sets the price investors are willing to pay for a share of a company's stock.161

Transcription

1 TIEC and Cities argued that regulator$y protections make utilities less risky than firms operating in competitive markets.15 However, the Non-Utility Proxy Group is limited to a low-risk group of companies that represent the pinnacle of corporate America. Moreover, while utilities operate under a regulatory regime that differs from firms in the competitive sector, any risk-reducing benefit of regulation is already incorporated in the overall indicators of investment risk Dr. Avera presented. 159 Critically, the impact of regulation on a utility's investment risks is one of the key elements considered by credit rating agencies and investment advisory services, such as S&P and Value Line, when establishing corporate credit ratings and other risk measures. As a result, the impact of regulatory protections is already reflected in the risk analysis Dr. Avera presented. 160 For these reasons, the Commission should not adopt the PFD's position that consideration of the Non-Utility Proxy Group was an error and, at a minimum, should reject the PFD's position that Dr. Avera relied on the Non-Utility Proxy Group for each of his analyses when that it clearly not the case. 2. DCF Analysis The constant growth DCF model attempts to replicate the market valuation process that sets the price investors are willing to pay for a share of a company's stock.161 The model assumes that investors evaluate the risks and expected rates of return from all securities in the capital markets.162 By estimating cash flows investors expect to receive from the stock in the way of future dividends and capital gains, one can calculate their required rate of return. Importantly, in performing his DCF analysis, Dr. Avera looked at both a utility proxy group and a non-utility proxy group. For the utility proxy group, the dividend yields for the firms ranged from 1.9% to 5.6%.163 Dr. Avera then considered long-term growth expectations 158 TIEC Ex. 2 at 564 (Direct Testimony of Michael P. Gorman); Cities Ex. 3 at 87 (Direct Testimony of Stephen G Hill). 159 Lone Star Ex. 16 at 112 (Amended Direct Testimony of William E. Avera). 161 Id at Id. at Id. at Id. at

2 for each firm based on the assumption that earnings, dividends, book value, and market price will grow in lockstep, and the growth horizon of the DCF model is infinite.164 Combining the dividend yields and respective growth projections for each utility, and after eliminating illogical low and high-end values, Dr. Avera calculated common equity estimates ranging from 9.4% to 10.7%.16s Performing the same analysis using the non-utility proxy group resulted in cost of common equity estimates ranging from 10.6% to 12.0%.166 As mentioned above, the DCF analyses performed by other witnesses included numerous flaws. Lone Star identifies many such flaws below. TIEC witness Mr. Gorman's constant growth DCF analysis relies on understated growth estimates that lack empirical analysis and create an unreasonable result. Mr. Gorman's analysis resulted in an average constant growth DCF that is suspect because it is dependent on many estimates of the utilities' future ROE, the percentage of earnings retained, future stock issues, and the relationship of the price of those issues to book value of each company. 167 As shown on Lone Star Ex. 26 at Exhibit WEA-R-5, simply screening Mr. Gorman's DCF results based on analyst growth rates to eliminate illogical, low-end values results in a revised DCF cost of equity estimate of 9.89%, which is 89 basis points higher than the one he proposed, and within the range calculated by Dr. Avera using his constant growth DCF model. TIEC witness Mr. Gorman's multi-stage growth DCF model is flawed because Mr. Gorman failed to establish that his growth projections are actually used by investors. The evidence establishes that Mr. Gorman used average growth rate values to determine the initial dividend cash flows for each utility without any consideration for the reasonableness of the underlying data.' 68 S&P recently noted that despite slow economic growth, capital spending in the electric utility industry is rising significantly,169 and Mr. Gorman's own source notes that the electric utility industry 164 Id. 165 Id. at 62-63, Exhibit WEA Id. at 63, Exhibit WEA Lone Star Ex. 26 at (Rebuttal Testimony of William E. Avera). 168 Id. at Standard & Poor's, U.S. Utilities' Capital Spending Is Rising, And Cost Recovery Is Vital, RATINGSDIRECI' (May 14, 2012). 51

3 "may boost capex spending by 30% in the years ahead." 170 This contradicts his suppositions regarding GDP growth and supports the reasonableness of the analysts' growth estimates relied on by the Company. Mr. Gorman also incorrectly equates cash flows with the current market price of the stock. Correcting for this and the fact that there is no support for the key assumption underlying Mr. Gorman's multi-stage DCF models-i.e., that investors expect growth rates for electric utilities to converge to GDP growth in five years' 71-produces an average DCF estimate of 9.69%.172 Staff witness Ms. Winker's multi-stage DCF model does not provide reliable results. Ms. Winker's multi-stage DCF model is inconsistent with the traditional, wellaccepted constant growth DCF model routinely relied on by Commission Staff.173 The Company demonstrated at hearing that, had Ms. Winker relied on the traditional constant growth model, the median of her results exclusive of outliers would have been 9.67%."a Like Mr. Gorman, Ms. Winker attempted to predict the cash flows that would accrue to investors over the next 150 to 200 years. There is, however, no support for the key assumption underlying Ms. Winker's and Mr. Gorman's multi-stage DCF models; namely, that investors expect growth rates for electric utilities to converge to GDP growth in five years.175 Cities witness Mr. Hill's DCF analysis is flawed and does not provide a reasonable basis for determining the appropriate ROE. Mr. Hill's constant growth DCF analysis relies on growth estimates that include outliers and produce unreasonable results.176 The internal growth rates Mr. Hill calculated are biased downward because of computational errors.' 77 Mr. Hill's reliance on dividends per share ("DPS") growth rates as a determinant of long-term sustainable growth is misplaced and does not provide a meaningful guide to investors' current growth expectations because utilities have significantly altered 170 TIEC Ex. 2 at 10 (Direct Testimony of Michael P. Gorman). 171 Lone Star Ex. 26 at (Rebuttal Testimony of William E. Avera). 172 This is equal to the average of the mean and median values, consistent with Mr. Gorman's methodology. TIEC Ex. 2 at 36 (Direct Testimony of Michael P. Gorman); see Lone Star Ex. 26 at Exhibit WEA-R-4 (Rebuttal Testimony of William E. Avera). 173 Docket No , Direct Testimony of Slade Cutter at (Apr. 3, 2012). 174 Lone Star Ex Lone Star Ex. 26 at 63 (Rebuttal Testimony of William E. Avera). 16 Cities Ex. 3 at Attachment B (Direct Testimony of Stephen G. Hill). 177 Bangor Hydro-Elec. Co., 122 FERC 61,265 (2008). 52

