`
`
`
`UNITED STATES OF AMERICA
`BEFORE THE
`FEDERAL ENERGY REGULATORY COMMISSION
`
`Pacific Gas and Electric Company
`
`
`
`
`
`
`)
`
`
`Docket No.
`
`ER16-2320-002
`
`
`REPLY BRIEF OF
`COMMISSION TRIAL STAFF
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Washington, DC 20426
`February 12, 2021
`
`Gregory J. Feeney
`Peter V. Black
`Commission Trial Staff Counsel
`
`
`
`Docket No. ER16-2320-002
`
`TABLE OF CONTENTS
`
`I. Summary ........................................................................................................................ 1
`
`II. Argument........................................................................................................................ 4
`
`A. PG&E’s request to alter the proxy group is inconsistent with Commission
`policy and does not accurately reflect PG&E’s risk. ..................................... 5
`
`B. A natural break test should be applied to potential proxy group members... 7
`
`C. PG&E’s application of the constant growth DCF Model is inconsistent with
`Commission policy. ........................................................................................ 9
`
`D. The ECAPM model is fundamentally inconsistent with Commission policy.
`
` ..................................................................................................................... 11
`
`E. PG&E’s Risk Premium analysis erroneously includes ROE observations
`that do not meet the Commission’s criteria. ................................................ 14
`
`F. The Expected Earnings model has no place in the Commission’s ROE
`methodology. ................................................................................................ 16
`
`G. The proper measure of central tendency for PG&E is the median.............. 22
`
`H. PG&E is not entitled to an ROE in the upper third of the zone of
`reasonableness. ............................................................................................. 23
`
`I. PG&E’s purported Opinion No. 569-A-compliant ROE analysis contains
`errors. ............................................................................................................ 24
`
`III. Conclusion.................................................................................................................... 26
`
`
`
`
`
`
`
`
`
`i
`
`
`
`Docket No. ER16-2320-002
`
`TABLE OF AUTHORITIES
`
` Page(s)
`
`Cases
`
`Allegheny Power, 103 FERC ¶ 63,001 (2003), aff’d in relevant part,
`Opinion No. 469, 106 FERC ¶ 61,241 (2004), reh’g denied, Opinion
`No. 469-A, 108 FERC ¶ 61,151, vacated in part on other grounds sub
`nom. Allegheny Power v. FERC, 437 F.3d 1215 (D.C. Cir. 2006) ................................ 7
`
`Association of Busisnesses Advocating Tariff Equity v. Midcontinent
`Independent System Operator, Inc., 153 FERC ¶ 63,027 (2015) ................................ 12
`
`Association of Businesses Advocating Tariff Equity v. Midcontinent
`Independent System Operator, Inc., Opinion No. 569,
`169 FERC ¶ 61,129 (2019), order on reh’g, Opinion No. 569-A,
`171 FERC ¶ 61,154 (2020) .................................................................................... passim
`
`Bluefield Waterworks & Improvement Co. v. Public Service Commission of
`West Virginia, 262 U.S. 679 (1923)................................................................................ 5
`
`Consumers Energy Company, Opinion No. 429, 85 FERC ¶ 61,100 (1998) ...................... 7
`
`FPC v. Hope Natural Gas Co., 320 U.S. 591 (1944) .......................................................... 5
`
`Golden Spread Electric Cooperative, Inc., Opinion No. 501, 123 FERC ¶
`61,047 ........................................................................................................................ 7, 22
`
`Pacific Gas & Electric Co., 165 FERC ¶ 63,001 (2018) (Initial Decision) .................... 2, 5
`
`Pacific Gas & Electric Co., Opinion No. 572, 173 FERC ¶ 61,045 (2020)
`(Order on Initial Decision). ......................................................................... 1, 2, 6, 11, 23
`
`Transcontinental Gas Pipeline Corp., Opinion No. 414-A, 84 FERC ¶
`61,084, (1998), aff’d, Opinion 414-B, 85 FERC ¶ 61,323 ........................................... 22
`
`Secondary Sources
`
` Roger A. Morin, New Regulatory Finance (2006) ....................................................... 12
`
`
`
`ii
`
`
`
`Docket No. ER16-2320-002
`
`UNITED STATES OF AMERICA
`BEFORE THE
`FEDERAL ENERGY REGULATORY COMMISSION
`
`
`
`Pacific Gas and Electric Company
`
`
`)
`
`
`
`
`Docket No.
