulent statements made by the taxpayer with respect to any issue relating to its federal tax liability. See section 5.02(4) of Rev. Proc. 97-27.

None of the items of property subject to the taxpayer's Form 31 15 is subject to a general asset account election under § 168(i)(4) and the regulations thereunder;

None of the items of property subject to the taxpayer's Form 31 15 is subject to a mass asset account election under former § 168(d)(Z)(A);

The taxpayer does not determine depreciation for any of the items of property subject to the taxpayer's Form 31 15 in accordance with § 1.167(a)-I 1 (regarding the Class Life Asset Depreciation Range System (Am); 6

American Water Works Company, Inc. & Subsidiaries CAM-I 08421-09

(7) None of the items of property subject to the taxpayer‘s Form 31 15 is subject to the repair allowance under § 1.167(a)-1 l(d)(2) (including expenditures incurred after December 31, 1980, for the repair, maintenance, rehabilitation, or improvement of property placed in service before January I, 1981);

8) None of the cost (or a portion thereof) of the assets subject to the taxpayer’s Form 31 15 is expensed or amortized under any provision of the Code, regulations, or other published guidance in the Internal Revenue Bulletin (for example, § 179D, § 14001); and,

9) If any item of property subject to the taxpayer’s Form 31 15 is public utility property within the meaning of § 168(i)(IO) or former 167(1)(3)(A):

(A) A normalization method of accounting (within the meaning of § 168(i)(9), former $$ 168(e)(3)(8), or former 3 167(1)(3)(G), as applicable) must be used for such public utility property;

B) As of the beginning of the year of change, the taxpayer must adjust its deferred tax reserve account or similar reserve account in the taxpayer’s regulatory books of account by the amount of the deferral of federal income tax liability associated with the § 481(a) adjustment applicable to such public utility property; and

C) Within 30 calendar days of filing the federal income tax return for the year of change or of receiving this letter ruling, whichever is later, the taxpayer must provide a copy of its Form 31 15 (and any additional information submitted to the Service in connection with such Form 31 15) to any regulatory body having jurisdiction over such public utility property.

EFFECT OF THIS ACCOUNTING METHOD CHANGE

The accounting method change granted in this letter is a letter ruling pursuant to 9 601.204(c) of the Statement of Procedural Rules. See also section 2.01 of Rev. Proc. 2010-1, 2010-1 I.R.B. at 6 (or any successor). The taxpayer ordinarily may rely on this letter ruling subject to the conditions and limitations described in Rev. Proc. 97-27.

However, the consent granted under this letter ruling for the taxpayer’s requested change is not a determination by the Commissioner that the taxpayer is using the appropriate unit of property for determining dispositions of tangible depreciable property and does not create any presumption that the proposed unit of property is permissible 7

American Water Works Company, Inc. & Subsidiaries CAM-I 08421-09

for such purposes. The director will ascertain whether the taxpayer’s determination of its unit of property for dispositions of tangible depreciable property is correct.

Further, the taxpayer should not infer approval of any tax treatment not specifically stated in this letter ruling. For example, this letter does not address the application of § 263A, which generally requires taxpayers to capitalize certain direct and indirect costs of property produced or acquired for resale, or the propriety of the taxpayer’s classification of property under § 168(e) or Rev. Proc. 87-56, 1987-2 C.B. 678. Further, this letter ruling does not imply approval of any tax treatment (including amounts that are part of the $481(a) adjustment) when the Code, the regulations, or other published guidance provides specific limitations and/or prohibitions. The Service expresses no opinion on the propriety of the unit(s) of property the taxpayer proposes to use in determining the deductibility of repair and maintenance costs. The unit of property determination is a factual one within the jurisdiction of the director.

The director must apply the ruling in determining the taxpayer’s liability unless the director recommends that the ruling should be modified or revoked, The director will ascertain whether (1) the representations upon which this ruling was based reflect an accurate statement of the material facts, (2) the change in method of accounting was implemented as proposed in accordance with the terms and conditions of the Consent Agreement and Rev. Proc. 97-27, (3) there has been any change in the material facts upon which the ruling was based during the period the method of accounting was used, (4) there has been any change in the applicable law during the period the method of accounting was used, (5) the amount of the § 481(a) adjustment was properly determined, and (6) the taxpayer’s determination of its unit of property is correct. In the case of (I),(Z), (3), or (4) above, if the director recommends that the ruling should be modified or revoked, the director will forward the matter to the national office for consideration before any further action is taken. Such a referral to the national office will be treated as a request for technical advice, and the provisions of Rev. Proc. 2010- 2, 2010-1 I.R.B. 90 (or.any successor) will be followed, & section 1I .01 of Rev. Proc. 97-27.

As noted above, the Department of the Treasury has published proposed regulations that clarify the application of fj§ 162 and 263 to expenditures paid or incurred to repair, improve, or rehabilitate tangible property. See Guidance Regarding Deduction and Capitalization of Expenditures Related to Tangible Property, 73 FR 12838-01 (March IO, 2008), 2008-1 C.B. 871. If final or temporary regulations are adopted with positions that are inconsistent with the method of accounting that the taxpayer implements in accordance with this letter ruling, the taxpayer will be required to follow any instructions in those final or temporary regulations concerning methods of accounting for the repair,

I improvement, or rehabilitation of tangible property for future taxable years, 8

American Water Works Company, Inc. & Subsidiaries CAM-108421 -09

AUDIT PROTECTION

An examining agent may not propose that the taxpayer change the same method of accounting as the method changed by the taxpayer under this ruling for a year prior to the year of change provided the taxpayer implements the change as proposed, in accordance with the terms and conditions of this ruling and Rev. Proc. 97-27, and the ruling is not modified or revoked retroactively because there has been a misstatement or an omission of material facts. See sections 9.01 and 9.02(1) of Rev. Proc. 97-27.

However, the Service may change the taxpayer's method of accounting for the same item for taxable years prior to the requested year of change if there is any pending or future criminal investigation or proceeding concerning (a) directly or indirectly, any issue relating to the taxpayer's federal tax liability for any taxable year prior to the year of change, or (b) the possibility of false or fraud.ulent statements made by the taxpayer with respect to any issue relating to its federal tax liability for any taxable year prior to the year of change. See section 9.02(4) of Rev. Proc. 97-27.

CONSENT AGREEMENT

If the taxpayer agrees to the terms and conditions set forth above, an individual with the authority to bind the taxpayer in such matters must sign and date the attached copy and return it within 45 days from the date of this letter to:

Internal Revenue Service Attention: lnnessa Glazman, CC:ITA:BOI P.O. Box 14095 Benjamin Franklin Station Washington, D.C. 20044

The signed copy constitutes an agreement regarding the terms and conditions under which the change is to be effected ("Consent Agreement") within the meaning of § 481(c) and as required by 5 1.481-4(b). The Consent Agreement shall be binding on both parties except that it will not be binding upon a showing of fraud, malfeasance, or misrepresentation of a material fact. In addition, a copy of the executed Consent Agreement must be attached to the taxpayer's federal income tax return for the year of change. For further instructions, see section 8.1 1 of Rev. Proc. 97-27. Alternatively, a taxpayer that files its returns electronically may satisfy this requirement by attaching a statement to its return that provides the date and control number of this letter ruling. 9

American Water Works Company, Inc. & Subsidiaries CAM-10842 1-09

The rulings contained in this letter are based upon information and representations submitted by the taxpayer and accompanied by a penalty of perjury statement executed by an appropriate party. While this office has not verified any of the material submitted in support of the request for rulings, it is subject to verification on examination.

The accounting method change granted in this letter is directed only to the taxpayer and may not be used or cited as precedent. See section 11.02 of Rev, Proc. 2010-1 , 2010-1 I.R.B. at 49. Final or temporary regulations under § 167 or $j168 pertaining to one or more of the issues addressed in this letter ruling have not yet been adopted. Therefore, if final or temporary regulations under § 167 or § 168 should be adopted with positions that are inconsistent with the conclusions reached in this letter ruling, the method of accounting utilized as a result of the letter ruling will no longer be regarded as a proper method of accounting and would be subject to change within the framework of §§ 446 and 481. 10

American Water Works Company, Inc. & Subsidiaries CAM-1 08421-09

In accordance with the provisions of a power of attorney currently on file, we are sending a copy of the ruling letter to your authorized representatives.

Sincerely yours,

J~HNP. MORIARTY Chief, Branch 1 Office of the Associate Chief Counsel (Income Tax and Accounting) cc: Internal Revenue Service Industry Director, LM:NRC Natural Resources and Construction 1919 Smith Street, Stop 1OOOHOU Houston, TX 77083

Robert Weiss PricewaterhouseCoopers LLP 1301 K Street, NW, Ste 800W Washington, DC 20005

Gwynneth H. Stott, CPA PricewaterhouseCoopers LLP 2001 Market Street, Ste 1700 Philadelphia, PA 19103

/o ,& Signed this day

(Name and corporate title of parent officer)

West Virginia American Water Company Case Nos . 18-0573-Mi-42T and 18-0576-S-42T Rebuttal Testimony of Anne E . Bulkley Page 2 of 46

TABLE OF CONTENT$

I . WITNESS IDENTIFICATI N AND QUALIFICATI NS ...... 3 I1 . SUNI~ARY VERVIEW ...... 5 I11 . NSE TO STAFF WITNESS MS . SHANIBLIN ...... 10 ...... 12 CF A~ALYSES...... 13 NALYSES ...... 18 ...... 31 E . RECENTLY AUTHORIZE^ ROES ...... 33 ...... 38 ...... 38 N ...... 41 ON ...... West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 3 of 46

ST

1

2

3 A. My name is Ann E. Bulkley. I am employed by Concentric Energy Advisors, Inc.

4 (“Concentric”) as a Senior Vice President. My business address is 293 Boston

5 Post Road West, Suite 500, Marlborough, Massachusetts 01 752.

6

7 A. I am submitting this Rebuttal Testimony on behalf of West-Virginia American

8 Water Company (“WVA~C’’or the “Company”), a wholly-owned subsidiary of

9 American Water (‘bA%7K’‘).

10

11 A. Yes. My Direct Testimony was included as part of the Company’s case-in-chief

12 that was filed on April 30,2018.

13

14 A. I am responding to the testimony presented by Ms. Kaitlyn Shamblin on behalf of

15 the Staff of the Public Service Commission of West Virginia (“Staff’) as it relates

16 to the just and reasonable Return On Equity (“ROE”) and the appropriate capital

17 structure for WVA~C. Specifically, I will respond to the methodologies

18 developed by Ms. Shamblin and her final recommendations. Rather than

19 providing a response to every point of disa~eemen~,my response identifies the West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 4 of 46

1 most significant differences between our respective analyses and

2 recommendations. However, the fact that there may be issues and statements

3 raised by Ms. Shamblin that have not been addressed in this Rebuttal Testimony

4 does not constitute agreement with her on those matters. I also respond to the

5 weighted average cost of capital relied on by Consumer Advocate Division’s

6 witness Mr. Ralph Smith in his calculation of the revenue requirement. In addition

7 I respond to Consumer Advocate Division‘s witness Mr. Scott Rubin regarding his

8 proposed adjustment to the ROE for the implementation of decoupling.

