United Airlines, Inc. v. FERC , 827 F.3d 122 ( 2016 )


Menu:
  •  United States Court of Appeals
    FOR THE DISTRICT OF COLUMBIA CIRCUIT
    Argued March 22, 2016                   Decided July 1, 2016
    No. 11-1479
    UNITED AIRLINES, INC., ET AL.,
    PETITIONERS
    v.
    FEDERAL ENERGY REGULATORY COMMISSION AND UNITED
    STATES OF AMERICA,
    RESPONDENTS
    BP WEST COAST PRODUCTS LLC, ET AL.,
    INTERVENORS
    Consolidated with 12-1069, 12-1070, 12-1073,
    12-1086, 15-1101, 15-1105, 15-1107
    On Petitions for Review of Orders of the
    Federal Energy Regulatory Commission
    Thomas J. Eastment argued the cause for Shipper
    Petitioners. With him on the briefs were Gregory S. Wagner,
    Richard E. Powers, Jr., Melvin Goldstein, and Steven A.
    2
    Adducci. Frederick G. Jauss and Marcus W. Sisk Jr. entered
    appearances.
    Charles F. Caldwell argued the cause for Petitioner SFPP
    L.P. With him on the briefs were Dean Lefler and Daniel W.
    Sanborn. Deborah R. Repman entered an appearance.
    Ross R. Fulton and Lisa B. Luftig, Attorneys, Federal
    Energy Regulatory Commission, argued the causes for
    respondents. On the brief were William J. Baer, Assistant
    Attorney General, U.S. Department of Justice, James J.
    Fredricks and Robert J. Wiggers, Attorneys, Robert H.
    Solomon, Solicitor, Federal Energy Regulatory Commission,
    Beth G. Pacella, Deputy Solicitor, and Elizabeth E. Rylander,
    Attorney.
    Steven A. Adducci, Thomas J. Eastment, Gregory S.
    Wagner, Richard E. Powers Jr., and Melvin Goldstein were
    on the brief for Shipper Intervenors in support of Federal
    Energy Regulatory Commission.
    Charles F. Caldwell, Dean H. Lefler, and Daniel W.
    Sanborn were on the brief for intervenor SFPP, L.P. in
    support of respondents. Elizabeth B. Kohlhausen entered an
    appearance.
    Before: GRIFFITH and KAVANAUGH, Circuit Judges, and
    SENTELLE, Senior Circuit Judge.
    Opinion for the Court filed by Senior Circuit Judge
    SENTELLE.
    SENTELLE, Senior Circuit Judge: Petitioners SFPP, L.P.
    (“SFPP”) and several shippers—“i.e., firms that pay to
    transport petroleum products over SFPP’s pipelines,”
    3
    ExxonMobil Oil Corp. v. FERC, 
    487 F.3d 945
    , 947 (D.C. Cir.
    2007)—challenge aspects of three orders from the Federal
    Energy Regulatory Commission (“FERC”) related to filings
    by SFPP for cost-of-service tariffs on its pipelines. SFPP
    disputes FERC’s choice of data for calculating SFPP’s return
    on equity and the Commission’s decision to grant only a
    partial indexed rate for the 2009 index year. The shipper-
    petitioners (the “Shippers”) claim that FERC’s tax allowance
    policy for partnership pipelines, such as SFPP, is arbitrary or
    capricious and results in unjust and unreasonable rates. We
    grant-in-part and deny-in-part SFPP’s petition and grant the
    Shippers’ petition for review.
    I.     BACKGROUND
    SFPP is a Delaware limited-partnership, common-carrier
    oil pipeline. The pipeline transports refined petroleum
    products from California, Oregon, and Texas to various
    locations throughout the southwestern and western United
    States. On June 30, 2008, SFPP filed tariffs to increase rates
    on its West Line, which transports petroleum products
    throughout California and Arizona. These new tariffs had an
    effective date of August 1, 2008. Also on June 30, 2008,
    SFPP made a separate tariff filing to decrease the rates on its
    East Line, which runs from West Texas to Arizona. The
    purported impetus for these filings was increased throughput
    on SFPP’s East Line due to a recently completed expansion,
    which accordingly decreased throughput on the West Line.
    Several shippers protested the West Line tariff filing by
    raising challenges to SFPP’s cost of service.
    On December 2, 2009, an administrative law judge issued
    an Initial Decision addressing the shippers’ arguments.