`
`ER16-2320-002
`
`
`REPLY BRIEF OF
`COMMISSION TRIAL STAFF
`
`Pursuant to the Commission’s October 15, 2020 Order on Initial Decision,1
`
`Commission Trial Staff (Trial Staff) submits this Reply Brief (Reply Brief) and the Reply
`
`Affidavit of Trial Staff witness Douglas M. Green (Reply Affidavit) in the above-
`
`captioned proceeding concerning whether and how to apply the Commission’s revised
`
`rate of return on equity (ROE) methodology set forth in Opinion Nos. 569 and 569-A.2
`
`I.
`The Commission directed the participants to submit initial and reply briefs that
`
`SUMMARY
`
`address whether and how to apply the revised ROE methodology adopted in Opinion
`
`
`1 Pac. Gas & Elec. Co., Opinion No. 572, 173 FERC ¶ 61,045 at P 28 (2020) (Order on
`Initial Decision); Pac. Gas & Elec. Co., “Notice of Extension of Time,” Docket No.
`ER16-2320-002 (Dec. 16, 2020).
`2 Ass’n of Bus. Advocating Tariff Equity v. Midcontinent Indep. Sys. Operator, Inc.,
`Opinion No. 569, 169 FERC ¶ 61,129 (2019), order on reh’g, Opinion No. 569-A, 171
`FERC ¶ 61,154 (2020), order on reh’g, Opinion No. 569-B, 173 FERC ¶ 61,159 (2020).
`Opinion No. 569-B was issued after the Commission’s Order on Initial Decision.
`
`
`1
`
`
`
`Docket No. ER16-2320-002
`
`Nos. 569 and 569-A.3 Initial briefs and affidavits were submitted by the participants on
`
`December 14, 2020.4 Pacific Gas and Electric Company (PG&E) proposes an ROE of
`
`13.29 percent, which is higher than the 10.4 percent originally requested by PG&E. As
`
`discussed in Trial Staff’s Initial Brief and the Initial Affidavit of Trial Staff witness
`
`Douglas M. Green (Initial Affidavit), Trial Staff’s application of Opinion Nos. 569 and
`
`569-A to the facts in this proceeding results in an ROE for PG&E of 8.89 percent.
`
`PG&E’s proposal is based on departures from Commission policy and essentially
`
`recycles the same arguments that the Commission has previously rejected.
`
`First, PG&E requests that the Commission alter the proxy group screening criteria
`
`to include companies that are not rated within one notch of PG&E by both Standard and
`
`Poor’s (S&P) and Moody’s. This would add twenty new companies to the seven-member
`
`proxy group approved by the Presiding Judge. All of these new companies have worse
`
`credit ratings than those of the companies in the Initial Decision’s proxy group,
`
`asymmetrically biasing the dataset towards overstating PG&E’s cost of equity.
`
`Second, PG&E’s analysis fails to apply the Commission-mandated natural break
`
`test on potential proxy group members. The natural break test is an indispensable
`
`safeguard to ensure that illogical or unsustainable observations do not skew the ROE
`
`analysis. It should, therefore, be applied in this proceeding.
`
`
`3 Order on Initial Decision, 173 FERC ¶ 61,045 at P 28.
`4 Trial Staff replies herein to the Initial Brief filed by Pacific Gas and Electric Company
`(PG&E).
`
`2
`
`
`
`Docket No. ER16-2320-002
`
`Third, PG&E eschews the Commission-approved two-step discounted cash flow
`
`(DCF) model, which includes an observation of long-term gross domestic product (GDP)
`
`growth forecasts (weighted at 20 percent of the overall model). Instead, PG&E
`
`illogically ignores long-term trends in economic growth, using a “constant growth” DCF
`
`model that considers only short-term growth projections. This is foreclosed by
`
`Commission precedent.
`
`Fourth, PG&E applies the lesser-known empirical capital asset pricing model
`
`(ECAPM) instead of the Commission-approved capital asset pricing model (CAPM).
`
`ECAPM is not widely accepted by financial scholars, nor has the Commission ever
`
`adopted it. The Commission should reject PG&E’s use of ECAPM and apply the CAPM
`
`it adopted in Opinion Nos. 569 and 569-A.
`
`Fifth, PG&E’s Risk Premium analysis improperly includes five ROE observations
`
`that fail the Commission’s screening criteria. These observations should be excluded.