9

10 A. Yes, Schedules AEB-14 through AEB-16 were prepared by me or under my

11 direction.

12 ow is the rema uttal Testimony organize

13 A. The remainder of my Rebuttal Testimony is organized as follows:

14 In Section TI: I provide a summary and overview of my Rebuttal Testimony

15 and the important factors to be considered in establishing the ROE for

16 WV AWC .

17 In Section 111, I respond to Staff witness Ms. Shamblin’s analyses and

18 recoimendations.

19 In Section IV, I provide a discussion of recent capital market conditions.

20 In Section V, I respond to CAD witness Ralph Smith’s cost of capital used

21 in the revenue requirements calculation West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-4271 Rebuttal Testimony of Anne E. Bulkley Page 5 of 46

I In Section VI, I summarize my conclusions and recommendations.

2

3 hat are your ey conclusions and r~co~~endatio~sregarding the

4 ita1 structure €or C?

5 A. My key conclusions are as follows:

6 1) Ms. Shamblin contends that WVAWC‘s cost of equity is only 9.27

7 percent. Ms. Shamblin has not reconciled her recommendation with the

8 fact that the Commission‘s currently authorized ROE for ~VAWCis

9 9.75 which was authorized at a time of lower interest rates and investor

10 expectations. Ms. Sharnblin is proposing to reduce that equity retuni by

11 48 basis points, despite her recognition that short-term interest rates have

12 been increasing. Over the past year, interest rates have increased

13 significantly, with the yield on the 10-year Treasury increasing more than

14 60 basis points since December 2017. Investors expect that trend to

15 continue into 20 I9 and forward. In addition, utilities are facing declining

16 cash flow metrics resulting from the TCJA. Credit rating agencies have

17 reported concerns about these metrics and are looking to regulators and

18 utility management to implement solutioiis to stabilize financial metrics,

19 citing increasing ROES or equity ratios as possible solutions. Moody’s

20 downgraded its outlook on AWK in January 2018 and on the entire

21 utilities sector in June 2018 related to tax reform. Ms. Shamblin’s

22 proposal that the Conipany’s return be decreased by 48 basis points is West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 6 of 46

unsupported and unconstructive given current market conditions and

would exacerbate the financial community’s concerns about cash flow

metrics.

4 2) Ms. Shamblin is also proposing an ROE of 9.27 percent when the median

5 recent ROE for water companies authorized across the nation is at least

6 9.6%. Given WVAWC’s additional risk factors as compared with the

7 proxy group, including its small size, declining use and regulatory lag, it

8 is not reasonable for the Company to have an authorized ROE lower than

9 the average for water distribution companies across the nation, especially

10 in a rising interest rate environment. Consideration of these risk factors

11 supports an ROE above the average authorized return for water

12 companies in other jurisdictions.

13 3) While Ms. Shamblin establishes an outlier test in her Constant Growth

14 DCF analysis, she establishes the low-end screen at the cost of debt,

15 which is not reflective of the premium for the additional risk that is

16 required to hold equity as opposed to debt. This unreasonably low floor

17 results in a significant understatement of the cost of equity using the

18 Constant Growth DCF model. Conducting a DCF analysis using all of

19 the relevant projected EPS growth rates, rather than using the screening

20 criteria relied on by Ms. Shamblin establishes mean DCF results of 9.09

21 percent and a high DCF result of 10.79 percent. Taking into

22 consideration a forward-looking market risk premium in the CAPM West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 7 of 46

using Ms. Shamblin's various assumptions for the risk-free rate

' establishes a range of returns of 11.48 percent to 11.73 percent. Weighing

these DCF and CAPM results equally, as Ms. Shamblin has done in her

Direct Testimony results in a range of the cost of equity of 9.77 percent

to 1 1.26 percent with a mean result of 10.47 percent. My recommended

ROE of 10.80 percent is between the mean and high result, whicli is

appropriate based on the company-specific risk factors that are discussed

in my Direct Testimony that differentiate ~AWCfrom the proxy

10 4) Ms. Shamblin's proposed ROE and capital structure, taken together are

11 not consistent with more realistic and constructive regulation seen in

12 recent authorized ROES in other jurisdictions. Regulators across the

13 country have been addressing this issue in authorized returns and equity

14 ratios. While some of the financial models used to estimate the cost of

15 equity produce results between 8.00 and 9.00 percent or below, utility

16 regulators across the nation have recognized that such low returns are

17 unrealistic and not compensatory for investors. The recently authorized

18 ROEs for water utilities demonstrate that the range of authorized ROEs

19 for water distribution companies has been from 9.20 percent to 10.50

20 percent in 2018. The corresponding equity ratios have been in a range

21 from 51.89 percent to 57.00 percent. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-5-42T Rebuttal Testimony of Anne E. Bulkley Page 8 of 46

1 5) Ms. Shamblin’s recornmended ROE and equity ratio do not meet the

2 Hope and Bluefield Standards. While Ms. Shamblin provides financial

3 metria, she does not address the comparable return standard. As

4 discussed in my Direct Testimony, the comparable return standard

5 established in these landmark cases require that the return that is

6 established for a utility be commensurate with the return on other

7 investments of similar risk. Despite rising interest rates, the equity return

8 (equity ratio x ROE) that results from Ms. Shamblin‘s recommended

9 ROE of 9.27 percent and her proposed equity ratio of 45.59 percent is

10 lower than the equity rate that was authorized for the Company in its last

11 rate proceeding. Ms. Shamblin provides no explanation as to why this is

12 appropriate.

13 In addition, the equity rate that results from Ms. Shamblin’s proposal is

14 below the equity return that has been authorized for any water utility in

15 2018.l Ms. Shamblin has provided no explanation as to why this is

16 supportable for a WVAWC which is smaller in size and is exposed to

17 significant risk factors.

18 6) In contrast to Ms. Shamblin’s proposal, the range and recommended

19 ROE presented in my Direct Testimony is supported by the analyses

20 presented in my Direct Testimony and the revisions to Ms. Shamblin’s

- Direct Testimony of Kaitlyn J. Shamblin, at 4. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 9 of 46

1 analyses presented in my Rebuttal Testimony. Furthermore, the equity

2 rate that is established by my recommended ROE and equity ratio is

within the range established by recently authorized retunis and equity

ratios. My recommendation is consistent with the expectation of a

continued rising interest rate environment and also provides a reasonable

response to credit rating agency concerns regarding cash flow coverage

ratios. Finally, my recommendation appropriately considers the

additional risks faced by a smaller water utility.

9 7) Mr. Smith‘s revenue requirement report is developed using two scenarios

10 for the ROE of 8.75 percent and 9.75 percent. r. Smith provides no

11 analysis to support the development of these scenarios. Based on the

12 results of the DCF and CAPM methodologies that are presented in my

13 Direct Testimony, the analyses presented in my response to Ms.

14 Shamblin, and recently authorized ROES, Mr. Smith’s unsubstantiated

15 recommended ROE of 8.75 percent is more than 100 basis points below

16 the low end of the range of results. Furthermore, Mr. Smith‘s 9.75

17 percent ROE - also relied on without any analytical foundation - is at the

18 low end of the range and is unreasonable given the rising interest rate

19 environment and other financial risk factors discussed in my Direct and

20 Rebuttal Testimonies.

21 8) Mr. Smith’s use of the historical capital structure for WVAWC is not

22 appropriate because it does not reflect the capital structure that will be West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-3-42T Rebuttal Testimony of Anne E. Bulkley Page 10 of 46

1 used to finance the Company’s rate base over the Addendum Period. The

2 Company’s proposed capital structure is based on the test year actual

3 capital structure as of December 3 1, 201 9, adjusted for financing that is

4 projected in the Addendum Period. Therefore, the Company’s proposed

5 capital structure is representative of the actual capitalization of the

6 company over the rate period and should be used in setting the revenue

7 requirement.

8 9) Mr. Rubin’s suggestion that the implementation of an RSM necessitates

9 a reduction in the ROE is unsubstantiated, is inconsistent with the

10 methodology used by regulatory cornmissions to set the ROE and is

11 contrary to recent analysis on the overall risk and return requirements for

12 utilities.

13

14

15 reco~~end~tio~s.

16 A. Ms. Shamblin recommends an ROE for ~AWCof 9.27 percent, by weighing

17 equally the results of one Constant Growth Discounted Cash Flow (“DCF’’)

18 analysis and the results of one scenario of her Capital Asset Pricing Model

19 (“CAPM”) analysis2 Ms. Shamblin adjusts the Conipany‘s capital structure

20 proposal based on the Company’s December 31, 2017 test year-end capital

Direct Testimony of Kaitlyn J. Shamblin, at 7 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 11 of 46

1 structure and the average daily balances of short-term debt over the test period.

2 Based on this analysis, Staff recommends a capital structure that consists of 5.03

3 percent short term debt, 49.29 percent long-term debt, 0.09 percent preferred stock

4 and 45.59 percent eq~ity.~Ms. Shamblin calculates various coverage ratios as a

5 test of the reasonableness of her overall recommendation.

6 amblin’s conclusions?

7 A. No, I do not. As is discussed in the remainder of my Rebuttal Testimony, Ms.

8 Shamblin disregards some of her own analyses that demonstrate the ROE is higher

9 than her recommended ROE. In fact, reasonable changes to Ms. Shamblin’s DCF

10 analyses indicate a range of returns that is consistent with the recently

11 authorized ROES for water utilities in other regulatory jurisdictions and

12 demonstrate that Ms. Shamblin’s estimate of the cost of equity is insufficient.

13 Furthermore, the analyses presented in my Direct and Rebuttal testimonies

14 demonstrate that Ms. Shamblin‘s recommended capital structure does not reflect

15 the financing requirements of the Company over the Addendum Period and is

16 below the mean capital structure of the proxy companies. Taken together, Ms.

17 Shamblin’s proposed equity return (cost of equity x equity ratio) is significantly

18 below anything that has been authorized for a water utility in 201 8. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 12 of 46

1 roxy Group

2 Q. Please sum~~ar~~the proxy groups refie

3 A. Ms. Shamblin does not apply any screening criteria to determine the comparability

4 of companies to WVAWC. Rather, she relies on the entire Value Line universe of

5 water utilities which includes nine companies.

6 oes this group differ fro the proxy group you relied on in your Direct

7 Testimony?

8 A. As discussed in my direct testimony, I relied on several screening criteria that were

9 applied to establish a proxy group of comparable risk to WVAWC. My final proxy

10 group did not include Consolidated Water Company. Consolidated Water

11 Company develops and operates seawater desalinization projects and water

12 distribution systems in areas where potable water is scarce. These areas include

13 the Cayman Islands, the Bahamas, Belize, the British Virgin Islands, and BaL4

14 These operations have significantly different operating risk than WVAWC and

15 therefore Consolidated Water Company should be excluded from any proxy group

16 of comparable risk.

' Value Line Report on Consolidated Water Company, July 13,2018. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page I3 of 46

1

2

3 A. Ms. Shamblin develops a Constant Growth DCF analysis that relies on historical

4 and projected dividend per share (“DPS”) and historical and projected earnings per

5 share (“EPS”) growth rates. Ms. Shamblin eliminates growth rates from her

6 analysis using two screening criteria: 1) DCF results that are less than the

7 Company‘s most recently issued debt, which was 4.20 percent and 2) DCF results

8 that are greater or less than the average DCF result by 300 basis points.

