Commissioner Allison Clements Statement
March 17, 2022
Docket No. ER16-2320-002
As an initial matter, I agree that today’s order reasonably applies the Commission’s return on equity (ROE) policy established in Opinion No. 569-A[1] to the facts in this proceeding. I support the finding that Pacific Gas and Electric Company was an average risk utility during the rate period in question. I also support the view that using Bloomberg-based alternative betas is a reasonable method of addressing the imperfect correspondence between the CAPM risk premium calculation, which uses data from the S&P 500, and Value Line betas, which are derived from the NYSE.
However, I dissent in part from today’s order because of my continuing concerns with the current ROE policy, which I believe applies a flawed methodology that does not adequately protect consumers and does not yield just and reasonable rates. I explained my concerns with our current ROE policy in a prior dissent,[2] so I will not repeat them here. However, I will raise two specific and significant points.
First, I am troubled by the remarkable lack of parity between the treatment of proxy companies on the high- and low-ends of the spectrum under the high- and low-end outlier tests.
Opinion No. 569 adopted a high-end outlier test that would exclude extreme observations by removing any proxy company whose cost of equity estimated under the two-step DCF and CAPM models is more than 150 percent of the median result of all the potential proxy group members in the models, subject to a natural-break analysis.[3] Opinion No. 569-A relaxed this 150 percent threshold to more than 200 percent of the median result of all the potential proxy group members in each model, subject to a natural-break analysis.[4]
In Opinion No. 569-A, the majority asserted that it was necessary to relax the high-end outlier threshold from 150 to 200 percent to “reduce the risk that [] rational results are inappropriately excluded.”[5] But if this were truly the case, the rationale should have applied equally to the low-end outlier test.
As today’s record reflects, it does not. Instead, the high-end outlier screen identifies no outliers,[6] and a 634 basis point gap exists between the highest DCF result in the proxy group and the high-end outlier screen.[7] The low-end outlier screen, on the other hand, removes two companies from the proxy group. The low-end companies’ “outlier” DCF results are just 102 and 62 basis points below the low-end outlier screen.[8]
Opinion No. 569-A’s stated intent to guard against the inappropriate exclusion of rational results applies asymmetrically to high-end and low-end results. Indeed, while the low-end outlier test has consistently removed low-end “outliers” from proxy groups, the high-end outlier test has yet to identify even a single result in any proceeding to be too high.[9]
Second, I agree with Joint Parties that there have been, from the very start, “serious questions raised about the soundness of the size premium adjustment” that lead inexorably to the conclusion that applying a size adjustment to the CAPM analysis is neither appropriate nor supported.[10]
Today’s order relies on the mantra—first articulated in Opinion No. 531-B and again in Opinion Nos. 569 and 569-A—that the size premium adjustment is “a generally accepted approach to CAPM analyses.”[11] The support for this premise is thin.[12]
In fact, even if a size premium adjustment to the CAPM were to make sense generally, a size effect does not appear to exist in context of the utility industry.[13] Industry practice, the facts, and the academic literature—including the text on which Opinion No. 531-B relies for support—caution against applying a size premium adjustment to the CAPM analysis.[14]
Given my skepticism that the current methodology reflects a reasonable approach that balances utility and consumer interests, I cannot conclude that the resulting ROE in this proceeding is just and reasonable. For these reasons, I respectfully concur in part and dissent in part.
[1] Ass’n of Bus. Advocating Tariff Equity v. Midcontinent Indep. Sys. Operator, Inc., Opinion No. 569-A, 171 FERC ¶ 61,154 (2020).
[2] Entergy Arkansas, Inc., 175 FERC ¶ 61,136 (2021) (Clements, Comm’r, dissenting); see also Constellation Mystic Power, LLC, 177 FERC ¶ 61,106 (2021) (Clements, Comm’r, dissenting), order on reh’g, 178 FERC ¶ 61,116 (2022) (Clements, Comm’r, dissenting).
[3] Ass’n of Bus. Advocating Tariff Equity v. Midcontinent Indep. Sys. Operator, Inc., Opinion No. 569, 169 FERC ¶ 61,129, at PP 375, 533 (2019).
[4] Opinion No. 569-A, 171 FERC ¶ 61,154 at P 154.
[5] Id.
[6] Pacific Gas and Electric Company, 178 FERC ¶ 61,175, App. (2022).
[7] Id.
[8] Id.