`
`Sixth, PG&E asks the Commission to adopt the Expected Earnings model. The
`
`Commission has rejected the Expected Earnings model several times because it is not a
`
`market-based measure of the cost of equity. Moreover, PG&E has not demonstrated that
`
`investors directly rely on the Expected Earnings model when making investment
`
`decisions. The Commission should reject PG&E’s use of this methodologically unsound
`
`model.
`
`Seventh, PG&E uses the midpoint as the measure of central tendency of its
`
`computed zone of reasonableness. Commission policy requires that single-utility filers,
`
`such as PG&E, use the median as the measure of central tendency. Furthermore, the
`
`3
`
`
`
`Docket No. ER16-2320-002
`
`median is the more robust statistical measure. PG&E’s ROE should be determined using
`
`the median, not the midpoint of the zone of reasonableness.
`
`Eighth, PG&E argues that it is an above-average risk utility, and therefore it is
`
`entitled to an ROE in the upper third of the zone of reasonableness. As Trial Staff
`
`witness Green testifies, however, PG&E fails to prove that it is of above-average risk. In
`
`fact, if the Commission were to adopt PG&E’s “expanded” proxy group (which it should
`
`not), PG&E would actually be of lower risk than the average proxy group member.
`
`PG&E should, therefore, be assigned an ROE at the median of the zone of
`
`reasonableness.
`
`Finally, after arguing for a wholesale reimagination of Commission ROE policy,
`
`PG&E purports to present an ROE analysis that complies with Opinion Nos. 569 and
`
`569-A as written. As Mr. Green shows, this analysis contains errors that must be
`
`corrected.
`
`Accordingly, PG&E fails to provide sufficient evidence to justify such departures
`
`from Commission policy, and the Commission should reject these arguments for the
`
`same reasons it has rejected them in the past.
`
`II.
`While PG&E claims to have conducted an ROE analysis consistent with Opinion
`
`ARGUMENT
`
`No. 569 and 569-A, in fact they suggest extensive modifications to the Commission-
`
`approved methodology. Taken as a whole, PG&E’s preferred methodology
`
`systematically biases the analysis toward overstating the required ROE. As summarized
`
`above and further discussed below, PG&E proposes to modify the Commission’s policy
`
`4
`
`
`
`Docket No. ER16-2320-002
`
`in eight different ways. This asymmetrical bias fails to fulfill the balancing of ratepayer
`
`and shareholder interests required by the Hope5 and Bluefield6 precedents.
`
`A. PG&E’s request to alter the proxy group is inconsistent with Commission
`policy and does not accurately reflect PG&E’s risk.
`
`PG&E contends that the Commission should adopt an “expanded proxy group”
`
`containing 27 companies.7 As discussed below, this argument contravenes the
`
`Commission’s proxy group screening criteria. The Commission should instead use the
`
`seven-member proxy group adopted by the Initial Decision.8
`
`The DCF and CAPM models both require a proxy group to estimate a company’s
`
`cost of equity.9 The Commission applies several screening tests to potential members of
`
`the proxy group to exclude companies that do not accurately reflect the subject utility’s
`
`risk profile. One of these tests specifies that companies may be included in the proxy
`
`group only if they have credit ratings within one notch of the subject utility.10 In Opinion
`
`No. 569, the Commission expressly noted that it “requires use of both Standard and
`
`Poor’s corporate credit ratings and Moody’s issuer ratings when both are available.”11
`
`This means that potential proxy group members pass the credit-rating test only if both
`
`
`5 FPC v. Hope Nat. Gas Co., 320 U.S. 591 (1944) (Hope).
`6 Bluefield Waterworks & Improvement Co. v. Pub. Serv. Comm’n of W. Va., 262 U.S.
`679 (1923) (Bluefield).
`7 PG&E Initial Brief at 16.
`8 See Pac. Gas & Elec. Co., 165 FERC ¶ 63,001, at PP 54-60 (2018) (Initial Decision).
`9 Opinion No. 569, 169 FERC ¶ 61,129 at P 365.
`10 Opinion No. 569, 169 FERC ¶ 61,129 at P 365.
`11 Opinion No. 569, 169 FERC ¶ 61,129 at P 365 n.748 (emphasis added).
`
`
`5
`
`
`
`Docket No. ER16-2320-002
`
`their S&P and their Moody’s credit ratings are within one notch of the subject utility.
`
`The Commission did not alter this test in Opinion No. 569-A.