    FERC reviewed the Initial Decision in Opinion 511, 
    134 FERC ¶ 61,121
     (2011), considered a request for rehearing of
    4
    that opinion in Opinion 511-A, 
    137 FERC ¶ 61,220
     (2011),
    and then reviewed a request for rehearing of Opinion 511-A
    in Opinion 511-B, 
    150 FERC ¶ 61,096
     (2015). Both SFPP
    and the Shippers 1 petition this Court for review of these three
    FERC orders.
    SFPP makes two arguments in its petition. First, it claims
    that FERC arbitrarily or capriciously failed to utilize the most
    recently-available data when assessing its so-called real return
    on equity. Second, SFPP asserts that FERC erred when it
    declined to apply the full value of the Commission’s
    published index when setting SFPP’s rates for the 2009 index
    year. We grant SFPP’s petition with respect to the first issue
    but deny the petition with respect to the second.
    The Shippers raise a separate challenge to FERC’s current
    policy of granting to partnership pipelines an income tax
    allowance, which accounts for taxes paid by partner-investors
    that are attributable to the pipeline entity. Specifically, the
    Shippers claim that because FERC’s ratemaking methodology
    already ensures a sufficient after-tax rate of return to attract
    investment capital, and partnership pipelines otherwise do not
    incur entity-level taxes, FERC’s tax allowance policy permits
    partners in a partnership pipeline to “double recover” their
    taxes. We agree that FERC has not adequately justified its
    tax allowance policy for partnership pipelines and grant the
    Shippers’ petition.
    1
    The Shippers are: United Airlines, Inc.; Delta Air Lines,
    Inc.; Southwest Airlines Co.; US Airways, Inc.; BP West
    Coast Products LLC; Chevron Products Co.; ExxonMobil Oil
    Corporation; Valero Marketing and Supply Company; and
    Tesoro Refining and Marketing Company LLC.
    5
    II.    ANALYSIS
    Under the standard dictated by the Administrative
    Procedure Act, we will vacate FERC ratemaking decisions
    that are arbitrary or capricious. See 
    5 U.S.C. § 706
    (2)(A).
    Conversely, “FERC’s decisions will be upheld as long as the
    Commission has examined the relevant data and articulated a
    rational connection between the facts found and the choice
    made.” ExxonMobil, 
    487 F.3d at 951
    . “In reviewing FERC’s
    orders, we are ‘particularly deferential to the Commission’s
    expertise’ with respect to ratemaking issues.” 
    Id.
     (quoting
    Ass’n of Oil Pipe Lines v. FERC, 
    83 F.3d 1424
    , 1431 (D.C.
    Cir. 1996)). While we have not expressly stated whether we
    review for substantial evidence FERC’s factual findings
    within orders under the Interstate Commerce Act, “in their
    application to the requirement of factual support the
    substantial evidence test and the arbitrary or capricious test
    are one and the same.” Butte Cty. v. Hogen, 
    613 F.3d 190
    ,
    194 (D.C. Cir. 2010) (citation omitted); cf. Farmers Union
    Cent. Exch., Inc. v. FERC, 
    734 F.2d 1486
    , 1499 n.39 (D.C.
    Cir. 1984) (noting the uncertainty surrounding whether the
    substantial evidence standard applies to FERC’s ratemaking
    decisions under the Interstate Commerce Act).
    The statutory regime governing FERC’s ratemaking for
    oil pipelines is unique. In 1906, as an amendment to the
    Interstate Commerce Act (the “ICA”), Congress delegated
    regulatory authority over oil pipelines to the Interstate
    Commerce Commission. Pub. L. No. 59-337, § 1, 
    34 Stat. 584
    , 584. But in 1977, Congress transferred regulatory
    authority over oil pipelines to FERC. Department of Energy
    Organization Act, Pub. L. No. 95-91, § 402(b), 
    91 Stat. 565
    ,
    584 (1977); see also 
    49 U.S.C. § 60502
    . Congress then
    repealed the ICA in 1978 except as related to FERC’s
    regulation of oil pipelines. Pub. L. No. 95-473, § 4(c), 92
    
    6 Stat. 1337
    , 1470. For such regulation, the ICA continues to
    apply “as [it] existed on October 1, 1977 . . . .” 
    Id.
     The
    relevant provisions of the ICA were last reprinted in the
    appendix to title 49 of the 1988 edition of the United States
    Code, to which we refer as necessary. Cf. BP West Coast
    Prods., LLC v. FERC, 
    374 F.3d 1263
    , 1271 n.1 (D.C. Cir.
    2004).