9 As shown on Schedule 2, sheet 6 of Ms. Shamblin’s exhibits: the growth rate that

10 she relies on in her DCF analysis (shown on Schedule 2, sheet I) is the average of

I1 all growth rates that are not been eliminated by any of the above described

12 screening criteria. This average includes historical and projected EPS and DPS

13 growth rates for the proxy group companies.

14 rowth rates relie n in t

15 odd?

16 A. No. I disagree with the use of historical average growth in DPS and EPS. In

17 addition, I believe that projected EPS growth rates are the more appropriate growth

18 rates to use in the DCF analysis. The estimation of the ROE is a forward-looking

19 concept. Therefore, the use of projected data is more appropriate than backward-

20 looking historical data. In addition, historical growth rates are reviewed by West Virginia American Water Company Case Nos. 18-0573-W-427"and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 14 of 46

1 investors and incorporated in future growth rate projections, therefore this data is

2 already included in analysts' forward-looking growth projections.

3 Regarding the use of DPS and EPS growth rates, EPS growth rate projections are

4 more appropriately used in the DCF model. Earnings growth rates tend to be least

5 influenced by capital allocation decisions that companies may make in response

6 to near-term changes in the business environment. Since such decisions may

7 directly affect near-term dividend payout ratios, estimates of earnings growth are

8 more indicative of long-term investor expectations than are dividend growth

9 estimates. Furthermore, over the long run, dividend growth can only be sustained

10 by earnings growth. Therefore, as discussed in my Direct Testimony, projected

11 EPS growth rates are more appropriate for use in the DCF analysis.

12

13 A. No, I do not. Ms. Shamblin develops eighteen DCF analyses that establish a range

14 of returns that might be reasonable to consider in setting the recommended ROE

15 for ~VAWC.However, Ms. Shamblin then disregards this range of DCF results,

16 using these return estimates as an interim step to eliminate reported historical EPS

17 and DPS growth observations and the projected EPS and DPS estimates. Her final

18 recommendation relies on the mean result from one DCF analysis, averaged with

19 the result of one CAPM analysis. The selection criteria relied on by Ms. Shamblin

20 to eliminate growth rates used in the DCF model ignore the fundamental

21 expectations of equity investors. Ms. Sharnblin also incorrectly establishes a floor

22 on the values that result from the DCF model at the cost of the Company's most West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testii~onyof Anne E. ~ulkley Parze 15 of 46

recent debt issuance. In applying this floor, Ms. Sharnblin is implicitly stating that

2 there is no equity risk premium and that an equity investor would find the cost of

3 debt an appropriate return for the risk of holding equity. This contradicts financial

4 theory because equity holders are the last claimant in the event of bankruptcy,

5 behind holders of a company’s debt. It is unreasonable to expect that equity

6 investors would not require a premium above the cost of debt to compensate for

7 that incremental risk. Setting the floor of equity returns unrealistically low, at the

8 cost of debt, as Ms. Shamblin does, serves to reduce the average growth rate used

9 in the DCF model and understates the cost of equity.

10 01%’ w ’S yses?

11 A. As shown in Schedule AEB-14, using Ms. Shamblin’s data for the proxy group I

12 made the following four revisions to the analysis:

13 1) 1 excluded Consolidated Water Company from the proxy group for the

14 reasons discussed previously.

15 2) 1 calculated the ROE for each of the proxy group companies relying on

16 EPS growth projections reported by Value Line, Zacks and

17 Yalioo!Finance.

18 3) I revised Ms. Shamblin’s DCF floor to reflect a reasonable equity risk

19 premium, setting the floor at 7.00 percent as discussed in my Direct

20 Testimony. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 16 of 46

1 4) I calculated additional DCF scenarios using low mean and high EPS

2 growth rates for all proxy companies.

3 CF analysis is corrected for the errors you i

4 what range of results is produced?

5 A. As shown in Schedule AEB-14, the range of DCF results is between 8.06 percent

6 and 10.79 percent.

7 ore ~sefulthan a point esti~atein deter~ningthe ap

8 E for a Co~pany?

9 A. Yes. While the proxy group has been screened to establish a reasonably

10 comparable group, the screening criteria need to remain sufficiently broad to

11 produce a reasonable sized proxy group. Therefore, these criteria cannot identify

12 all of the risk factors of an individual subject company. Developing a single point

13 estimate for the DCF results, as Ms. Shamblin has done does not consider the full

14 range of investor expectations returns on the proxy group. In contrast, when the

15 range of published analysts' projections are considered in the DCF model, there is

16 a more meaningful range of returns that is developed which provides for the ability

17 to appropriately consider the factors that differentiate the subject company fi-om

18 the proxy group that could not be included in the screening criteria. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 17 of 46

1 E fo

2 C?

3 A. No, it does not. In my Direct Testimony, I discuss the effect of current market

4 conditions on the assumptions used in the ROE estimation models. As discussed

5 in my Direct Testimony, many regulatory conimissions have recognized that

6 market conditions have affected the ROE estimation models and have made

7 modificatiotis to the authorized ROES to adjust for those conditions. Furthermore,

8 as will be discussed later in my Rebuttal Testimony, recently authorized ROES

9 demonstrate that 8.06 percent is insufficient to meet investor expectations based

10 on the range of recently authorized ROES for water distribution companies.

11 s. s

12 A. Although Ms. Shamblin appropriately developed multiple DCF scenarios, she then

13 disregarded the results of those scenarios. Consequently, Ms. Shamblin’s

14 methodology is inherently circular, relying on CF results to screen out growth

15 rates that produce high and low DCF results only to use the average of the

16 remaining growth rates in her final DCF model. This methodology is an

17 unconventional approach to the developing the DCF model assumptions and

18 almost by design, narrows the range of likely returns produced by the DCF model.

19 The resulting analysis is a single estimate of the cost of equity that is lower than

20 the Company‘s currently authorized ROE and is at the lower end of the range of

21 recently authorized ROES. Furthermore, Ms. Shamblin’s DCF result is West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Aime E. Bulkley Page 18 of 46

1 inconsistent with her own recognition that interest rates have increased since

2 several of the recent cases that she cites in her Direct Testimony,

As discussed previously, by relying on all the reported analysts’ EPS growth rate

projections, the DCF analysis reflects a much broader range of results, as high as

10.79 percent, which is consistent with my recommended ROE in this case.

6. Response to s. Shamblin’s CAP

. Please su~~ar~es. Sha~b~ndevelope

A. Ms. Shainblin develops several estimates of the ROE using the CAPM,

summarized in Figure 1 below.

10

Current T-Bill Rf rate 2.39% 8.70% 9.1 1%

Historical Rfrate 3.40% 8.70% 10.12%

Projected Rf rate 3.23% 6.10% 7.94%

11

12 Ms. Shamblin bases her CAPM on three measures of the risk-free rate; the current

13 Treasury Bill rate, the projected yield on long-term Treasury bonds and a historical

14 yield on the 30-year Treasury bond. Ms. Shamblin recognizes that although the

15 Commission had once expressed a preference for the use of short-term U.S.

16 Treasury Bill rates as the risk-fkee rate in the CAPM, the Coinmission abandoned West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testiiiiony of Anne E. Bulkley Page 19 of 46

1 that practice in recent cases, determining that the rates on Treasury Bills were too

2 low to be reasonable' and adopting, instead, the historical return on long-term

3 Treasury Bonds as the measure of the risk-free rate.6 As shown in Schedule 3 of

4 her exhibits, based on these assumptions, her CAPM results range from 7.94

5 percent to 10.12 percent.

6 lin use the results of her CA ~alysisin her final

7

8 A. Despite the Commission's determinations in recent decisions acknowledging that

9 the interest rates on the U.S. Treasury Bills were too low to be reasonable, Ms.

10 Shamblin disregards the use of the projected yield on long-term Treasury Bonds

11 as a measure of the risk-free rate and relies exclusively on the CAPM analysis she

12 developed using the projected Treasury Bill rates. Ms. Shamblin averages the 9.1 1

13 percent ROE that results fi-om that CAPM scenario with the 9.42 percent result

14 CF analysis to establish her recommended ROE of 9.27 percent. Ms.

15 Shamblin concedes, however, that if she had relied on the 10.12 percent CAPM

16 result based on the historical Treasury bond yields weighted equally with her DCF

17 result, the ROE estimate would be 9.77 per~ent.~

Direct Testimony of Kaitlyn J. Shamblin, at 13 referencing Commission Orders in Case Nos. 11-1627- G-42T (~ountain~erGas Company) and 10-0920-W-42T (West Virginia-American Water Company) ' Direct Testimony of Kaitlyn J. Shamblin, at 13 referencing Commission Orders in Case Nos. 11-1627- G-42T. Appendix KJS-1. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 20 of 46

1 o YOU agree with a~blin’sappli~atio

2 A. No, I do not. Specifically, I do not agree with Ms. Shamblin’s the use of the

3 projected rate for the U.S. Treasury Bill as the risk-fiee rate and the Market Risk

4 Premium (“MW”) estimates that she relies on in her analysis.

5 lain why you isagree with the risk-free rate that

6 relies on.

7 A. It is a well-established convention to rely on long-term Treasury bond yields as a

8 measure of the risk-free rate, in particular the yield on 30-year Treasury bonds.*

9 In addition, financial theory, specifically Modigliani-~iller,generally assumes

10 that government debt is risk-free or that the risk is completely diversifiable.

11 Therefore, it is reasonable to rely on the yield on 30-year debt as the risk-fiee rate

12 not on short term Treasury Bills.9

13 Furthermore, while Ms. Shamblin relies on the rate on U.S. Treasury Bills as the

14 risk-fiee rate in her CAPM, she recognizes that in recent cases the Commission

15 has determined that the rate on these instruments is too low to be considered a

16 reasonable measure of the risk-free rate “due to the unprecedented economic and

17 financial crisis that rendered significant unexpected results in some of the

18 measures of the cost of capital that the Commission has used in the past.’O

* Public Utility Reports, Inc, New Reaulatow Finance, Morin, Roger, 2006, at 151. Ibid. Io Direct Testimony of Kaitlyn J. Shamblin, at 12 citing to the West Virginia Commission Order in Case NO, 10-0920-W-42T. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Parre 21 of 46

1 o the

2 to use the rate on U.S. Treasury e risk-free rate?

3 A. No. As noted in Ms. Shamblin’s Direct Testimony, the Commission has

4 recognized that unprecedented market conditions have affected the interest rate

5 assumptions that can be used in the CAPM. In those cases, the Commission

6 demonstrated reasoned judgement in its selection of the appropriate risk-free rate

7 to use in the ROE estimation models. In its decision in the Company’s 2012 case,

8 the Commission clearly reasoned that the historical Treasury Bill rates were

9 unreasonable to be considered in the CAPM in that case. Furthermore, in Case

10 No. 1 1-1627-6-42T, when the long-term Treasury Bond yield was at 2.75 percent,

11 the Commission also noted that the use of that yield in the CAPM may have

12 skewed the results of the CAPM and relied instead on a risk-fi-ee rate of 6.10

13 percent. Therefore, rather than providing unwavering support for the use of a

14 specific proxy for the risk-free rate, the Commission has demonstrated that it

15 recognizes the importance of considering how market conditions have affected the

16 assumptions used in the CAPM and whether the use of the assumptions provide

17 meaningful results for the cost of equity. In doing so, the Commission has

18 determined that the Treasury Bill rate is not useful in the CAPM based on recent

19 market conditions. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 22 of 46

1 s. Sha~blin’sanalysis o historical rates on U.S. Treasury

2 bills support her use of these instru~entsas t e risk-free rate?