4 their dividend policies in response to more accentuated business risks in the industry. 178 Mr. Hill erroneously calculates the average individual growth rates without any consideration for the reasonableness of the underlying data.179 As shown on page 2 of Mr. Hill's Schedule 5, almost one-half of the historical DPS growth rates for the companies in Mr. Hill's proxy group were less that 1.0%, with four being zero or negative. Combining a growth rate of 1.0% with Mr. Hill's dividend yield of 4.51 % (Schedule 6) implies a DCF cost of equity of approximately 5.5%.180 This implied cost of equity is not materially different than the yield from triple-b public utility bonds, which averaged 5.0% over the six-months ended June Thus, Mr. Hill's analysis leads to the absurd conclusion that utility stocks are no more risky than utility bonds. 182 Mr. Hill's method of removing outliers is unsound, contrary to FERC precedent, and inconsistent with Mr. Cutter's testimony for the Commission Staff in recent cases. Screening Mr. Hill's DCF cost of equity estimates based on historical EPS and book value per share growth rates to eliminate illogical values results in an implied cost of equity of 10.89%.183 The average cost of equity imlied by Mr. Hill's corrected DCF analysis based on projected growth rates is 10.17%. 84 OPUC witness Dr. Szerszen presents a DCF analysis that is flawed and unsound. Dr. Szerszen's constant growth DCF analysis is improperly based on historical data and fails to consider projected earnings.' 85 Dr. Szerszen's DCF analysis is further flawed by the fact that she averages individual growth rates with no consideration for the reasonableness of the underlying data. As shown on the five-year historical EPS growth rates reported on page 1 of Schedule CAS-4, the projected EPS growth rates Dr. Szerszen relied on in her DCF application included negative growth rates, which imply a cost of equity less than the utility's dividend yield. Dr. Szerszen should have eliminated negative growth rates that produce illogical DCF results from her analysis.' 86 17$ For example, the payout ratio for electric utilities fell from approximately 80 percent historically to on the order of 60 percent. Value Line, THE VALUE LINE INVESTMENT SURVEY at 161 (Sept. 15, 1995); Value Line, THE VALUE LINE INVESTMENT SURVEY at 136 (Feb. 24, 2012). 19 Id. at 75. "0 Id. at ' Moody's Analytics, Yields & Spreads Data, Lone Star Ex. 26 at 74 (Rebuttal Testimony of William E. Avera). 183 Id. at Exhibit WEA-R Id. at 77 & Exhibit WEA-R See id. at Id. at

5 3. Risk Premium Analysis Like the DCF model, the risk premium method is capital market oriented. 187 However, unlike DCF models, which indirectly impute the cost of equity, risk premium methods directly estimate investors' required rate of return by adding an equity risk premium to observable bond yields. 188 Lone Star witness Dr. Avera's estimates of equity risk premiums for electric utilities are properly based on surveys of previously authorized rates of return on common equity. 181 It is, however, important to recognize that the historical focus of the risk premium studies will not fully capture the significantly greater risks that investors now associate with providing electric utility service.190 As a result, they are likely to understate the cost of equity for a firm operating in today's electric power industry. 191 As illustrated on page 1 of Dr. Avera's Exhibit WEA-8, with the yield on average public utility bonds in November 2011 being 4.37%, this implied a current equity risk premium of 5.26% for electric utilities.' 92 Adding this equity risk premium to the yield on triple-b utility bonds of 4.93% produces a current cost of equity of approximately 10.2%.193 Incorporating a forecasted yield for and adjusting for changes in interest rates since the study period implied an equity risk premium of 4.45% for electric utilities.194 Adding this equity risk premium to the average implied yield on triple-b public utility bonds for of 6.80% resulted in an implied cost of equity of approximately 11.3% Lone Star Ex. 16 at 72 (Amended Direct Testimony of William E. Avera). 188 id. 189 Id. 19' Id. at Id. 192 Id. & Exhibit WEA-8. '93 Id at Id. at Id. 54