`
`Consistent with this policy, the Initial Decision correctly established a seven-
`
`member proxy group for PG&E.12 Specifically, the Initial Decision’s proxy group
`
`included only companies that were rated within one credit rating notch of PG&E during
`
`the study period by both S&P and Moody’s.13 By contrast, PG&E suggests that the
`
`Commission should include companies rated within one credit rating notch of PG&E by
`
`either S&P or Moody’s.14 PG&E’s construction of this “expanded proxy group” ignores
`
`the clear directives of Opinion Nos. 569 and 569-A. Moreover, PG&E’s deviation from
`
`Commission policy baldly “expand[s]” the proxy group to only include riskier
`
`companies. Therefore, as Trial Staff witness Green explains, this “expanded proxy
`
`group” would “produce ROE results that overstate [PG&E’s] cost of equity.”15 The
`
`Commission must reject this transparent, asymmetrical attempt to manipulate the
`
`Commission’s proxy group screening criteria to overstate its required ROE.
`
`As Mr. Green affirms, a seven-member proxy group is more than sufficient to
`
`produce reliable results, and PG&E’s concerns regarding the adequacy of the proxy
`
`group are unfounded.16 Indeed, the Commission has adopted even smaller proxy groups,
`
`
`12 Initial Decision, 165 FERC ¶ 63,001 at PP 54-60.
`13 Initial Decision, 165 FERC ¶ 63,001 at P 55.
`14 PG&E Initial Brief at 16.
`15 Trial Staff Reply Aff. ¶ 4.
`16 Trial Staff Reply Aff. ¶ 5.
`
`
`6
`
`
`
`Docket No. ER16-2320-002
`
`of as few as three or four companies.17 This shows that PG&E’s assertion that a seven-
`
`member proxy group is too small to produce a reliable estimate of PG&E’s cost of capital
`
`is unfounded, and it confirms that the Commission’s policies on proxy group composition
`
`prioritize quality over quantity.18 The Commission should apply its policy without
`
`modification, and confirm the proxy group selected in the Initial Decision and supported
`
`by Mr. Green.
`
`B. A natural break test should be applied to potential proxy group members.
`PG&E’s ROE analysis does not apply a natural break test to potential proxy group
`
`members.19 This approach is inconsistent with Commission policy.
`
`Commission policy mandates that the proxy group include only “companies whose
`
`ROE results pass threshold tests of economic logic, including both a low-end outlier test
`
`and a revised high-end outlier test . . . .”20 In addition to a numerical cutoff point, the
`
`Commission’s low-end outlier test and high-end outlier tests apply a “natural break”
`
`analysis to allow for pragmatic corrections to the proxy group composition.21 PG&E’s
`
`
`17 See, e.g., Consumers Energy Company, Opinion No. 429, 85 FERC ¶ 61,100, at 61,361
`(1998); Allegheny Power, 103 FERC ¶ 63,001, at P 23 (2003), aff’d in relevant part,
`Opinion No. 469, 106 FERC ¶ 61,241 (2004), reh’g denied, Opinion No. 469-A, 108
`FERC ¶ 61,151, vacated in part on other grounds sub nom. Allegheny Power v. FERC,
`437 F.3d 1215 (D.C. Cir. 2006).
`18 See Golden Spread Elec. Coop., Inc., Opinion No. 501, 123 FERC ¶ 61,047, at P 62
`n.126 (2008) (“While a larger group is generally desirable, the group cannot include
`companies that are not reflective of the subject companies.”).
`19 PG&E Initial Brief at 28-29.
`20 Opinion No. 569, 169 FERC ¶ 61,129 at P 365.
`21 Opinion No. 569, 169 FERC ¶ 61,129 at PP 368, 375.
`
`
`7
`
`
`
`Docket No. ER16-2320-002
`
`proposed proxy group composition ignores this clear directive, and it does so in a manner
`
`that moves the analysis in only one direction: toward companies that are riskier than
`
`PG&E and, therefore, a higher ROE.22
`
`PG&E correctly “recognizes that not applying the natural break analysis here
`
`could be construed as a departure from the approach adopted in the Combined Order [sic]
`
`Nos. 569.”23 Not only does PG&E’s proposal deviate from the express text of the
`
`Opinion Nos. 569 and 569-A in a one-sided manner, it frustrates the fundamental purpose
`
`of the Commission’s high-end outlier test. The high-end outlier test is designed to
`
`exclude unsustainably high, economically illogical, or otherwise aberrant datapoints from
`
`the proxy group.24 To do this consistently, the Commission must maintain the flexibility
`
`to make “common sense” corrections to the dataset, and cannot rely solely on rigid,
`
`mechanical computations to reveal illogical datapoints.25
`
`As Mr. Green explains, the Commission recognizes that there is always “a risk . . .