3 A. No, it does not. Ms. Shamblin notes that the long-term average historical return

4 on U.S. Treasury Bills has been 3.40 percent over the period from 1927 through

5 2017. The current rate that she has relied on is 2.39 percent, which is still 100

6 basis points below the historical average rate on U.S. Treasury Bills. Therefore,

7 the current rate is still well below market levels prior to the financial market

8 collapse and the conditions that the Commission noted in Case No. 10-0920-W-

9 42T when it made a specific decision to discontinue the use of the Treasury Bill

10 rates as the measure of the risk-ti-ee rate.

11 riate r~sk-freerate to

12 ?

13 A. Based on financial theory and well accepted convention, I believe it is reasonable

14 and appropriate to rely on the yield on the 30-year Treasury Bond as a measure of

15 the risk-free rate in the CAPM. Financial theory and common practice are

16 supported by the Commission’s recognition of recent unprecedented market

17 conditions and its reasoned response to those conditions in the selection of the 30-

18 year Treasury bond yield as the risk-free rate in the CAPM. I. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 23 of 46

1 as

2 re of the risk-free rate?

3 A. Yes. Ms. Shamblin developed an estimate of the risk-free rate by averaging the

4 projected yield on the 30-year Treasury Bond as published in the Blue Chip

5 Financial Forecast, projections published by Value Line in The Quarterly

6 Econoinic Review and the 13-week historical average rate on long-term term

7 bonds reported by the Federal Reserve.'* The result of this analysis, which relies

8 on a historical MRP, was an ROE of 7.94 percent.

9 terest rate enviro~~~e

10 A. No, it does not. Ms. Shamblin's application of these assumptions results in a return

11 that is lower than any return authorized by any regulatory jurisdiction for a water

12 utility demonstrating that the results of this analysis are unreasonable and do not

13 reflect the current cost of equity.

14

15

16 A. Ms. Shainblin develops two historical measures of the market risk premium using

17 data fiom the Duff & Phelps publication; Stocks, Bonds Bills and Inflation

18 Valuation Edition 201 8 Yearbook. The first measure of the market risk premium

19 is calculated as the difference between the arithetic mean return on Large

20 Company Stocks and U.S. Treasury Bills for the period from 1929 through 2017.

l1 Direct Testimony of Kaitlyn J. Shamblin, at 13. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 24 of 46

This calculation results in an MRP of 8.70 percent.12 Ms. Shamblin’s second

measure of the historical MRP is the difference between the return on Large

Company Stocks and the historical return on long-term govemiient bonds as the

risk-free rate, resulting in an MRP of 6. 10 percent.

o you agree with s. Sha~b~in’$ca~cu~ation$ of the M

A. No. 1 disagree with the use of a historical MRP to set a forward-looking ROE. The

MRP is dynamic, not static, meaning that the MRP changes over time as economic

8 and capital market conditions change. Academic research has shown that there is

9 an inverse relationship between interest rates and the equity risk premium. That

10 is, as interest rates decrease, the MRP increases, and vice versa. Ms. Shamblin’s

11 historical MW fails to reflect the inverse relationship between interest rates and

12 market risk premia. It is not reasonable to rely on the historical MRP in the current

13 low interest rate environment because the historical average does not reflect

14 current market conditions.

1s In addition, Ms. Shamblin’s calculation of the MRP using historical data is

16 incorrect. Stocks, Bonds Bills and Inflation Valuation Edition 201 8 Yearbook

17 RP should be calculated as the difference between the historical

18 return on Large Company Stocks and the income only return on government bonds.

19 Therefore, using Ms. Shamblin’s calculation of the MRP based on the historical

Id., at 11. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 25 of 46

1 return on long-term government bonds, the MRP would be 7.10, which is 100 basis

2 points higber than the MRP that Ms. Shamblin relied on.

3 . Why do you disagree wit

4 A. In addition to the fact that the return that is being estimated is fonvard-looking and

5 that the historical return does not reflect the inverse relationship between the MRP

6 and interest rates, the calculation of the historical MRP has produced results that

7 are not consistent with investor sentiment and current conditions in capital

8 markets. For example, Morningstar observes:

9 It is important to note that the expected equity risk premium, as it is 10 used in discount rates and the cost of capital analysis, is a fonvard- 11 looking concept. That is, the equity risk premium that is used in the 12 discount rate should be reflective of what investors think the risk 13 premium will be going forward.

14 Using historical data, Figure 2 demonstrates why the historical data that Ms.

15 Shamblin relied on should not be used to develop the MRP. Figure 2 summarizes

16 the MRP that would be estimated using the corrected estimate of the MRP

17 discussed previously; the difference between the return on Large Company stocks

18 and the income only return on government bonds for the 2007-2009 time-period,

19 which marks the most significant decline in the stock market since the Great

20 Depression.

l3 Morningstar Inc., 2010 Ibbotsoii Stocks, Bonds, Bills, and Inflation, Valuation Yearbook, at 55. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 26 of 46

1 As shown in Figure 2, using the data that Ms. Shamblin relied on would suggest

2 that 2007-2009 the historical MRP decreased even as market volatility (the

3 primary statistical measure of risk) over this time-period significantly increased.

4 arket Risk Premium and arket Volatili~

arket Volatili~

5

6 The assumption that investors would expect or require a lower risk premium

7 during periods of increased volatility is counter-intuitive and leads to unreliable

8 analytical results.

9 The relevant objective in the application of the CAPM is to ensure that all three

10 components of the model (Le., the risk-free rate, Beta, and the MRP) are consistent

11 with market conditions and investor perceptions. Assuming a lower MRP during

12 periods when interest rates are artificially suppressed by Federal Reserve monetary

13 policy is at odds with that premise.

l4 Morningstar Inc., 2010 Ibbotson Stocks, Bonds, Bills, and Inflation, Valuation Yearbook, at 23, 2009 Ibbotson SBBI Valuation Yearbook at 23, and 2008 Ibbotson Valuation Yearbook, at 28. Historical Market Risk Premium equals total return on large company stocks less income only return on long-term government securities. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 27 of 46

1

2 A. The ROE that is being set in this case is intended to be forward-looking. Therefore,

3 it is appropriate that the ROE estimation models that are relied on reflect those

4 forward-looking market conditions. As discussed above, Ms. Shamblin's CAPM

5 analysis using historical measures of the MRP is not reflective of expected market

6 conditions and therefore should be disregarded. Furthermore, for the reasons

7 discussed above, I continue to support the use of the projected yield on the long-

8 term Treasury bond as the appropriate estimate of the risk-free rate.

9 re

10

I1 A. As shown in Schedule AEB-15, I updated Ms. Shamblin's analysis that was

12 presented in Schedule 3 her Direct Testimony. I excluded Consolidated Water

13 Company from the analysis for the reasons discussed previously in my Rebuttal

14 Testimony. 1 updated s. Shamblin's CAPM analysis to rely on the MRP used in

15 my Direct Testimony and the three estimates of the risk-free rate that Ms.

16 Shamblin relied oil. As shown in that schedule, the results of the CAPM using a

17 projected MW range from 1 1.48 to I 1.73 percent.

18 et

19

20 A. Yes, Figure 3 summarizes estimates of the S&P 500 return calculated using

21 earnings growth projections from Bloomberg Professional, Yahoo!Finance, and West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-3-42T Rebuttal Testimony of Anne E. Bulkley Page 28 of 46

1 Standards and Poor’s. The calculated returns for the S&P 500 range from 12.89

2 percent (Yahoo!Finance) to 16.85 percent (Bloomberg Professional). These data

3 support the total return for the S&P 500 Index of 14.51 percent that I used to

4 determine the forward-looking market risk premium in my CAPM analysis.

5

Yahoo!Finance August 31,2018 1.80% 1 1.OO% 12.89%

Standard and Poor’s August 23,201 8 1.80% 13.90% 15.82%

6

7 e results of the cQrrecte

8 a~blin’sa~alysis.

9 A. As shown in Figure 4 below, the DCF results range from 8.06 percent to 10.79

10 percent and the CAPM results range from 11.48 percent to 11.69 percent.

11 Weighing equally the DCF results with each of the CAPM results establishes a

12 range of returns from 9.77 percent to 1 1.26 percent.

Is Bloomberg and Yahoo!Finance do not report a dividend yield for the S&P 500; therefore, the most recent 12-month average dividend yield as of July 2018 reported in the August 23, 2018 S&P 500 Earnings and Estimate Report was used to calculate the total return. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 29 of 46

1

CAP

Weighted ROE Estimates I I/ li I 9.77% I 9.87% 1 9.89% I Mean DCF (9.09%) and CAPM 10.29% 10.39% 10.41% 1 Mean High DCF (1 0.79%) and CAPM I 11.13% 11.24% 11.26% 2

3 you c

4 A. As discussed previously, the Mean Low DCF results are below any authorized

5 ROE that has been authorized for any water utility and therefore are not reasonable

6 representations of the cost of equity. The weighted ROE results using the Mean

7 and Mean High growth rates in the DCF model result in returns that are within the

8 range established in my irect Testimony. Furthermore, considering market

9 conditions and the Company’s small size the results of this analysis support my

10 recommended ROE of 10.80 percent.

11

12

13 A. While Ms. Shamblin did not propose an adjustment in her final recommended

14 ROE, she suggested that “[Eilf Staff were to adjust the ~o~ii~issionapproved rate West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 30 of 46

1 to reflect a reduction in risk due to the Company utilizing a Distribution System

2 Improvement Charge (DSIC), the ROE would be 25 basis points below the

3 Commission approved rate or 9.5 1%."16

4 Do you agree with s. Shamblin that the use of a SIC should result in a

5 ustment to the authorize E?

6 A. No, I do not. Ms. Shamblin offers no support for the suggested 25 basis point

7 reduction in the ROE. The ROE analyses that Ms. Sharnblin and I have conducted

8 are based on proxy groups of risk comparable companies. Therefore, to determine

9 that the DSIC was risk reducing, as Ms. Shamblin suggests, it would be necessary

10 to provide information about the risk of the proxy group regarding these

11 mechanisms. To the extent that the proxy group has implemented capital trackers

12 similar to the DSIC, then the Company's implementation of a DSIC does not

13 reduce its risk as compared to that benchmark group. Ms. Shamblin has provided

14 no evidence to suggest that the proxy group has greater risk than WVAWC as a

15 result of the implementation of the DSIC. Therefore, her suggested reduction in

16 the ROE is unsubstantiated and should be disregarded.

17 cted any analysis of the ~frastructurereplaceme~t trackers

18 t ave ente ies?