6 reasons: Intervenor recommendations using the risk premium analysis are faulty for the following TIEC witness Mr. Gorman provided a Risk Premium Analysis that is incomplete. Like Staff witness Ms. Winker, Mr. Gorman subjectively chose to truncate the data available to apply his risk premium approach by excluding all data before 1986.I96 By choosing a truncated time period for his risk premium study, Mr. Gorman unnecessarily introduces a subjective bias that taints his analysis and artificially lowers his results. Correcting Mr. Gorman's risk premium analysis to include all available data and to account for the inverse relationship between bond yields and equity risk premiums results in a current cost of equity estimate for Lone Star of 10.23%, or 11.50% after incorporating projected bond yields. 197 Staff witness Ms. Winker provides a similarly incomplete Risk Premium Analysis. Ms. Winker's Risk Premium Analysis is incomplete because it subjectively truncates the data available and ignores all observations prior to Ibbotson Associates (now Morningstar) noted the pitfalls of such a subjective approach: Some analysts estimate the expected risk premium using a shorter, more recent time period on the basis that recent events are more likely to be repeated in the near future... This view is suspect...i98 4. CAPM Analysis Dr. Avera's CAPM supports an ROE as high as 12.0%. The CAPM is a theory of market equilibrium that measures risk using the beta coefficient.199 Assuming investors are fully diversified, the relevant risk of an individual asset (e.g., common stock) is its volatility relative to the market as a whole, with beta reflecting the tendency of a stock's price to follow changes in the market. 200 The dividend yield for each company in the S&P 500 that pays dividends was obtained from Value Line, and the growth rate was equal to the consensus earnings growth projections for '96 TIEC Ex. 2 at 37 (Direct Testimony of Michael Gorman). 197 Lone Star Ex. 26 at Exhibit WEA-R-10, p. 2 (Rebuttal Testimony of William E. Avera). 198 Ibbotson Associates, 2005 YEARBOOK, VALUATION EDITION, 80 (2006). 199 Lone Star Ex. 16 at 64 (Amended Direct Testimony of William E. Avera). 200 id. 55

7 each firm published by IBES, with each firm's dividend yield and growth rate being weighted by its proportionate share of total market value. 01 Based on the weighted average of the projections for the 370 individual firms, current estimates imply an average growth rate over the next five years of 11.0%202 Combining this average growth rate with a year-ahead dividend yield of 2.5% results in a current cost of common equity estimate for the market as a whole of approximately 13.5%.203 Subtracting a 3.0% risk-free rate based on the average yield on 30-year Treasury bonds produced a market equity risk premium of 10.5%.204 Dr. Avera also utilized Morningstar's size premiums to account for the level of a firm's market capitalization in determining the CAPM cost of equity.205 As shown on page 1 of Exhibit WEA-7, adjusting the 10.8% theoretical CAPM result to incorporate this size adjustment results in an average indicated cost of common equity of 11.6%.206 In addition, because there is widespread consensus that interest rates will increase materially in the near future, it is likely that current bond yields understate the capital market requirements that will exist when this case is concluded. To compensate for this fact, Dr. Avera reasonably applied the CAPM based on the forecasted long-term Treasury bond yields. These long-term Treasury bond yields were developed based on projections published by Value Line, IHS Global Insight and Blue Chip.207 After incorporating a forecasted yield for (the time period rates from this proceeding will likely be in effect), the result was a theoretical CAPM 201 Id. at Id. 203 Id. 204 Id. Lone Star used Value Line as the source of the beta values, is the most widely referenced source for beta in regulatory proceedings. 2 5 Id. at 67. zo6ld. at Id. 56

8 estimate of approximately 11.2% for the Utility Proxy Group, and an implied cost of equity of 12.0% after accounting for the impact of firm size. 208 Again, analyses offered by other witnesses in this case relied on flawed assumptions or included significant errors as follows: TIEC witness Mr. Gorman's "forward-looking" CAPM approach is based on historical data and should be disregarded entirely. Mr. Gorman's "forward-looking" CAPM is based almost entirely on historical data. This approach incorrectly assumes that investors' assessment of the relative risk differences, and their required risk premium, between Treasury bonds and common stocks, is constant and equal to some historical average. When Mr. Gorman's flawed analysis is corrected, it produces a market equity risk premium of 10.6%.209 Mr. Gorman's CAPM analysis also disregards the impact of size distinctions, as measured by the average market capitalization, and fails to incorporate the market capitalization adjustments recommended by Morningstar. When this error is corrected, Mr. Gorman's proxy group shows an unadjusted CAPM result of 10.65% and an adjusted ROE of 11.33%?1 Further, incorporating projected bond yields implied unadjusted cost of equity estimates of approximately 11.1% and 10.8% for Mr. Gorman's and Mr. Hill's proxy group, and adjusted ROEs of 11.8% and 11.6%, respectively. 211 Similarly, Staff witness Ms. Winker's CAPM analysis is incorrectly based on historical data. Like Mr. Gorman's analysis, Ms. Winker's CAPM analysis is incorrectly based on historical data. Because she failed to look directly at the returns investors are currently requiring in the capital markets, her 6.86% CAPM estimates fall woefully short of investors' current required rate of return. When the CAPM analysis is correctly performed and applied to the firms in Ms. Winker's proxy group, the analysis results in an unadjusted ROE of 10.4% for the firms in her proxy group. Furthermore, Ms. Winker, like Mr. Gorman, does not include an adjustment for market capitalization.212 Incorporating such an adjustment increases the ROE to 11.2% Id. at Lone Star Ex. 26 at 99 (Rebuttal Testimony of William E. Avera) Id. at Exhibit WEA-R-8, p Id at Exhibit WEA-R-9, pp Id. at id. 57