`
`that the theoretical model will fail to predict or represent the real phenomenon that is
`
`being modeled.”26 Therefore, the pragmatic check on potentially anomalous datapoints
`
`provided by the natural break test is made doubly important by Opinion No. 569-A’s
`
`
`22 See PG&E Initial Brief at 28-29.
`23 PG&E Initial Brief at 28.
`24 See Opinion No. 569, 169 FERC ¶ 61,129 at PP 376, 378.
`25 See Opinion No. 569, 169 FERC ¶ 61,129 at P 376; Opinion No. 569-A, 171 FERC ¶
`61,154 at PP 145-146.
`26 Trial Staff Reply Affidavit at ¶ 13 (quoting Opinion No. 569, 169 FERC ¶ 61,129 at P
`38).
`
`
`
`8
`
`
`
`Docket No. ER16-2320-002
`
`increase of the cutoff point of its high-end outlier test from 150 to 200 percent of the
`
`median result of all potential proxy group members before any high- or low-end outlier
`
`test is applied. By increasing the cutoff point, the risk likewise increases that an
`
`unsustainable ROE observation may be inadvertently included.27 Thus, Mr. Green
`
`testifies that the Commission was correct to retain the natural break test in Opinion Nos.
`
`569 and 569-A because without the flexibility to exclude observations that appear
`
`illogical, there is a greater risk that the final ROE results will include outlier
`
`observations.28 The Commission should uphold its policy here and continue to apply the
`
`natural break test.
`
`C. PG&E’s application of the constant growth DCF Model is inconsistent
`with Commission policy.
`
`PG&E’s ROE analysis applies the “constant growth” DCF model, which ignores
`
`long-term GDP projections and relies solely on short-term growth rates.29 The
`
`Commission’s ROE methodology, however, requires the DCF model to consider long-
`
`term GDP projections as well. PG&E’s analysis, therefore, does not comply with the
`
`Commission’s ROE policy.
`
`The DCF methodology approved by the Commission in Opinion Nos. 569 and
`
`569-A is a “two-step” DCF model. This means that the input for the expected future
`
`
`27 Trial Staff Reply Affidavit at ¶ 10 (citing Opinion No. 569-A, 171 FERC ¶ 61,154 at P
`154; Opinion No. 569-B, 173 FERC ¶ 61,159 at PP 140 and 142).
`28 Trial Staff Reply Affidavit at ¶ 13 (quoting Opinion No. 569, 169 FERC ¶ 61,129 at P
`38).
`29 PG&E Initial Brief at 29-30.
`
`
`9
`
`
`
`Docket No. ER16-2320-002
`
`growth rate of dividends (itself an input to the Commission’s DCF formula used to
`
`determine the required ROE) “is calculated using both short-term and long-term growth
`
`projections.”30 The Commission uses projected GDP growth to represent long-term
`
`growth. By including projected GDP growth in its analysis, the Commission accounts for
`
`the fact that, “over the long-term, companies cannot maintain their short-term growth
`
`rates and must, to some extent, converge on the growth rate of the overall economy.”31
`
`The Commission currently affords the short-term growth component an 80 percent
`
`weighting and the long-term growth component a 20 percent weighting.32 This nominal
`
`consideration of long-term market trends “aid[s] in normalizing any distortions that might
`
`be reflected in short-term data limited to a narrow segment of the economy.”33
`
`PG&E argues that long-term GDP projections are not relevant to the analysis, and
`
`relies solely on short-term growth rates.34 As Mr. Green testifies, PG&E’s proposed
`
`model does not faithfully reflect investor assumptions, nor is it an accurate representation
`
`of real market conditions. It is not reasonable to assume that three- to five-year growth
`
`estimates will be sustained for the 50-year time horizon used for the Commission’s two-
`
`step DCF model.35 Sophisticated investors realize this and use GDP estimates in their
`
`
`30 Opinion No. 569, 169 FERC ¶ 61,129 at P 89; Opinion No. 569-A, 171 FERC ¶ 61,154
`at PP 48, 57.
`31 Opinion No. 569-A, 171 FERC ¶ 61,154 at P 49.