[9] See e.g. Mystic Power, 177 FERC ¶ 61,106 at App. (three low-end companies removed from DCF proxy group based on low-end outlier test, no high-end companies removed from DCF proxy group based on high-end outlier test (435 basis point gap between high-end threshold and highest DCF result)); Mystic Power, 176 FERC
¶ 61,019 at App. (no low- or high-end outliers removed from CAPM proxy group (158 basis point gap between low-end threshold and lowest CAPM result, 808 basis point gap between high-end threshold and highest CAPM result)); DATC Path 15, LLC, 177 FERC ¶ 61,115, App. (2021) (one low-end company removed from DCF proxy group based on low-end outlier test, no high-end companies removed from DCF proxy group based on high-end outlier test (712 basis point gap between high-end threshold and highest DCF result), no low- or high-end outliers removed from CAPM proxy group (50 basis point gap between low-end threshold and lowest CAPM result, 786 basis point gap between high-end threshold and highest CAPM result)); Entergy Arkansas, 175 FERC ¶ 61,136 at App. (four low-end companies removed from DCF proxy group based on low-end outlier test, no high-end companies removed from DCF proxy group based on high-end outlier test (667 basis point gap between high-end threshold and highest DCF result), no low- or high-end outliers removed from CAPM proxy group (193 basis point gap between low-end threshold and lowest CAPM result, 876 basis point gap between high-end threshold and highest CAPM result)).
[10] Joint Parties Initial Br. 26 (quoting Ex. CIT-0102 (Testimony of Breandan T. Mac Mathuna) at 41: 11-15).
[11] Opinion No. 569-A, 171 FERC ¶ 61,154 at P 61 (citing Opinion No. 569, 169 FERC ¶ 61,129 at P 296); Opinion No. 569, 169 FERC ¶ 61,129 at P 296 (citing Coakley v. Bangor Hydro-Elec. Co., Opinion No. 531-B, 150 FERC ¶ 61,165, at P 117 (2015)); Opinion No. 531-B, 150 FERC ¶ 61,165 at P 117 (citing Roger A. Morin, New Regulatory Finance 187 (2006) (New Regulatory Finance)).
[12] As support, Opinion No. 531-B relies on New Regulatory Finance, which states in relevant part, “Given the evidence of a small firm premium … the CAPM understates the risk of smaller utilities, and a cost of equity based purely on a CAPM beta will therefore produce too low an estimate for those small companies. This has led some analysts to add a premium to the estimated cost of equity for smaller companies.” However, on the very same page of New Regulatory Finance, Dr. Morin also cautions against the use of the size premium adjustment in the context of the regulated utility industry and suggests a superior alternative. Joint Parties Initial Brief at 26 (citing New Regulatory Finance 187 & n.8 (cautioning that applying size premium adjustment “opens the door to a whole series of adjustments” and thus “a superior alternative to considering the size premium explicitly is to identify the economic reasons for the premium and develop more direct measures of risk.”)). Joint Parties also note that “Dr. Morin himself did not employ the size adjustment in recently-filed testimony filed at FERC, which calls into question whether he ‘endorses’ the need for a size premium adjustment in the CAPM analysis.” Joint Parties Initial Br. 26 & n.114 (internal quotation omitted) (referring to Fifth Transmission Owner Formula Rate Tariff Filing, San Diego Gas & Elec., Docket No. ER19-221-000 (filed Oct. 30, 2018) – Prepared Direct Testimony of Dr. Roger A. Morin, Ex. SD-0019).
[13] Joint Parties Initial Br. at 26 (citing Ex. CIT-0102 at 38:10-19); see also Opinion No. 569, 169 FERC ¶ 61,129 at PP 285, 296, 303 (Noting study finding “that there is no significant difference and there is no discernible pattern of increase among the risk indicators of publicly-traded electric utility of different sizes[,]” and two “regression analyses demonstrating that a size premium adjustment is not appropriate for the CAPM” but nonetheless continuing to find such adjustment to be “a generally accepted approach to CAPM analyses.”)).
[14] Joint Parties Initial Br. at 25-27 (citing Ex. CIT-0102 at 34-41); see also Ex. CIT-0102 at 34-41 (citing Clifford S. Ang, Why We Shouldn’t Add a Size Premium to the CAPM Cost of Equity, National Ass’n of Certified Valuators and Analysts (Feb. 15, 2017) (“[s]ince the mid-1980’s, however, there has been no size premium after adjusting for market risk”), John H. Cochrane, Asset Pricing 452 (Princeton Univ. Press, rev. ed. 2005) (noting decline of anomalous risk premia have been in decline, with the small-firm effect disappearing in 1980), Dr. Aswath Damodaran, The Small Cap Premium: Where’s the Beef?, Musings on Markets (Apr. 11, 2015) (the premium “seems to have dissipated since 1981”), Annie Wong, Utility Stocks and the Size Effect: An Empirical Analysis, Journal of the Midwest Fin. Ass’n 96-101 (1993) (concluding “there is no need to adjust the firm size in utility rate regulation” and noting “weak evidence that firms size is a missing factor from the CAPM”), and New Regulatory Finance 187 & n.8 (cautioning that applying size premium adjustment “opens the door to a whole series of adjustments” and thus “a superior alternative to considering the size premium explicitly is to identify the economic reasons for the premium and develop more direct measures of risk.”))).