19 A. Yes. As shown in Schedule AEB-9, I have reviewed these mechanisms for the

20 operating subsidiaries of each of my proxy group companies. As shown in that

l6 Direct Testimony of Kaitlyn J. Shamblin, at 15 West Virginia American Water Company Case Nos. 18-O573-W-42T and 18-057643-421” Rebuttal Testimony of Anne E. Bulkley Page 31 of 46

1 schedule, tlie majority of the proxy group operating subsidiaries have implemented

2 some form of infrastructure surcharge tracking mechanism.

3 Q. at do you conclude from this analysis?

4 A. I conclude that WVAWC’s implementation of the DSIC is not risk-mitigating

5 when compared with the proxy companies. Therefore, since the data that is used

6 to estimate the ROE for WVAWC is market data that reflects the risks of the proxy

7 companies, including the subsidiaries shown in Schedule AEB-9, there is no need

8 to make any adjustment to the ROE to reflect the implementation of the DSIC.

9 Conversely, if the Company does not have a DSIC mechanism, it would be

10 exposed to greater business risk than the proxy group and that increased risk

11 should be reflected through an upward adjustment to the ROE.

12

13 ease s ’s rec

14 A. Ms. Shamblin is recommending a capital structure that consists of 5.03 percent

15 short-term debt, 49.29 percent long-term debt, 0.09 percent preferred stock, and

16 45.59 percent common equity.I7

17

18 A. Ms. Shamblin’s capital structure recommendation is based on the Company’s

19 capital structure as of December 3 1 , 201 7 with adjustments to short-term debt to

” Ibid. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Parre 32 of 46

reflect the historical average daily balances and adjustments to include recent long-

2 term debt issuances and equity infbsions.18

3 o you agree with this methodology for setting the capital structure for the

4 rate period that is t e subject of this proceeding?

5 A. No, I do not. The capital structure that is determined in a rate proceeding should

6 be based on the capital structure that will be in effect over the period that rates will

7 be in effect. While Ms. Shamblin makes some adjustments, her proposed capital

8 structure fails to reflect the changes in capital structure that are expected to occur

9 over the Addendum Period and is therefore not reflective of the Company’s

10 financing costs over the rate period.

11 As discussed in the testimony of the Mr. Nevrauskas, the Company’s proposed

12 capital structure is based on the actual 13-month average period ending December

13 31, 2017 and is adjusted for the changes that are expected to occur through

14 February 29, 20 19, the Addendum Period. Therefore, the Company‘s proposed

15 capital structure, which includes 2.97 percent short-term debt, 47.15 percent long-

16 term debt, 0.06 percent preferred stock and 49.82 percent common equity reflects

17 the capitalization of the Company over the rate period.

18 s the structure reaso

19 A. Yes, it is. As discussed in my Direct Testimony, the equity ratios of the proxy

20 group companies, excluding AWK, are in the range of 46.22 percent to 58.22

Id., at 4. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 33 of 46

1 AWC’s capital structure is within that range established by this group

2 and is therefore reasonable. Ms. Shamblin’s equity ratio lies below the lower

3 bound of that range. It is also unreflective of the actual equity that will be

4 supporting plant in West Virginia during the Addendum Period.

5 E.

7

8 A. Schedule AEB-16, summarizes the 2018 authorized ROES and equity ratios for

9 water utilities. In this schedule I calculate the equity return to investors (equity

10 ratio x ROE) resulting from those 2018 water utility rate cases. 1 also compare

11 that return to the equity return that would result from Ms. Shamblin’s proposed

12 ROE and equity ratio and my proposal to the range established by recently

13 OEs. As shown in Schedule AEB-16, the range of authorized equity

14 returns is 4.74 percent to 5.45 percent. Ms. Shamblin’s proposed ROE and equity

15 ratio results in an equity return of 4.23 percent, which is well below the low end

16 of the range established by recent authorizations. In contrast the equity return that

17 results from my proposed ROE is 5.38 percent which is above the median but

18 within the range established by those recent authorizatioiis. Based on the size of

19 WVAWC, I believe that a return that is slightly above the mean of recently

20 authorized equity returns is reasonable and appropriate. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 34 of 46

at suggests that t e cost of equity has been increasing?

A. Yes. A number of factors have weighed on utility stock prices. These include (i)

higher interest rates, (ii) greater stock market volatility, and (iii) the TCJA. There

4 have been seven one-quarter percentage point increases in the Federal Funds rate

since the FOMC began to normalize interest rates following the financial crisis

and the Great Recession. Going forward, there is an expectation of possibly two

additional interest rate increases in 201 8 and three more increases in 2019. Along

8 with these increases, and contrary to Ms. Shamblin‘s testimony that increases in

9 the Federal Funds rate have not corresponded with increases in longer-term bond

10 yields, the yield on 10-year Treasury notes has increased above the 3.00 percent

11 level in 2018 for the first time since 2014 and the yield on A-rated public utility

12 bonds has increased to 4.17 percent in August 201 8 &om 3.79 percent in December

13 20 17 -- a 3 8 basis point increase. Higher interest rates have been weighing heavily

14 on stocks, and utilities in particular.

15 cies react to a

16

17 A. The Company’s currently authorized ROE is 9.75 percent. A 48 basis point

18 reduction in ~A~7C’sreturn at a time when rating agencies are focused on cash

19 flow metrics would likely be viewed negatively. As discussed in my Direct

20 Testimony, the TCJA has been viewed negatively for utilities by the credit rating

21 agencies, due to concerns about cash flow coverage ratios. In January 2018

22 Moody‘s do~~ngradedthe outlook for several utilities, including AWK, the parent West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-3-42T Rebuttal Testimony of Anne E. Bulkley Page 35 of 46

1 company of WVAWC. In June 201 8, Moody’s issued a report in which the rating

2 agency downgraded the outlook for the entire regulated utility industry from stable

3 to negative for the first time ever. Moody’s cites ongoing concerns about the

4 negative effect of the TCJA on cash flows of regulated utilities. While noting that

5 “[rlegulatory commissions and utility management teams are taking important first

6 steps”I9 and that “we have seen some credit positive developments in some states

7 in response to tax reform,”2o Moody‘s concludes that “we believe that it will take

8 longer than 12-18 months for the majority of the sector to show any material

9 financial improvement from such efforts.”21

IO

11

12 A. The investment coinrnunity would find it very concerning if the Commission were

13 to adopt Ms. Shamblin‘s proposal to cut the Company‘s equity return by 48 basis

14 points. Investors would see West Virginia regulation as less supportive of the

15 Company at a time when high levels of capital investment are required, interest

16 rates are increasing, and the TCJA has imposed adverse financial circuinstances

17 on water, and other, utilities. The ROE established by the Commission to set rates

18 embodies in a single numerical value a clear signal of regulatory support for the

19 financial strength of the utilities it regulates. Although cost allocations, rate design

l9 Moody’s Investors Service, “Regulated utilities - US: 2019 outlook shifts to negative due to weaker cash flows. continued high leverage”, June 18, 2018, at 3. 2o Ibid. 21 Ibid. West Virginia American Water Company Case Nos. 18-0573-Vir-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 36 of 46

1 issues, and regulatory policies relative to the cost of service are important

2 considerations, the opportunity to earn a reasonable ROE represents a direct signal

3 to the investment community of regulatory support for the utility’s financial

4 strength (or lack thereof). In a single figure, the ROE utilized to set rates provides

5 a common and widely understood benchmark that can be compared from one

6 company to another, and is the basis by which returns on all financial assets (stocks

7 - both utility and non-regulated, bonds, money market instruments, and so forth)

8 can be measured. The allowed ROE is universally understood and communicates

9 to investors the types of returns that they can reasonably expect from an investment

10 in utilities operating in West Virginia.

11 ssion to consi~erinvestor ex

12 A. For a utility to obtain new capital and refinance existing capital at reasonable cost

13 and on reasonable terns, the authorized ROE must be high enough to satisfy

14 investors with returns that are commensurate with the risk of their investments.

15 The cost of equity proposed by Ms. Shainblin, if adopted by the Commission,

16 would provide a signal to the investment community of less supportive regulation

17 for WVAWC. That is to say, if the Commission were to adopt Ms. Shamblin’s

18 proposal, it would tend to discourage capital commitments by investors to West

19 Virginia because more attractive risk-adjusted returns are available in other states.

20 This holds equally true for investors in other companies and for the allocation of

21 investment capital within the AWK system. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42" Rebuttal Testimony of Anne E. Bulkley Page 37 of 46

1

2

3 A. The forecasted returns on equity for the water utility industry, as published in the

4 July 13, 201 8 edition of Value Line, are as follows:

5 igure

American States Water Co AWR 12.00% 13.00% 14.00% American Water Works Co, Inc. AWK 10.50% 10.50% 10.50% Aqua America, Inc. WTR 13.00% 13.00% 12.50% California Water Service Group CWT 11 .OO% 11.OO% 11.50% Connecticut Water Services CTWS 7.50% 9.50% 11.OO% Middlesex Water Company MSEX 11.OO% 11.OO% 12.50% SJW Corporation SJW 1 'I .50% 13.00% 14.00% York Water Company YORW 10.00% 10.00% 13.50% Mean 10.44% 11.38% 12.44%

6 Knowledgeable investors are aware of these returns and price the stocks of the

7 water utilities accordingly. Again, these data support an equity return for the

8 Company at 10.40 percent to 12.40 percent, validating the 10.80 percent ROE that

9 the Company has requested.

22 Value Line Water Utility Reports, July 13,2018. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-3-42T Rebuttal Testimony of Anne E. Bulkley Page 38 of 46

1 F.

2 s. ~hamb~ndeclines to acce t an a~jus~mentto the authorize^

3 C based on the small size of the Company relative to the proxy group.

4 lease co~ment.

5 A. Academic research has shown that a variety of factors explain the risk

6 compensation required by investors that exceeds the risk-free rate of return (i.e.,

7 the yield on Treasury obligations), It is for this reason that multi-factor models

8 have been developed in the academic community to explain investor-expected

9 returns. One of the more famous studies was conducted by Fama and French (see

10 "The Cross-Section of Expected Stock Returns," The Journal of Finance, June

11 1992), which identified size as a separate factor that helps explain returns. Fama

12 and French identified the size of a firm as a separate factor that must be recognized

13 in addition to the beta measure of systematic risk in explaining investor-expected

14 returns. As explained in my Direct Testimony, WVAWC is significantly smaller

15 than the proxy group coinpanies. While I have not made an explicit adjustment to

16 the authorized ROE for sinal1 size, I have considered that factor among other risks

17 that influence the investor-required ROE for ~AWC.

18

19 se su

20 A. Mr. Smith calculated the recommended revenue requirement using two

21 alternatives, 1) ai1 overall rate of return of 6.96 percent, relying on an ROE of 8.75

22 percent and 2) ai1 overall rate of return of 7.43 percent relying on an ROE of 9.75 West Virginia American Water Company Case Nos. 18-0573- -4211 and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 39 of 46

1 percent. In both scenarios Mr. Smith relies on a capital structure that includes 6.14

2 percent short-term debt, 45.99 percent long-term debt, 0.14 percent preferred stock

3 and 47.73 percent equity, which is the Company’s historical test year capital

4 Mr. Smith’s recommended ROE is 8.7.5 percent with a revenue

5 increase of no more than $1 7.5 million based on an ROE of 9.75 percent.