9 Moreover, there is widespread consensus that interest rates will increase materially from current historical lows. Incorporating a forecasted Treasury bond yield for implies an unadjusted cost of equity of approximately 10.9% for the utilities in Ms. Winker's proxy group, or 11.7% after accounting for firm size. 214 Cities witness Mr. Hill provides a CAPM analysis that is also improperly based on historical data and contains errors. Mr. Hill's CAPM analysis improperly considers historical data and contains errors. 215 Moreover, the historical analysis Mr. Hill relies on is not even accurate because Mr. Hill, like Ms. Winker, calculated the equity risk premium using the total return for Morningstar's long-term government bond series. As a result, the equity risk premium falls far below what his data source reports and the resulting CAPM cost of equity estimate is understated. The market returns and equity risk premiums reported by Mr. Hill do not make economic sense in light of current capital market conditions, and in fact they contradict his own testimony.216 Mr. Hill cites an implied market risk premium of approximately 3.5%-the midpoint of his 3% to 4% range.217 But combining a market equity risk premium of 3.5% with average yield on 30-year Treasury bonds for June 2012 of 2.7% results in an indicated cost of equity for the market as a whole of 6.2%, which is 280 basis points below Mr. Hill's ROE recommendation for Lone Star in this case. When Mr. Hill's CAPM analysis is corrected, the calculation results in cost of equity estimates of 10.2% and 11.1%. Moreover, when the CAPM is applied to Mr. Hill's proxy groups based on projections published by Value Line, IHS Global Insight and Blue Chip, the result is unadjusted cost of equity estimates of approximately 10.8% for Mr. Hill's proxy group and an adjusted ROE of 11.6% Corroborating Analyses Lone Star does not address this issue. 214 Id. at Exhibit WEA-R-9. z'5 Lone Star Initial Brief at Lone Star Ex. 26 at 91 (Rebuttal Testimony of William E. Avera). 21 Cities Ex. 3 at 93 (Direct Testimony of Stephen G. Hill). 218 Lone Star Ex. 16 at Exhibit WEA-9 (Amended Direct Testimony of William E. Avera). 58

10 6. Flotation Cost Adjustment The PFD's ROE recommendation is also premised on the conclusion that there should be no flotation adjustment.219 This is contrary to the evidence Lone Star presented and should be rejected. The common equity used to finance the investment in utility assets is provided from either the sale of stock in the capital markets or from retained earnings not paid out as dividends.220 When equity is raised through the sale of common stock, there are costs associated with "floating" the new equity securities, including services such as legal, accounting, and printing, as well as the fees and discounts paid to compensate brokers for selling the stock to the public.221 In addition, "market pressure" from the additional supply of common stock and other market factors may further reduce the amount of funds that a utility nets when it issues common equity.222 These costs are recorded on the books of the utility, amortized over the life of the issue, and increase the effective cost of debt capital. However, equity flotation costs are not included in a utility's rate base because that portion of the gross proceeds achieved from the sale of common stock used to pay flotation costs is not available to invest in plant and equipment and is not capitalized as an intangible asset.223 To account for these costs, Dr. Avera appropriately recommended an upward adjustment to the cost of common equity in the amount of 15 to 45 basis points.224 The Commission should approve such an upward flotation cost adjustment. 219 PFD at Lone Star Ex. 16 at 78 (Amended Direct Testimony of William E. Avera). 221 Id. 222 Ia'. 223 Id. at ICa 59

11 D. Cost of Debt [Issue 5] 1. Phase I The Commission should reject the PFD's proposal to update Lone Star's actual cost of debt for Phase I to account for more recent Federal Reserve Board data on average swap rates as proposed by Cities, which results in a recommended cost of debt of 4.536%.225 While the ALJs determined that Cities' proposed adjustment was reasonable, there is no rational policy basis to update only one component of several included in the Company's overall Phase I long-term cost of debt of 4.794%. Lone Star's actual long-term cost of debt of 4.794% for Phase I should be approved because it is based on actual debt financing arrangements for the construction period that Lone Star recently negotiated and executed with an international consortium of lenders.226 As Lone Star argued in briefing, it is reasonable to base Lone Star's requested long-term cost of debt for Phase I rates on the Construction Bank Loan because: (i) it is currently the only outstanding source of debt for the Lone Star CREZ project, and (ii) it is the product of an extensive bid solicitation and thorough negotiation process that was recently conducted with outside lenders 227 Updating Lone Star's requested cost of debt for Phase I to account for Cities' proposed adjustment should be rejected. Lone Star appropriately used the interpolated 17-month LIBOR swap rate that was effective on the closing date for the Construction Bank Loan.228 The cost of debt has several components including issuance costs, commitment fees and the interest rate. It is arbitrary to update just one component of Lone Star's overall cost of Phase I debt after the loan was issued. 225 PFD at Lone Star Ex. 15 at 3 (Amended Direct Testimony of Aldo E. Portales). 227 Id. 228 Id. at 6. 60

12 For these reasons, Lone Star's requested cost of debt for Phase I rates of 4.794% should be approved rather than the PFD's recommended 4.536% cost of debt. 2. Phase II The Commission should approve Lone Star's requested cost of debt of 6.106% for Phase II rates and reject the PFD's recommendation to rely on a 4.536% cost of debt approved for Phase I. The ALJs agree, as Lone Star argued, that it is "virtually indisputable" that the cost of Phase II debt will be higher than that of Phase I because long-term debt is more expensive than short-term debt. 229 Nevertheless, the recommendation in the PFD is based on the conclusion that the cost of debt for which Lone Star seeks approval in Phase II rates is not known and measurable.230 The PFD's recommendation is not consistent with the record evidence and should be rejected. a. Lone Star's Phase II cost of debt is known and measurable because it is based on financing proposals for long-term debt. Lone Star's cost of debt for Phase I rates is based on debt financing arrangements for the construction period.231 In contrast, Lone Star's long-term cost of debt for Phase II rates is based on the expected interest rate the Company will achieve on a long-term permanent debt issuance planned for early 2013 that will include the retirement of the construction bank loan.232 Thus, the bases for the two separate costs of debt are distinct, which means that Phase I cost of debt should not serve as a proxy for Phase II cost of debt. The Company undertook a diligent process to determine the Phase II cost of debt that refutes the PFD's conclusion that 6.106% cost of debt for Phase II rates is not known and measurable. To determine the long-term cost of debt for Phase II rates, Lone Star requested that 229 PFD at Id. at ' Lone Star Ex. 15 at 3 (Amended Direct Testimony of Aldo E. Portales). 232 Id. 61