`32 Opinion No. 569-A, 171 FERC ¶ 61,154 at P 57.
`33 Opinion No. 569-A, 171 FERC ¶ 61,154 at P 60 (quoting Opinion No. 531, 147 FERC
`¶ 61,234, at P 38 (2014)).
`34 PG&E Initial Brief at 29-30.
`35 Trial Staff Reply Affidavit ¶ 14.
`
`
`10
`
`
`
`Docket No. ER16-2320-002
`
`projections.36 Indeed, the underlying theory of the DCF model itself requires the use of a
`
`long-term growth projection, and recent, reputable academic work supports its use.37
`
`Thus, the Commission has repeatedly rejected the use of a constant-growth DCF model.38
`
`The Commission should do the same here.
`
`D. The ECAPM model is fundamentally inconsistent with Commission
`policy.
`
`Opinion No. 569 added the CAPM model to the Commission’s ROE
`
`methodology.39 However, PG&E’s analysis replaces the Commission-approved CAPM
`
`methodology with the “ECAPM” methodology,40 which is inconsistent with Commission
`
`policy. By contrast, Trial Staff’s analysis complied with the Commission’s instructions41
`
`and applied the CAPM methodology described in Opinion Nos. 569 and 569-A.42
`
`Because PG&E’s analysis does not use the Commission-approved CAPM model, it must
`
`be rejected.
`
`Indeed, as Trial Staff witness Green observes, the Commission has already had the
`
`opportunity to consider adopting the ECAPM model, but it has declined to do so.43
`
`
`36 Trial Staff Reply Affidavit ¶ 14.
`37 Trial Staff Reply Affidavit ¶ 14.
`38 Opinion No. 569, 169 FERC ¶ 61,129 at PP 151-159; Opinion No. 569-A, 171 FERC ¶
`61,154 at PP 56, 59-60.
`39 Opinion No. 569, 169 FERC ¶ 61,129 at P 236; aff’d, Opinion No. 569-A, 171 FERC ¶
`61,154 at PP 43-44.
`40 PG&E Initial Brief at 30-31.
`41 See Order on Initial Decision, 173 FERC ¶ 61,045 at P 28.
`42 Trial Staff Initial Brief at 12-13.
`43 Trial Staff Reply Affidavit at ¶ 15.
`
`
`11
`
`
`
`Docket No. ER16-2320-002
`
`Specifically, the Presiding Judge in the case that led to Opinion No. 551 rejected the
`
`ECAPM model, observing that it is “relied upon by no more than a few ‘financial
`
`scholars.’”44 The Presiding Judge further noted that the ECAPM model would
`
`“inevitably” increase the estimated cost of equity for proxy group companies with betas
`
`less than 1.0, and concluded that the Commission had no reason to adopt “an obscure,
`
`and arguably more controversial,” version of the CAPM model.45
`
`Mr. Green testifies that the Commission’s continued preference for the traditional
`
`CAPM model over the ECAPM model rests on sound analytical footing.46 As Mr. Green
`
`explains, PG&E witness McKenzie relies upon Dr. Roger A. Morin’s New Regulatory
`
`Finance to support his use of the ECAPM model.47 However, Dr. Morin himself
`
`acknowledged that the overwhelming majority of financially sophisticated actors use the
`
`traditional CAPM model to estimate capital costs, not the ECAPM.48 Moreover, Mr.
`
`Green finds unpersuasive Mr. McKenzie’s citations of other regulatory agencies that have
`
`used the ECAPM in the past. Mr. McKenzie failed to demonstrate that the ECAPM
`
`model allegedly used by these agencies is the same model Mr. McKenzie used here.49
`
`
`44 Ass’n of Bus. Advocating Tariff Equity v. Midcontinent Indep. Sys. Operator, Inc., 153
`FERC ¶ 63,027, at P 330 (2015).
`45 Id.
`46 Trial Staff Reply Affidavit at ¶ 15.
`47 Trial Staff Reply Affidavit at ¶ 15; Ex. PGE-0014 at 62-63; see Roger A. Morin, New
`Regulatory Finance at 442 (2006).
`48 Trial Staff Reply Affidavit at ¶ 16 (citing New Regulatory Finance at 442).
`49 Trial Staff Reply Affidavit at ¶ 17.