6

7 A. No, I do not. Mr, Smith has not provided any analysis to support his

8 recoininendation, nor ’has he offered any testimony that explains why he believes

9 it would be warranted to reduce the Company’s currently authorized ROE. As

10 noted previously, the Company’s currently authorized ROE is 9.7.5 percent. Mr.

11 Smith has provided no justification for his recommended ROE that is 100 basis

12 points below the currently authorized ROE. As discussed in my response to Ms.

13 Shamblin, recommendations to decrease the Company’s ROE are unwarranted and

14 unconstructive based on current market conditions, the rising interest rate

15 environment and the rating agencies’ concerns regarding cash flow metrics for

16 utilities following tax reform.

17 r. S se o or

18 r~turnto ysis?

19 A. No, it is not. The analysis presented in my Direct Testimony demonstrates a mean

20 result of 9.41 percent to high results of 1 1.67 percent for the CF model. Using

23 Direct Testimony of Ralph C. Smith, at 7 and Exhibit LA-I, Schedule D

V West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-427 Rebuttal Testimony of Anne E. Bulkley Pas40 of 46

1 the CAPM, the results are in the range of 11.50 percent to 12.1 1 percent. In

2 addition, reasonable corrections to Ms. Shamblin’s analyses result in weighted

3 returns, using DCF and CAPM methodologies in the range of 9.77 percent to 1 1.26

4 percent, with a mean result of 10.47 percent, considering all scenarios developed

5 in Figure 4 above. Mr. Smith’s unsubstantiated recommended ROE of 8.75 percent

6 is inore than 100 basis points below the low end of the range of results.

7 Furthermore, Mr. Smith’s 9.75 percent ROE - also relied on without any analytical

8 foundation - is at the low end of the range and is unreasonable given the rising

9 interest rate environnient and other financial risk factors discussed in my Direct

10 and Rebuttal testimonies.

11 structure reasona

12

13 A. It is not. As discussed in my response to Ms. Shamblin, the return that is set in

14 this proceeding is intended to be forward looking. Therefore, it is necessary to

15 rely on the expected capitalization of the Company over the forward-looking rate

16 period. As discussed by Mr. Nevirauskas, the Company’s proposed capital

17 structure began with the December 31, 2017 test year and was adjusted for

18 financing that is projected in the Addendum Period. Therefore, the Company’s

19 proposed capital structure is representative of the actual capitalization of the

20 company over the rate period and should be used in setting the revenue

21 requirement. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page41 of46

1

2

3 A. As shown in Schedule AEB-16, Mr. Smith’s proposed equity rate (equity ratio x

4 equity return) is lower than the equity rate that is proposed by Ms. Shamblin. As

5 shown in that exhibit, both Mr. Smith’s recommended equity rate, using an 8.75

6 percent ROE and his alternative, using a 9.75 percent ROE are below the low end

7 of the range established by recently authorized ROES.

8 hat are your eo r. ’s reco ations?

9 A. First, Mr. Smith has not provided any basis for his recommended ROE of 8.75

10 percent. As shown in my Direct and Rebuttal testimonies, based on the results of

11 the ROE estimation models, Mr. Smith’s recommended ROE is not reasonable.

12 Furthermore, comparing Mr. Smith‘s recommendation to recently authorized

13 returns for other water utilities, it is clear that his proposal does not meet the Hope

14 and Bluefield standards and should therefore be rejected.

15

16 ubin’s testi~on~~wit

17

18 A. Mr. Rubin suggests that among other reasons, the SM mechanism should be

19 rejected because ‘&itdoes not include a lower return on equity to recognize the shift

20 in business risk from share~oldersto

24 Direct Testimony of Scott Rubin at 28. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 42 of 46

1 ere any merit to ubin’s suggestion that the ROE should be re

is implemented?

A. No. Mr. Rubin’s point of comparison is flawed, which renders his conclusions

irrelevant. Mr. Rubin’s point of comparison is whether the Company has greater

or less risk after the implementation of the RSM. This is not the relevant point of

comparison when considering whether or not it is appropriate to adjust the ROE

for the implementation of any revenue recovery mechanism. The appropriate

8 question is whether or not the Company’s risk is increased or decreased reZutive

9 to the proxy gru~pthat is used to set the ROE. Mr. Rubin has not provided any

10 analysis of a risk-appropriate proxy goup or the appropriate return on equity.

11 Therefore, any recommendations that he makes with respect to the return on equity

12 are unsubstantiated and should be rejected.

13 ted any analys~sof the relat~verisks of

14

15 A. Yes, I have. As explained in my Direct Testimony, I conducted an analysis of the

16 relative risks of the proxy group companies and WAWC. This analysis is

17 appropriate because the ROE that is being set is based on data for a proxy group

18 of companies. As shown in Schedule AEB-9, many of the proxy companies have

19 forward test periods, revenue stabilization mechanisms and capital trackers. In

20 addition, a recent Brattle Group report summarizes several other regulatory

21 mechanisms that have been implemented by the proxy group companies to West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-427’ Rebuttal Testimony of Anne E. Bulkley Page 43 of 46

stabilize revenue and secure timely recovery of The approval of these

types of adjustment clauses, revenue decoupling mechanisms such as RSM, ROE

incentives iiders, trackers, forward test years, and cost recovery mechanisms by

regulatory commissions is widespread in the utility business and is already largely

5 embedded in financial data, such as stock prices, bond ratings, and business risk

6 scores. Moreover, it is important to note that investors generally do not associate

7 specific increments to their return requirements with specific rate structures.

8 Rather, investors tend to look at the totality of alternative regulatory mechanisms

9 in place relative to those in place at comparable companies when assessing risk.

10 The evidence demonstrates that the proxy companies have implemented some

11 form of alternative ratemaking mechanism to increase the companies’ ability to

12 achieve the revenue requirement that was authorized by the regulatory

13 commission. Based on this review, the returns for the proxy companies used in

14 my analysis already reflect any risk-reducing features of these mechanisms.

15 Therefore, it is not necessary to adjust the ROE to reflect the implementation of

16 the RSM.

17 f aIter~ativereg~lato

18

The Braale Group, “Alternative Regulation and Ratemaking Approaches for Water Companies: Supportiiig the Capital Investment Needs of the 2Is‘ Century”, September 30,2013. West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 44 of 46

1 A. Yes, there is. A comprehensive study by the Brattle Group26 investigated the

2 impact of a particular alternative regulatory mechanism, namely, revenue

3 decoupling, on risk and the cost of capital and found that its effect on risk and cost

4 of capital, if any, is undetectable statistically.

5 o you agree with ubin that there is a ri$~-shiftfrom $harehol~er$to

6 custo~ersresulting from decoup

7 A. No, I do not. The premise of rate regulation is that utilities should have a

8 reasonable opportunity to recover their costs and earn their authorized ROE. As

9 discussed in the Direct Testimony of Mr. Nevirauskas, 94.5 percent of WVAWC's

10 water system costs are fixed but only 45 percent of the revenues are fixed.

11 Therefore, the Company is recovering almost two-thirds of its fixed costs on a

12 variable cost basis. Under this structure, any reduction in volume directly affects

13 the Company's ability to recover its fixed costs. The existing rate structure,

14 coupled with declining use, do not provide the Company a reasonable opportunity

15 to recover its costs or earn its authorized ROE.

16 's reco~~e~de

17

18

26 Wharton, Vilbert, Goldberg & Brown, The Impact of Decoupling on the Cost of Capital: An Empirical Investigation, The Brattle Group, February 201 1; Wharton and Vilbert, Decoupling and the Cost of Capital - The Electricity Journal, September 08, 2015 West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Page 45 of 46

I A. There is no statistical analysis that demonstrates that the implementation of

2 decoupling mechanisms lower the cost of capital. As discussed in my

3 Rebuttal Testimonies, each of the proxy companies have implemented decoupling,

4 capital trackers or forward test years to increase the ability to achieve the revenue

5 requirement that was authorized by their regulatory commissions. Therefore, the

6 returns of the proxy companies already reflect any risk-reducing features of these

7 mechanisms. Because of the broad implementation of these mechanisms in the

8 proxy companies, WVAW~would be more comparable to the comparison group

9 if the RSM is approved. In the event that the RS is not approved for ~A~C,

10 the Company would have greater risk than the proxy group companies and that

I1 incremental risk should be reflected in the cost of equity.

12

13

14 A. I continue to support the analyses and recoini~e~idationcontained in my

15 Testimony. Specifically, the range of reasonable ROE results for the proxy group

16 companies is between 10.00 percent and 10.80 percent, and a reasonable cost of

17 equity for ~VAW~is 10.80 percent. My recommendation is supported by the

18 analyses filed in my Direct Testimony as well as the range of results from Ms.

19 Shamblin's Constant Growth DCF assumptions, as corrected in my Rebuttal

20 Testimony and the results of Ms. Shamblin's CAP using the entire Value Line

21 universe in the calculation of the market return. Furthermore, the equity rate that West Virginia American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Anne E. Bulkley Parre 46 of 46

1 results &om my recommended ROE and equity ratio is within the range established

2 by the recently authorized ROES around the country.

3 My recommendation is consistent with recent changes in market conditions, the

4 expectations for interest rates to continue to increase as well as the credit rating

5 agencies’ recommendations for steps that can be taken to stabilize utilities’ cash

6 flow metrics following tax refom.

7 Consistent with the recent conclusions of other regulators, my recommendation

8 takes into consideration both the results of the DCF model and risk premium

9 methodologies, specifically the forward-looking CAPM, as well as projected

10 returns for water companies from Value Line. In addition, my recommendation

11 considers other factors in determining the appropriate ROE, including company-

12 specific risk factors such as the small size of ~AWCrelative to the proxy group

13 companies and the capital investment requirements for the Company. Finally,

14 based on my review of the alternative rate mechanisms implemented by the proxy

15 companies, there is no basis to make any adjustment to the ROE for the Company’s

16 proposed RSM.