13 a group of experienced capital market participants familiar with the Company provide recommendations and indicative proposals of the pricing they expect for a Lone Star long-term debt issuance in early These recommendations primarily consisted of private placements of senior secured notes. 234 The proposals also provided estimates of forward treasury rates as well as Lone Star's credit spread to arrive at an interest rate.235 In addition, each proposal included an estimate of placement fees and other issuance costs based upon what is currently standard in the market.236 Lone Star took the proposal that provided the most attractive pricing (in terms of credit spreads) and added a consensus forecast of the long-term treasury rate coinciding with the expected issuance date and the period the long-term debt financing is expected to be outstanding in order to arrive at the relevant expected coupon rate.237 Finally, Lone Star divided the total annual interest expense (including amortization of issuance costs) by net proceeds in order to arrive at an "all-in" long-term cost of debt for Phase II rates. 238 Thus, the undisputed record evidence confirms that the Company's request is based on indicative proposals received from knowledgeable market participants and was determined in a manner that is unrelated to the determination of the Phase I cost of debt. b. Relying on the Phase I cost of debt ignores the distinct nature of the Phase I financing versus the Phase II financing. Just as the Intervenor recommendations to rely on the Phase I cost of debt for Phase II were unreasonable, so too, is the PFD's similar conclusion. The Phase I cost of debt is based on 233 jla at id. 235 id. 236 id. 237 jqa at 9; Lone Star Ex. 2 at WP/II-C-2.4 (Phase II) Amended (showing an average of three rate forecasts from Blue Chip Financial Forecasts, Value Line Investment Survey, and IHS Global Insight). 238 Lone Star Ex. 15 at 9 (Amended Direct Testimony of Aldo E. Portales). 62

14 the construction bank loan, and the construction period will have ended when the Phase II facilities are placed in service and the Phase II rates (including the Phase II cost of debt) are implemented. In fact, no party disputes that Lone Star plans to retire its existing construction bank loan when construction is complete and Lone Star has rates in place for its Phase II facilities.239 For this reason alone, the construction bank loan should not serve as the basis for the Phase II cost of debt because it represents a significant change in circumstances and its shortterm advantages are precisely the reason it should not be a proxy for the longer-term debt that will need to be in place when Phase II rates are implemented. After the Company energizes its Phase II facilities, Lone Star expects to refinance its existing debt and enter into a longer-term facility. 40 This means that the only appropriate cost of debt for Phase II is the rate that Lone Star expects to have to pay for that longer-term facility. The best-and only-evidence of that rate is the set of indicative rate proposals that Lone Star received from experienced capital markets participants 241 Those indicative proposals establish that a 6.106% long-term cost of debt for Phase II rates most accurately reflects Lone Star's expected long-term cost of debt,during the period Phase II rates will be in effect.242 c. Setting a cost of debt that is lower than can be reasonably expected for Phase II means Lone Star will not recover its full costs through rates. Interest rates are near historic lows, and Lone Star intends to refinance soon after the Phase II facilities are energized to lock in favorable interest rates for a 30-year period.243 Even with low interest rates, the Company knows with near certainty-and no party disputed-that the 239 Lone Star Ex. 25 at 7 (Rebuttal Testimony of Aldo E. Portales). Z4 Id. at Id. 241 Id. 242 Id. 243 Id. at 8. 63

15 true cost of long-term debt for Phase II will be higher than the Phase I cost of debt. 244 However, by imputing the Phase I cost of debt, which is based on the construction bank loan, to Phase II rates, Lone Star will not be able to recover the actual costs of financing its Phase II debt.245 One of the fundamental pillars of ratemaking is that rates should reflect the conditions under which the utility will operate at the time rates are implemented. The Commission should reject the PFD's Phase II cost of debt proposal because it violates this fundamental principle. d. Cities' witness Mr. Hill's proposed 4.536% cost of debt for Phase II is wrong, inconsistent with his own analysis and is an improper basis for the Company's Phase II cost of debt. The PFD relies on Mr. Hill's proposed cost of debt for Phase I as the basis for the Phase II cost of debt recommendation. However, the evidence shows that Mr. Hill's 4.536% cost of debt for Phase II is not reliable and should be rejected for the following reasons: The 4.536% rate does not correspond to the 30-year maturity contemplated for Lone Star's debt issuance. 246 Mr. Hill implicitly concedes that the 4.536% rate is not proper for long-term financing by calling it a "near-term estimate" of Lone Star's current cost of debt.247 The fact that actual interest rates may deviate from current forecasts is no excuse for adopting a rate that is far below the long-term debt rate that Lone Star will likely incur. z4 Mr. Hill's argument is internally inconsistent because he asserts that the future is unknowable, yet speculates about why interest rates might stay low in the future, and on that basis he urges the Commission to use a near-term rate that is far below a reasonable cost of long-term debt.249 The investment community anticipates that interest rates will move higher, and Mr. Hill concedes that "when and if the U.S. economy begins to be more robust and the 244 Lone Star Ex. 26 at 38 (Rebuttal Testimony of William E. Avera). 245 Lone Star Ex. 25 at 8 (Rebuttal Testimony of Aldo E. Portales). 24e Lone Star Ex. 26 at 38 (Rebuttal Testimony of William E. Avera). 241 Id. at d. at Id. 64