`
`
`12
`
`
`
`Docket No. ER16-2320-002
`
`Additionally, Mr. Green explains that the ECAPM, unlike the traditional CAPM,
`
`is not a forward looking model because it relies on an input, “alpha,” that is calculated
`
`using historic returns.50 Furthermore, the wide range of estimates of the “alpha” suggests
`
`that financial experts have been unable to confirm the impact of “alpha.”51 Mr. Green
`
`also testifies that the assumptions underpinning the ECAPM may have been overcome by
`
`events, particularly changes in the tax rates on capital gains and dividend income in 2002.
`
`Sufficient research has not been conducted since then to support the ECAPM’s current
`
`relevance. Therefore, Mr. Green concludes that the ECAPM may be obsolete, and there
`
`is insufficient evidence to support the Commission jettisoning the widely-used CAPM
`
`model in favor of a less-proven alternative.52
`
`Mr. Green notes that Mr. McKenzie’s ECAPM appears to include adjustments to
`
`inputs based on the size of the firms observed. Mr. McKenzie’s analysis, however,
`
`already includes an explicit size adjustment.53 Therefore, PG&E’s proposed ECAPM
`
`model may adjust for size twice, thereby overstating its cost of equity.54 Because the
`
`ECAPM is unproven, inconsistent with Commission policy, and methodologically
`
`unsound, the Commission should reject PG&E’s use of the model.
`
`
`50 Trial Staff Reply Affidavit at ¶ 18 (citing New Regulatory Finance at 189-92).
`51 Trial Staff Reply Affidavit at ¶ 18.
`52 Trial Staff Reply Affidavit at ¶ 18.
`53 Trial Staff Reply Affidavit at ¶ 19 (citing Ex. PGE-0120).
`54 Trial Staff Reply Affidavit at ¶ 19.
`
`
`13
`
`
`
`Docket No. ER16-2320-002
`
`Finally, PG&E advocates for the inclusion of projected bond yields in the model.55
`
`Projected bond yields, however, reflect expectations of the cost of debt in the future.56
`
`They are, therefore, not relevant to an ROE analysis based on the current cost of capital.
`
`The CAPM and Risk Premium models must use actual bond yields from the study period
`
`to estimate the current cost of equity, since those yields “fully reflect all investor
`
`expectations of the future at that time, including any projected interest rate and bond
`
`yield changes.”57 To argue that contemporaneous yields do not include all such
`
`expectations is to suggest that the market is not efficient, which defies economic logic.58
`
`Thus, the Commission has repeatedly rejected the use of projected bond yields in its
`
`CAPM and Risk Premium models.59 PG&E has not shown good cause for the
`
`Commission to revisit this well-reasoned analysis.
`
`E. PG&E’s Risk Premium analysis erroneously includes ROE observations
`that do not meet the Commission’s criteria.
`
`In addition to the CAPM model, Opinion No. 569-A also added the Risk Premium
`
`model to the Commission’s ROE methodology.60 PG&E’s attempt to apply the
`
`
`55 PG&E Initial Brief at 31.
`56 Trial Staff Reply Affidavit at ¶ 36.
`57 Trial Staff Reply Affidavit at ¶ 36.
`58 Trial Staff Reply Affidavit at ¶ 36.
`59 Opinion No. 569-B, 173 FERC ¶ 61,159 at P 121 (citing Ass’n of Bus. Advocating
`Tariff Equity v. Midcontinent Indep. Sys. Operator, Inc., Opinion No. 551, 156 FERC ¶
`61,234, at P 194 (2016)); Opinion No. 569, 169 FERC ¶ 61,129 at P 230 n.479.
`60 Opinion No. 569-A, 171 FERC ¶ 61,154 at P 104.
`
`
`14
`
`
`
`Docket No. ER16-2320-002
`
`Commission’s Risk Premium model includes several incorrect inputs.61 These inputs
`
`must be removed.
`
`The Risk Premium model rests on a simple premise: all else being equal, investors
`
`will require greater returns to invest in riskier investments.62 By observing the ROEs
`
`obtained by other market participants and comparing these ROEs to the return on the
`
`risk-free asset (i.e., U.S. Treasury bonds), one may solve for the implied “risk premium”
`
`demanded by investors.63 In the form used by the Commission, this model requires as
`
`inputs observations of “the returns on equity allowed by regulatory commissions for
`
`utilities over some past period . . . .”64 For the model to produce accurate results, these
`
`inputs must include only reliable datapoints that reflect the true equity risk premium
`
`required of utilities.