17

18 A. Yes, it does. West Virginia-AmericanWater Company Schedule AEB-14 Case Nos. 18-0573-W-42Tand 18-0576-5-421 Page 1of 1

Growth Rate Estimates

Value Line Value Line Zack's YahooiFinance Historical Projections Projected Projected Average Average High EPS Low EPS Price Dividend Dividend Yield DPS EPS Average DPS EPS EPS EPS EPS (A8C) Growth Rate Growth Rate (8) (AI ( C) American States Water Co .$ 5976 $ 110 184% 10 50% 7 00% 8 75% 8 OU?? 6 0036 6 00% 4.00% 533% 736% 6 00% 4 00% American Water Works Co. $ 8736 $ 185 2 12% 8 5P? 7 50% 8 00% 1000% 1000% 7 79% 8 10% 863% 888% 10 00% 7 79% Aqua America, 1%. $ 3662 $ 088 2.40% 8 00% 9 5056 8 75% 900% 750% 5 33% 5 Off? 594% 790% 7 50% 5 00% California Water Service Gmup $ 4078 $ 075 184% 2 50% 4 OO?? 3 25% 6 50% 9 50% 7 OP? 9 80% 877?? 6 17?? 9 8wb 7 00% Connecticut Water Service, Inc. 111 Consolidated Water Company Middlesex Water Company o 4480 $ 091 203% 2.00% a ow 5 00% 5 50% a 00% NA 2.70% 535% 528% 8 W? 2 70% SJW Corporation B 6329 $ 112 1779 5 00% 18 50% 11 75% 8.50% 6 00% NA 1400% 1O.W? 10.08% 1400% 6 00% York Water Company $ 3122 $ 067 t.15% 650M ___5 Off? -800% __9 00% -NA -49PA ___695% __665% 90056 a 2 Off? 5 44% a 94% 7 19% 7 63% 7 69% 6 53% 6 81% 709% 73096 8 79% 5 36%

Adjusted ROE Estlmates

Value Line Zacks YahoolFinance Low Average High Dividend Yield EP5 EPS EPS EPS EPS EPS American Stales Water Co 1 84% 734% 7.84% 5 84% 7 17% 7.84% Amencan Water Works Co 2 12% 12 12% 9 91% 10 22% 9 91% 10 75% 12 12% Aqua Amenca. Inc 2 40% 9 90Yo 7 73% 7 40% 7 40% 8 35% 9 90% Calrfomia Water Sewice Group 1 84% 11 34% 8 84% 11 64% 8 84% 1061% 1164% Connecticut WaterService. Inc 1 87% 7 37% 7 87% 7 37?? 7 62% 7 87% Consolidated Water Company Middlesex Water Company 2 03% 10 03% 4 73% 7 36% 10 03% SJW Corporation 1 77% 7 77% 15 77% 7 7P? 11 77% 15 7Ph York Water Company 2 15% i1 15% 7 05% 7 05% 9 10% 11 15% Mean 2 00% 9 69% 8 58% 8 81% 8 06% 9 09% 10 79%

Notes [I)Remwed Consolidated Water Company [2) Relied on EPS growth rates [3]Established a 7 0% floor on ROE results 14) Calculated ROE results based on low, mean and high EPS growth rates West Virginia-American Water Company Schedule AES-15 Case Nos. 180573-W42T and 180576-W2T Page 1of 2

CAPiTAL ASSET PRICING MODEL K=R,+p{Rm -R,)

CAPM USING HISTORICAL 30-YEAR TREASURY BOND YIELD AND PROJECTED MARKET RISK PREMIUM PI [21 [31 141 151 Market Risk-Free Market Risk Rate Beta Return Premium ROE Pmxy Group IRfl (6) fRml fRm -RI) 6) American States Water Co 3.40% 0.80 14.51% 11.11% 12.29% American Water Works Co. 3.40% 0.65 14.51% 11 11% 1062% Aqua America, Inc 3.40% 0.75 1451% 1111% 11.73% California Water Service Gmup 3.40% 0.80 14.51% 11.11% 12.29% Connecticut Water Service, inc 3.40% 0.65 14.51% 11.11% 10.62% Middlesex Water Company 3.40% 0.80 14.51% 11 11% 12.29% SJW Corporation 3.40% 0.75 14.51% 11.11% 11.73% York Water Company 340% 0.80 14.51% 11.11% 12.29% 0.75 11.73% Notes: [I] Source: Appendix US-1 Schedule 3. Sheet 2 [2] Source: Appendix IUS1 Schedule 3, Sheet 2 [3] Source: Schedule AE55 141 Equals [3] - [I] 151 Equals [I]+ 121 x [4]

CAPM USING MS. SHAMBLIN'S PROJECTED LONG-TERM RISK FREE RATE AND PROJECTED MARKET RISK PREMIUM

Risk-Free Market Risk Rate Beta Return Premium ROE Proxy Gmup fRr) fBI fRm) IRm -RIJ fW American States Water Co 3.23% 0.80 14.51% 11.28% 12.25% American Water Works Co. 3.23% 0.65 14.51% 11.28% 10.56% Aqua America Inc. 3.23% 0.75 14.51% 11.28% 11.69% California Water Service Group 3.23% 0.80 14.51% 11.28% 12.25% Connecticut Water Service, Inc 3.23% 0.65 14.51% 11.28% 10.56% Middlesex Water Company 3.23% 0.80 14.51% 11.28% 12.25% SJW Corporation 3.23% 075 14.51% 11.28% 11.69% York Water Company 3.23% 0.80 14.51% 11.28% 12.25% 0.75 11.69%

Notes' [I] Source: Appendix IUS-1 Schedule 3, Sheet 3 [2] Source: Appendix IUS1Schedule 3, Sheet 3 [3]Source: Schedule AE55 [4] Equals [3] - [I] [5] Equals [I] + [2] x [4]

CAPM USING MS. SHAMBLIN'S TREASURY BILL RATE AND PROJECTEDMARKET RISK PREMIUM [I1 (21 131 141 [51 Market Risk-Free Market Risk Rate Beta Return Premium ROE Pmxy Group IRt) fP) (Rm) {Rm- Rt) fKJ American States Water Co 2.39% 060 14.51% 12.12% 12.09% American Water Works Co. 2.39% 0.65 14.51% 12.12% 10.27% Aqua America, Inc. 239% 0.75 14.51% 1212% 11 48% California Water Service Gmup 2.39% 0.80 14.51% 12.12% 12.09% Connecticut Water Service. Inc. 2.39% 0.65 14.51% 12.12% 10.27% Middlesex Water Company 2.39% 0.60 14.51% 12.12% 12.W% SJW Corporation 2.39% 0.75 14.51% 12 12% 11 48% York Water Company 2.39% 0.80 14.51% 12.12% 12 09% 0.75 11.48%

Notes: [I]Source. Appendix IUS-1 Schedule 3, Sheet I p] Source: Appendix IUS1 Schedule 3, Sheet 1 [3] Source: Schedule AE55 [4] Equals 131 - [I] [SI Equals [I]+ [2] x [4] * West Virginia-American Water Company Schedule AEB-16 Case Nos. 18-0573-W-42T and 18-0576-5-42T Page 1of 1

Comparison of RRA Major Rate Case Decisions to Ms, Shamblin's Recommendation

Order date Company ROR ROE Equity Ratio Equity Return 3/7/2018 Aqua Illinois 7.52% 9.60% 53.22% 5.11% 3/22/2018 California American Water Co. 7.61% 9.20% 55.39% 5.10% 3/22/2018 California Water Service Co. 7.48% 9.20% 53.40% 4.91% 3/22/2018 Golden State Water Co. 7.91% 8.90% 57.00% 5.07% 3/22/2018 San Jose Water Co. 7.64% 8.90% 53.28% 4.74% 3/24/2018 Middlesex Water Co. 6.50% 9.60% 52.75% 5.06% 5/2/2018 Missouri American Water Co. NA NA NA NA 5/2/2018 Carolina Water Service, Inc. 8.40% 10.50% 51.89% 5.45% Mean 5.06% Min 4.74% Max 5.45%

Shamblin Proposal 9.27% 45.59% 4.23% Smith Alternative ROE 9.75% 47.73% 4.65% Smith Proposal 8.75% 47.73% 4.18% Bulkley Recommendation 10.80% 49.82% 5.38% PUBLIC SERVICE COMMISSION OF WEST VIRGINIA CHARLESTON

CASE NO. 18-0573-W-42T

WEST VIR~IN~A-AMERICANWATER COMPANY, Rule 42T application to increase sewer rates and charges and

CASE NO. 18-0576-S-42T

WEST VIRG~IA-A~E~ICANWATER COMPANY, Rule 42T application to increase water rates and charges.

UTTAL TESTIMONY OF

PATRICK L. BARYENBRUCH

October 12,2018 West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 1 of 10

1 Q. DID YOU PREVIOUSLY FILE DIRECT TESTIMONY IN THIS CASE?

2 A. Yes, I filed direct testimony and a study that evaluated the reasonableness of charges from

3 and the necessity of services provided by American Water Works Service Company, Inc.

4 (“Service Company”) to West Virginia American Water Company (‘‘WAmerican”).

5

6 Q. WHY HAVE YOU PREPARED THIS REBUTTAL TESTIMONY?

7 A. I am responding to a portion of the direct testimony of Ralph C. Smith, witness for the

8 Consumer Advocate Division of the Public Service Commission of West Virginia.

9

10 Q. WHICH OF MR. SMITH’S COMMENTS DO YOU WISH TO AD

11 A. I will respond to his criticisms of my study. These comments can be found in section VI11

12 on pages 59-61 of his testimony.

13

14 Q. PLEASE RESTATE THE PURPOSE OF THE STUDY YOU PERFORMED FOR WV

15 AMERICAN.

16 A. My study was undertaken to answer four questions concerning the services provided by

17 the Service Company to WV American during 2017. First, were the Service Company’s

18 charges to Mnl’ American during 2017 reasonable? Second, was WV American charged

19 the lower of cost or market for managerial and professional services provided by the

20 Service Company during 2017? Third, were 2017 costs of the Service Company’s

21 customer accounts services, including those of the National Call Centers, comparable to West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-8-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 2 of 10

1 those of other utilities? Fourth, are the services WV American receives from Service

2 Company necessary?

3

4 Q. HOW WOULD YOU CHARACTERIZE MR. SMITH’S COMMENTS ON YOUR

5 STUDY?

6 A. Mr. Smith criticizes three aspects of my 45-page study: (1) that WV American is not

7 comparable to electric and combination electridgas utilities that are included in my study‘s

8 comparison groups, (2) that it covers 201 7 costs and not the higher charges from the Service

9 Company beyond 201 7 that have been requested for recovery by WV American, and (3)

10 that it used 2017 American customer counts to calculate costs per customer for

11 cornparison to other utilities.

12

13 Q. IN HIS FIRST CRITICISM, M . SMITH CLAIMS (P. 60) THAT ELECTRIC

14 COMPA~IESARE NOT CO PARABLE TO WV AMERICAN. IS MR. SMITH’S

15 UE, PRESUMABLY THAT YOU IMPR~PERLYINCLUDE

16 UTILITIES IN YOUR COMPARISO~GROUPS, VALID?

17 A. No. First, Mr. Smith simply says “no” when asked if electric companies are comparable

18 to WV American. This is hardly a reasoned explanation of why it might be improper to

19 compare the two. In any event, this comparison is not improper because, as utilities,

20 electric and water companies perform certain similar processes in the delivery of their

21 services. Please let me explain. y study compares V American‘s charges from the West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 3 of 10

1 Service Company with other utility companies in two of the study’s four questions.

2 Question 1 compares the Service Company’s cost of administrative and general (A&G)-

3 related services to the same costs of service companies that are part of utility holding

4 companies. Question 3 compares WV American‘s cost of customer account services

5 (customer contact, bill printing, collection, payment processing) to those of neighboring

6 electric and combination electridgas utilities.

7

8 WITH RESPECT TO QUESTION 1, WHY IS IT APPROPRIATE TO INCLUDE

9 SERVICES PROVIDED TO ELECTRIC COMPANIES IN YOUR COMPARISON?

10 A. It is appropriate to use electric company data because these service companies provide the

11 same type of A&G services to their regulated electric and combination electridgas utility

12 affiliates that the Service Company provides to Mrv American. Examples of these services

13 include finance, accounting, taxes, legal, human resources, information technology and

14 executive management. One reason for the similarity is the fact that different types of

15 utilities use the same A&G-related information systems. For instance, my clients,

16 Dominion Energy and Southern California Edison use SAP, the same enterprise resource

17 planning system used by American Water. My clients American Electric Power and Duke

18 Energy use Power Plant for fixed asset accounting, the same system used by American

19 Water. The workflow, or business processes, associated with these systems are the same

20 regardless ofthe type of utility service. West Virginia-American Water Company Case Nos. 18-0573-W-421' and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 4 of 10

On the other hand, I did not include operations and maintenance (O&M)-related

charges in my service company cost comparison because a water company's O&M

activities (e.&.,water treatment, transmission, distribution) are clearly different than those

4 of an electric utility (e.&.,power generation, transmission and distribution).