16 potential for inflation pressure increases, interest rates will probably start to rise from what have been historically low levels."250 Mr. Hill acknowledges that it is appropriate to lock in long-term rates but proposes a cost of debt that reflects only the costs of the shorter-term construction financing. 251 If the Commission adopts the PFD's recommendation and approves a lower cost of debt than Lone Star is likely to find in the capital markets, Lone Star will not be able to recover its cost of debt and would have to use part of its equity return to pay its debt costs. 252 Not only is this an unreasonable result for the Company, it will negatively affect the financial profile of the Company, possibly leading to an even higher cost of debt compared to what would be reflected in its base rates when Lone Star goes to the market in For these reasons, the Company's requested 6.106% cost of debt for Phase II should be approved. E. Overall Rate of Return [Issue 5] 1. Phase I The rate of return used to calculate the Company's requested Phase I rates is based on Lone Star's actual 48% debt and 52% equity capital structure. The appropriate cost of debt for Phase I rates is 4.79%. The Company also used a cost of equity of 11.0% as recommended by Lone Star witness Dr. Avera. Consistent with these recommendations and the supporting evidence, the Company's request for an overall rate of return of 8.02% for Phase I rates should be adopted. The overall rate of return for Phase I as recommended by the PFD should be rejected for the reasons addressed above. 211 Cities Ex. 3 at 50 (Direct Testimony of Stephen G. Hill). 25' Id. at Lone Star Ex. 25 at 9 (Rebuttal Testimony of Aldo E. Portales); Lone Star Ex. 26 at (Rebuttal Testimony of William E. Avera). 253 Lone Star Ex. 25 at 9 (Rebuttal Testimony of Aldo E. Portales). 65

17 2. Phase II The rate of return used to calculate the Company's requested Phase II rates is based on Lone Star's actual 48% debt and 52% equity capital structure. The appropriate cost of debt for Phase II rates is 6.11 %. The Company also used a cost of equity of 11.0% as recommended by Lone Star witness Dr. Avera. Consistent with these recommendations, the Company's request for an overall rate of return of 8.65% for Phase II rates should be adopted. The overall rate of return for Phase I as recommended by the PFD should be rejected for the reasons addressed above. VII.COST OF SERVICE AND O&M [ISSUES 2 AND 3] A. Operation and Maintenance Expense [Issue 151 The meager level of O&M expense the PFD suggests Lone Star should recover is another instance in which Lone Star, because it has no current rates, is being penalized for the way in which it filed its case. Indeed, the PFD recommends that Lone Star recover just over 35% of its $5.16 million in Phase I O&M expense and $6.5 million in Phase II O&M expense. The evidence establishes that the recommended level of O&M recovery is not sufficient to even pay for (1) Lone Star's existing employees, (2) required costs for communications among Lone Star's substations, control and data centers and ERCOT, (3) contractually based hardware and software maintenance and license costs, and (4) contractually based equipment maintenance required by good utility practice and NERC and ERCOT operating standards. Even though the ALJs appropriately acknowledge the "difficult task that Lone Star undertook to comb through its books and records to present its O&M (and A&G) in this case," the PFD erroneously concludes that Lone Star has not met its burden of demonstrating that its adjustments to historical test year 66

18 expenditures are known and measurable.254 The PFD also claims the Company's expenses are not reasonable and necessary compared to incumbent transmission service providers ("TSPs") and cost estimates presented in Docket No As addressed in detail below, the Company's O&M expense request should be approved in full because: The Company's requested O&M expenses are based on a historical test year ended March 31, 2012 adjusted for known and measurable changes necessary to reflect operation of the Phase I Facilities to be energized beginning in October 2012 and the Phase II Facilities when they are energized in March The known and measurable changes reflect requested Phase I rates necessary to operate and maintain Lone Star's Sam Switch and Navarro Substations, two Control Centers, two EMS facilities, its Austin Office, and its field office in Hillsboro during the first year of operations. The known and measurable changes reflect O&M expenses to be recovered through Phase II rates that are necessary to operate and maintain all of Lone Star's facilities, including more than 320 miles of transmission lines. 256 The PFD concludes that Lone Star's O&M expense should be based on Staff's calculation of the incumbent TSPs' average O&M cost of $4,808 per circuit mile, resulting in a recommended disallowance of $3,926,369 in Phase I O&M and $3,947,484 in Phase II O&M. 257 Lone Star urges the Commission to reject the PFD's O&M recommendation and instead adopt Lone Star's requested level of reasonable and necessary O&M expenses needed to provide safe and reliable electric service See PFD at p Lone Star Ex. 5 at 11, 40 (Direct Testimony of David K. Turner). 211 Id. at ' PFD at The Commission has rejected PFD recommendations on O&M expenses in prior cases. For example, in Docket No , the ALJs concluded that the utility had not met its burden of proof to include $1 million of vehicle maintenance expense in the cost of service. The Commission, however, decided that the utility had met its burden of proof. Accordingly, the Commission restored the $1 million of vehicle maintenance expense. Docket No , Order (Aug. 15, 2005). 67