`
`In attempting to apply the Commission’s Risk Premium methodology, PG&E
`
`states that its witness, Mr. McKenzie, “include[s] seven cases [in his Risk Premium
`
`analysis] that were inadvertently omitted from Opinion No. 569-B.”65 This statement is
`
`misleading at best. As Mr. Green explains, two of these seven observations appear to be
`
`appropriately included in the Risk Premium analysis. Therefore, Mr. Green adjusted his
`
`model accordingly. This changed Trial Staff’s Risk Premium results by one basis point
`
`
`61 See PG&E Initial Brief at 31.
`62 See Opinion No. 569, 169 FERC ¶ 61,129 at PP 304-05.
`63 Opinion No. 569, 169 FERC ¶ 61,129 at P 305.
`64 See Opinion No. 569, 169 FERC ¶ 61,129 at P 305 (quoting New Regulatory Finance
`at 123).
`65 PG&E Initial Brief at 31.
`
`
`15
`
`
`
`Docket No. ER16-2320-002
`
`and had no effect on the 8.89 percent overall median ROE result.66 As Mr. Green
`
`testifies, however, the exclusion of five of the seven ROE observations was not
`
`“inadvertent”; rather, their exclusion is necessary to maintain the integrity of the model.67
`
`In three of the five observations, the observed ROE was merely adopted from an existing
`
`ROE, such as an RTO-wide ROE, “without consideration of whether that ROE would be
`
`determined to be just and reasonable under fresh analysis.”68 As for the remainder, one
`
`ROE observation resulted from a proceeding that the Commission had never set for
`
`hearing and settlement proceedings, and the other was excluded “because the ROE was
`
`set for a definite future date, and the Commission could not have evaluated a risk
`
`premium for a future date.”69 The Commission should continue to exclude from its Risk
`
`Premium model the five ROE observations identified in paragraph 21 of Mr. Green’s
`
`Reply Affidavit.70
`
`F. The Expected Earnings model has no place in the Commission’s ROE
`methodology.
`
`PG&E argues that the Commission should adopt the Expected Earnings model as
`
`part of its ROE methodology.71 PG&E’s arguments are duplicative of arguments the
`
`
`66 Trial Staff Reply Affidavit at ¶ 20.
`67 Trial Staff Reply Affidavit at ¶ 21.
`68 Trial Staff Reply Affidavit at ¶ 21 (quoting Opinion No. 569-A, 171 FERC ¶ 61,154 at
`App’x I, Cases Removed from the Risk Premium Model).
`69 Trial Staff Reply Affidavit at ¶ 21 (quoting Opinion No. 569-A, 171 FERC ¶ 61,154 at
`App’x I, Cases Removed from the Risk Premium Model).
`70 The docket numbers of these observations are: ER07-549, ER08-1548, ER08-1233,
`ER09-14, and ER08-1457. Trial Staff Reply Affidavit at ¶ 21.
`71 PG&E Initial Brief at 32-37.
`
`
`16
`
`
`
`Docket No. ER16-2320-002
`
`Commission has already considered and rejected several times. PG&E has not offered
`
`any new arguments that justify inclusion of this model.
`
`The Commission’s ROE methodology does not and should not include the
`
`Expected Earnings model. Opinion No. 569 rejected attempts to incorporate the
`
`Expected Earnings model.72 Several parties petitioned for rehearing of the Commission’s
`
`exclusion of the Expected Earnings model, “largely repeat[ing] arguments [they]
`
`previously made and which the Commission addressed in Opinion No. 569.”73 The
`
`Commission was not persuaded by the parties’ repetitive arguments, and confirmed its
`
`prior decision.74
`
`Here, PG&E argues yet again for the inclusion of the Expected Earnings model,
`
`recycling many of the same arguments the Commission has considered and rejected at
`
`least twice already.75 The Commission should affirm its prior rulings on this model once
`
`more, as PG&E has failed to provide new evidence addressing the Commission’s
`
`concerns with the Expected Earnings model.
`
`First, PG&E argues that investors rely, at least indirectly, on the Expected
`
`Earnings analysis, because rating agencies such as S&P and Moody’s consider the same
`
`factors considered by the Expected Earnings model.76 Trial Staff disagrees. As Mr.
`
`
`72 Opinion No. 569, 171 FERC ¶ 61,154 at P 200.
`73 Opinion No. 569-A, 171 FERC ¶ 61,154 at P 125.
`74 Opinion No. 569-A,
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