5 I might add that it is not possible to perform a cost comparison for only water utility

6 service companies because their cost information is not publicly available. Service

7 companies that are part of an electric and combination electric/gas utility holding company

8 must file a Form 60 with the FERC. Information in the Form 60 is sufficiently detailed to

9 allow me to isolate applicable A&C expenses and make a valid comparison. The service

10 company comparison methodology I employed for tlie WV American study is the same I

11 have used in other cases and has been accepted by regulators in other states as evidence of

12 the reasonableness of service company charges to regulated affiliates.

13

14 Q. FOR YOUR STU Y'S QUESTION 3, D YOU COMPARE WV AMERICAN WITH

15 IC UTILITIES?

16 A. Question 3 compares WV American's cost of customer account services (customer contact,

17 bill printing, collection, payment processing) to those of neighboring electric and

18 combination electridgas utilities. Here, too, the nature of these activities is similar for

13 water utilities and electric utilities and so a comparison of the costs is quite appropriate. I

20 make this comparison in order to incorporate the Service Company's call center costs into

21 my benchmark comparisons. Again, there is no publicly available source of water West Virginia-American Water Company Case Nos. 18-0573-W-423:and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 5 of 10

1 company-only customer account services costs, including charges from service company

2 affiliates. As with question 1, this comparison approach has been accepted by regulators

3 in other states as evidence of the reasonableness of service company charges.

4

5 Q. IS MR. SMITH CORRECT THAT YOUR STUDY ONLY COVERS 2017 CHARGES

6 FROM THE SERVICE COMPANY?

7 A. Yes, by necessity, it covers actual 2017 charges from the Service Company to WV

8 American because I require all the detail available with actual costs incurred to properly

9 perform my cost comparisons of Service Company’s cost to costs rendered by other utility

10 service companies. Consequently, WV American 201 7 actual charges from the Service

11 Company must be compared to actual costs of comparison group utilities. There is no way

12 to compare projected WV American Service Company charges to the prospective costs of

13 other utilities because that information does not exist.

14

15 Q. IN HIS SECOND CRITICISM, MR. SMITH CLAIMS (AT PAGE 60) THAT BECAUSE

16 YOUR COMPARISON IS BASED ON ACTUAL DATA, “IT PROVIDES NO

17 JUSTIFICATION FOR THE INCREASE IN AFFILIATED SERVICE COMPANY

18 COSTS BEYOND 2017 THAT HAVE BEEN EQUESTED BY ~AWC.”WOULD

i 19 THE INCLUSION OF POST-2017 SERVICE COMPANY CHARGES TO WV

20 AMERICA^ HAVE CHANGE YOUR COST CQMPARISO~CONCLUSIONS? West Virginia- American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 6 of 10

I A. y study provides evidence that the Service Company’s 20 I 7 charges are reasonable.

2 If I factor in WV American’s requested percent increase in Service Company O&M

3 expenses into my A&G cost calculation, the Service Company’s charges are still

4 reasonable compared to other utility companies. The Service Company’s actual 201 7

5 O&M charges to WV American were $1 1.494 million. The Company is requesting recover

6 of $12.684 million in Service Company O&M charges, an increase of 10.4% fi-om 2017.

7 The Service Company’s 2017 actual A&G-related charges were $66 per WV American

8 customer. Increasing that aniount by 10.4% brings the cost per customer to $73. This is

9 still considerably lower than the utility comparison group‘s 2016 (the latest year available)

10 average of $1 13 per customer. At $73 per customer, 15 of 24 total comparison group

11 utilities still have a higher cost than WV American’s $73 proforma cost per customer. This

12 comparison demonstrates that Mr. mith‘s criticism that my study is invalid because it was

13 not based on projected costs is misplaced and misleading.

14

15 ITH‘S FINAL CRITICISM CONCERNING YOUR USE OF 201 7 ACTUAL

16 R COUNTS IN COST CALCULATIONS VALID?

17 A. No. Mr. Smith actually states (p. 60) that “Dividing an amount of affiliated Service

18 Company charges to ~VAWCby a lower customer count would thus presumably produce

19 higher per-customer amount for WAWC than considered in [my] study.“ This is

20 supposition, not evidence. In contrast, my methodology calls for the use of actual data

21 that has been recorded in my client‘s books of record. In the case of WV American, this West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 7 of 10

1 means I divide actual 2017 costs incurred by the actual 2017 customer count. This

2 produces a valid metric because the costs are properly matched with the number of

3 customers served at the time the measurement is made. For the reasons expressed above,

4 it is inappropriate to use projections to compare WV American’s costs to those of other

5 utilities that rest on historical data. For the same reasons, it is inappropriate to base the

6 comparison on projections of customers.

7 Mr. Smith‘s logic is wrong. I am comparing WV American‘s actual 2017 costs per

8 actual customers to the same 2017 per-customer costs of other utilities. To ensure the

9 results produce a proper cost comparison, I do not make prospective adjustments to any of

10 the underlying numbers.

11

12 Q. WOULD YOUR CONCLUSION CHANGE REGARDIlVG THE REASONABLENESS

13 OF SERVICE COMPANY A&G CHARGES IF YOU USED FUTURE COSTS AND

14 CUSTOMER COUNTS?

15 A. No. In the table below, I recalculate Service Company A&G charges by factoring in a

16 10.4% cost increase and the proforma customer count. The resultant proforma A&G cost

17 per WV American customer increases from $66 to $74. This amount is still below than

18 the 2016 comparison group average of $1 13. WV American’s proforma amount is still

19 lower than 15 of the comparison group’s total 24 utility companies. It should be noted this

20 analysis compares the Service Company’s proforma 201 9 costs to the comparison group’s West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 8 of 10

1 2016 costs. This comparison to other utility companies of the Service Company’s

2 proforma A&G charges to WV American Water shows them to be reasonable.

Actual Proforma 2017 Adjustment Amount Service Company Charges A&G Charges (A) $11,061,897 $ 1,145,074 $12,206.970 WV American Customer Count (B) 167,127 (2,175) 164,952 WV American A&G SC Charges/Customer $ 66 $ 14

Comparison Group Average (2016) $ 113 $ 113 Total Utility Companies in Comparison Group 24 24 igher Costs per Customer 11 15

Note A: Adjustment is percent increase requested in total O&M expenses as calculated below. Requested O&M charges $12,684,000 201 7 actual O&M charges $11,494,179 Dollar increase from 201 7 actual $ 1,189,821 Percent Increase from 20 17 actual 10.4% Note B: Adjustment is based on the projected wstomer attrition as calculated below. Actual customer count at 1213 112017 167,127 Proforma customer count I64?952 3 Decrease in customers (2,175)

4

5 Q. WOUL YOUR CONCLUSION CHANGE REGA INC THE REASONARLENES~

6 OF SERVICE COMPANY CUSTOMER ACCOUNT SERVICES CHARGES IF YOU

7 USED FUTURE COSTS AND CUSTOMER COUNTS?

8 A. No. In the table below, I recalculate Service Company customer account services charges

9 by factoring in a 10.4% cost increase and the proforma custoiner count. The resultant

10 proforma customer account services cost per WV American customer increases f?om $32

11 to $35. This amount is somewhat above the 2016 comparison group average of $33. WV

12 American‘s proforma amount is still lower than 7 of the comparison group’s total 25 utility West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick L. Baryenbruch Page 9 of 10

companies. It should be noted this analysis compares the Service Company’s proforma

2019 costs to the comparison group’s 2016 costs. This comparison to other utility

companies of the Service Company’s proforma customer account services charges to WV

American Water shows them to be reasonable.

Actual Proforma 2017 Adjustment Amount A&G Service Company Charges (A) $ 5,266,849 $ 545,199 $ 5,812,047 WV American Customer Count (B) 167,127 (2,175) 164,952 WV American A&G SC ChargesICustorner $ 32 $ 35

Comparison Group Average (2016) $ 33 $ 33 Total Utility Companies in Comparison Group 25 25 Number with Higher Cost per Customer 10 7

Note A: Adjustment is percent increase requested in total O&M expenses as calculated below. Requested O&M charges $12,684,000 201 7 actual O&M charges $11,494,179 Dollar increase from 201 7 actual $ 1,189,821 Percent Increase from 2017 actual 10.4% Note B: Adjustment is based on the projected customer attrition as calculated below. Actual customer count at 1213 1/20 17 167,127 Profonna customer count 164,952 5 Decrease in customers (2,175)

6

7 Q. HAVE OTHER REGULATORS ACCEPTED THE METHODLOGY YOU EMPLOYED

8 IN PERFORMING THE MARKET COST COMPARISON STUDY OF WV

9 AMERICAN’S 2017 CHARGES FROM THE SERVICE COMPANY?

10 A. Yes. I have performed more than 100 market cost comparison studies for 37 utility clients

11 in 17 states. In over 70 of these engagements, I acted as my utility client’s expert witness

12 covering the reasonableness and necessity of affiliate transactions. In every case, my West Virginia-American Water Company Case Nos. 18-0573-W-42T and 18-0576-S-42T Rebuttal Testimony of Patrick t.Baryenbmch Page 10 of 10

client's regulators accepted my cost comparison methodology. The table below lists my

2 affiliate transactions-related assignments.

Rate Case Rate Case Client Year Purpose Witness? Client Year Purpose Witness? Zonnecticut American Water 1999 Rate Case Yes ay State Gas (MA) 2004 Rate Case Yes llinois American Water 2007 Rate Case Yes .olumbia Gas of Massachusetts 2011 Internal Info No ndiana American Water 2017 Rate Case Yes 2012 Internal Info No

--.~~___ 2010 Rate Case ___-~ Yes 'ennsylvania American Water 2008 Compliance No 201 1 Compliance No 2014 Compliance No rennessee American Water 2006 Rate Case Yes 2010 Rate Case . Yes Jirginia American Water 1996 Rate Case Yes 1999 Rate Case Yes 2000 Rate Case Yes 2001 Rate Case Yes 2003 Rate Case Yes 2007 Rate Case Yes 2009 Rate Case Yes 201 1 Rate Case Yes 2014 Rate Case Yes %st Virginia American Water 2002 Rate Case Yes 2006 Rate Case Yes 2007 Rate Case Yes 2009 Rate Case Yes 2012 Rate Case Yes 2010 Rate Case Yes 2014 Rate Case Yes 2017 Rate Case Yes 4tlanta Gas Light (AGL Resources) 2009 Rate Case Yes 4tmos Ener9 Corporation (VA) 2004 Compliance No Electric Transmission Texas 201 6 Rate Case Yes 3

4 Q. DOES THIS COMPLETE YOUR REBUTTAL TESTIMONY?

5 A. Yes.