19 As an alternative, the Commission could authorize Lone Star to establish a regulatory asset to track and seek future recovery of the difference between actual O&M expenses and the O&M expense level established for the Company in this proceeding. This option would: (1) allow Lone Star to operate and maintain its transmission facilities based on expense levels that the Company believes are necessary to provide safe and reliable service; (2) facilitate the development of a representative historic test year that the Commission will have the opportunity to review in the Company's next rate case, which will be filed within 16 months of placing the Phase II Facilities in service, and; (3) preserve the Company's ability to seek recovery of the incremental difference that will result from establishing rates based on an industry average instead of Lone Star's actual costs. 1. Lone Star met its burden of proof to show that its requested O&M expenses are known and measurable. No party disputes that Lone Star is a new entrant in the Texas transmission market that has not yet placed its facilities in service. Moreover, no party disputes that Lone Star must obtain a Commission-approved tariff before commencing operations. Consequently, and consistent with the Processing Agreement, the Company formulated its cost of service request based on provisions in PURA, Commission rules, and precedent that permit a utility to make "known and measurable" adjustments to historic test year data to more accurately reflect expenses the utility will incur during the time period rates will be in effect. Lone Star thoroughly analyzed its books and records and identified expenditures that would have been booked to O&M and A&G expense had the Company been operating.259 Lone Star, of course, prepared its filing in good faith consistent with the terms of the Processing Agreement, PURA, and Commission guidance. Even so, the Company finds itself in the position of having to defend 2s9 Lone Star Ex. 18 at 50 (Rebuttal Testimony of David K. Turner). 68

20 against unfair attacks that it does not have historical test year data tied to the operation of its facilities. The Commission enjoys considerable discretion to consider expenditures that occur outside the test year in order "to make the test year data as representative as possible of the cost situation `that is apt to prevail in the future."'260 The Commission has recognized two kinds of known and measurable changes to test year data: changes that are "actually realized" after the test year but before the rate year, and changes that "can be anticipated with reasonable certainty."261 Critically, these ratemaking principles allow a new entrant utility, such as Lone Star, to present its rate request consistent with Commission requirements despite the fact that it has no historical operating data associated with its CREZ facilities. Lone Star assessed its historic cost data for the period April 1, 2011 through March 31, 2012 in each functional area to determine whether its historical costs would be consistent with the expenses required to operate during Phases I and Where they were not, Lone Star made specific adjustments to account for known and measurable changes such as adjusting historical expenditures related to Third-Party O&M services out of the rate requests because the costs were not representative of services to be provided once Lone Star commences operation.263 Payroll and related expenses for both O&M and A&G also required known and measurable adjustments to annualize expenses for employees employed for only a portion of the historic period, 261 Suburban Utility Corp. v. Public Utility Commission, 652 S.W.2d 358, 366 (Tex. 1983). 26' Docket No , Order at CoLs (Aug. 15, 2005) (quoting Suburban Util. Corp., 652 S.W.2d at 362); see Federal Power Commission v. United Gas Pipe Line Co., 386 U.S. 237, 242 (1967). With regard to "anticipated" changes, the Commission has defined these as "those that will occur, can be measured, will affect future revenue requirements, and are a basis for determining forward-looking rates." Application of Oncor Electric Delivery Company LLC for Rate Case Expenses Pertaining to PUC Docket No , Docket No , Order at 3 (Sept. 21, 2009). The Commission has approved known and measurable changes for budgeted O&M expenses, O&M expenses supported by reliable test year data, labor-related expenses based on previously approved pay increases, and pension expense where adjustments are actuarially determined. Lone Star addressed this Commission precedent in detail in its Initial Brief at pages Lone Star Ex. 5 at 38 (Amended Direct Testimony of David K. Turner). 263 Id 69

21 removing expenses for terminated employees, and, adjusting pension and benefit costs accordingly.264 This approach to making known and measurable adjustments to historic data is consistent with the standard that the Commission has applied in prior rate cases, and Lone Star relied on this process to present its rate case. Despite Lone Star conducting this rigorous and thorough process, the PFD simply concludes that Lone Star's known and measurable O&M adjustments appear to be "projections." However, under FERC accounting rules, all expenditures are currently booked to capital investment because the Company is not yet operating,265 but they are the same expenses that will be incurred and classified as O&M or A&G when the Company's substations and lines are operational 266 The difference is one of accounting treatment, not substance. This is a concept the parties understood when they entered into the Processing Agreement, although the Opposing Parties did not comply with the terms of the agreement in presenting their evidence and arguments. 267 The facts are as follows: The Company's historical data was pulled directly from its books and records;268 The Company's accounting system, SAP, is a common enterprise resource planning system used in the utility business;269 The Company has presented its historical data consistent with the Commission's RFP instructions and requirements;270 Lone Star continues to incur costs that an operating utility incurs; and Id. 265 Lone Star Ex. 18 at 55 (Rebuttal Testimony of David K. Turner). 266 Id. 267 Id 261 Id. at Id. at Lone Star Ex. 17 at 7 (Amended Direct Testimony of Richard B. Cribbs). 27 Lone Star Ex. 18 at 50 (Rebuttal Testimony of David K. Turner). 70

22 Lone Star's O&M expenses are consistent with its lean staffing plan supported by shared services employees.272 The Company's O&M and A&G data is presented in the exact same manner that an operating utility would present in any rate case. 273 Nevertheless, the PFD forgoes analysis of Lone Star's proof in favor of Cities' argument that the workpapers supporting the Company's requested O&M amounts are "numbers categorized by account numbers and expense types."274 This statement unfairly simplifies the nature of the Company's evidence. The Company supplied its cost data by FERC account and specific expense categories for ease of analysis and reference. Moreover, the cost data must be considered in conjunction with the Company's testimony that explains the necessity for incurring the expenditures. The method used by Lone Star to develop its O&M and A&G expense request is also supported by the guidance Commissioner Anderson provided to Lone Star at the April 27, 2012 Open Meeting: and [A] llowing them to show known and measurable changes, which I think in this context, to the extent, for example, that outside of the test year they've already hired "X" number of employees who are likely to be permanent for O&M, to the extent that they've already made an investment, this would go into actually the capital part to the extent even though they hadn't started in whatever period, they've begun; they've ordered; they're in the process of fitting out an operation center for a control room or whatever or other facilities, then I think those are probably without prejudging known and measurable changes that we can include in the -- in the rates Lone Star Ex. 9 at 7, 11 and 12 (Amended Direct Testimony of Julie S. Rice) and Lone Star Ex. 7 at 17, 18 (Amended Direct Testimony of Cheryl L. Dietrich). 273 Lone Star Ex. 18 at 55 (Rebuttal Testimony of David K. Turner). 274 PFD at Open Meeting Tr. at 15 (Apr. 27, 2012). 71

23 [T]hey already have employees. I mean, they have some O&M now. And, again, to the extent that whatever period they use they can show known and measurable changes. 276 The integrity of Lone Star's supporting O&M data and analysis stands in stark contrast to that of the Opposing Parties. When asked at hearing by counsel for STEC and by the ALJ to identify the location of the Company's historical data in its schedules, Mr. Turner turned to Schedules II-D-1.1 and Schedules II-D Those schedules present the Company's Phase I and Phase II O&M and A&G expenditures during the historical period by month and by FERC account. 278 They also present the unadjusted amount of historical test year data.279 A cursory examination of Mr. Turner's voluminous electronic files reveals that each historical O&M and A&G expense is labeled with the Company's work breakdown structure ("WBS") Element, a description of the expense (such as "Payroll" and "Utilities"), the Account to which the expense was booked, and text further describing the expense or transaction. 280 Yet, witnesses for Staff and STEC admitted they conducted no substantive analysis of the thousands of pages in actual cost data presented by Lone Star.281 And, Cities witness Mr. Norwood admitted he was not familiar with the utility's SAP accounting software, which is widely used in the industry.282 The arguments regarding the Company's presentation of its O&M and A&G expenses are undermined by the different way in which Intervenors attack the Company's O&M and A&G 216 Id. at Tr. at (Jul. 23, 2012). 278 Lone Star Ex. 2 at 973, 975, 1193 & Id. at 973 (Column labeled "April 2011-March 2012 Total") (showing Phase I O&M); id. at 975 (Column labeled "April 2011-March 2012 Total") (showing Phase I A&G; id. at 1193 (Column labeled "April 2011-March 2012 Total") (showing Phase II O&M); id. at 1195 (Column labeled "April 2011-March 2012 Total") (showing Phase II A&G). 280 Lone Star Ex CD of Electronic Files, Amended Application Folder, Amended Schedule Workpapers Non- Confidential Folder, "WP-II-D-1 D-2 D-2.3 D-3 OM AG Schedule WP Amended (Turner) includes vol." Excel File. 281 Lone Star Ex. 18 at Exhibit DKT-R-1 (Deposition of F. John Meyer at 76); id. at Exhibit DKT-R-2 (Deposition of Mohammad Ally at 90), id. at Exhibit DKT-R-3 (Deposition of Scott Norwood at 43) (Rebuttal Testimony of David K. Turner). 282 Id. at

24 expenses, which is reflected in the PFD. The Company's requested O&M and A&G are operating expenses for which the Company seeks recovery based on a historical test year adjusted for known and measurable changes. Yet, when attacking the Company's O&M case, the Intervenors hang tightly to their argument that the Company's O&M request is nothing more than numbers in spreadsheets, disputing the accuracy of the Company's requested expenses. In contrast, when proposing adjustments to the Company's A&G request, those same parties do not question whether the costs presented are actual. Instead, they challenge the reasonableness of specific expenses and adjustments without arguing that they are not able to find the cost data in the Company's filing. This is despite the fact that the Company's O&M and A&G requests were derived from the same analysis of the Company's books and records. Further, the mountain of data and testimony supporting the reasonableness of the Company's O&M and A&G request is substantial given that the Company provided: Hundreds of pages of actual historical O&M and A&G data.283 Thousands of pages of actual historical O&M and A&G data taken directly from the Company's books and records using Lone Star's SAP accounting system complete with descriptions of each transaction, the type of cost, FERC account, date incurred, and whether each item should or should not be included in the Company's cost of service. 284 Testimony detailing the Company's O&M and A&G costs and activities, how Lone Star will use FPL to reduce O&M and A&G costs, what processes and controls are in place to ensure the reasonableness of Lone Star's O&M and A&G expenses, affiliate support, and Lone Star staffing levels.285 Testimony and exhibits responding to each proposed O&M and A&G disallowance, including RFI responses wherein the Company directed Staff and Intervenors to specific costs, schedule and workpaper information Lone Star Ex. 2 at Bates , , & Lone Star Ex. 2, CD of Electronic Files, Amended Application File, Amended Schedule Workpapers Non- Confidential File, "WP-II-D-1 D-2 D-3 OM AG Schedule WP Amended (Turner) includes vol.xlsx." 285 Lone Star Ex. 5 at (Amended Direct Testimony of David K. Turner). 286 Lone Star Ex. 18 at 1-31, 47-77, Exhibits DKT-R-16, DKT-R-17, DKT-R-19 and DKT-R-20 (Rebuttal Testimony of David K. Turner). 73