    Substantively, the ICA requires that all rates be “just and
    reasonable.” 49 U.S.C. app. § 1(5)(a) (1988). Just and
    reasonable rates are “rates yielding sufficient revenue to cover
    all proper costs, including federal income taxes, plus a
    specified return on invested capital.” ExxonMobil, 
    487 F.3d at 951
     (citation omitted).
    A. FERC’S CHOICE OF DATA FOR ASSESSING SFPP’S
    REAL RETURN ON EQUITY WAS ARBITRARY OR
    CAPRICIOUS
    SFPP challenges as arbitrary or capricious FERC’s
    reliance on cost-of-equity data from September 2008 when
    calculating SFPP’s so-called “real” return on equity and the
    Commission’s rejection of more recent data from April 2009.
    FERC argues in response that the more recent cost-of-equity
    data “encompassed the stock market collapse beginning in
    late 2008,” and was therefore anomalous. FERC’s Br. 31-32.
    We agree that FERC had substantial evidence to support its
    determination that the 2009 data did not reflect SFPP’s long-
    term cost of equity. However, because the Commission
    provided no reasoned basis to justify its decision to rely on
    the September 2008 data, we hold that it engaged in arbitrary
    or capricious decision-making and therefore grant SFPP’s
    petition on this issue.
    7
    The Supreme Court stated in Federal Power Commission
    v. Hope Natural Gas Co., that “the return to the equity owner
    [of a pipeline] should be commensurate with returns on
    investments in other enterprises having corresponding risks.”
    
    320 U.S. 591
    , 603 (1944). Further, “[t]hat return . . . should
    be sufficient to assure confidence in the financial integrity of
    the enterprise, so as to maintain its credit and to attract
    capital.” 
    Id.
     In accordance with these principles, FERC uses
    a so-called “discounted cash flow” model to determine a
    pipeline’s rate of return on equity. See Composition of Proxy
    Groups for Determining Gas and Oil Pipeline Return on
    Equity, 
    123 FERC ¶ 61,048
    , at 61,271-73 ¶¶ 3-9 (2008)
    (discussing the mechanics of the discounted cash flow
    model). “The premise of the [discounted cash flow] model is
    that the price of a stock is equal to the stream of expected
    dividends, discounted to their present value.” Williston Basin
    Interstate Pipeline Co. v. FERC, 
    165 F.3d 54
    , 57 (D.C. Cir.
    1999). Under the discounted cash flow model, FERC
    “examin[es] the percentage returns on equity the market
    requires for members of a proxy group.” Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶ 242. “The members of the proxy group
    must fall with[in] a reasonable range of comparable risks and
    have publically traded securities.” 
    Id.
     Based on the stock
    prices of securities within the proxy group, FERC “calculates
    the yield (the percentage return) by dividing the dollar
    amount of the distribution by the stock price.” 
    Id. ¶ 243
    .
    After applying the distribution over the long-term, FERC
    “discount[s] back at the first year’s percentage yield to obtain
    the return on equity required to attract capital to the firm.” 
    Id.
    The resulting figure is the “nominal” return on equity.
    Under its so-called “trended original cost” methodology,
    FERC splits the nominal return on equity into an inflation
    component and the so-called “real” return on equity, defined
    as the difference between the nominal return on equity and
    8
    inflation. See Williams Pipe Line Co., 
    31 FERC ¶ 61,377
    , at
    61,833-34 (1985). While the pipeline can recover its real
    return on equity in its current annual rates, inflation “is
    written-off or amortized over the life of the property.” 
    Id. at 61,834
    ; see also Ass’n of Oil Pipe Lines, 
    83 F.3d at 1429
    .
    When assessing the pipeline’s cost structure, FERC “uses
    a ‘test year’ methodology to determine a pipeline’s annual
    cost of service.” BP West Coast, 
    374 F.3d at 1298
    . This
    method starts with a “base period” that “consist[s] of 12
    consecutive months of actual experience” with some specified
    adjustments. 
    18 C.F.R. § 346.2
    (a)(1)(i). FERC then defines a
    “test period” that generally “must consist of a base period
    adjusted for changes in revenues and costs which are known
    and are measurable with reasonable accuracy at the time of
    [rate] filing and which will become effective within nine
    months after the last month of available actual experience
    utilized in the filing.” 
    Id.
     § 346.2(a)(1)(ii). In this case,
    FERC used a base period from January 1, 2007, through
    December 31, 2007, meaning that the “nine-month
    adjustment period for test period changes [wa]s from January
    1, 2008, through September 30, 2008.” Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶ 8.
    However, for the discounted cash flow analysis, “the
    Commission prefers the most recent financial data in the
    record,” 
    id. ¶ 208
    , “because the market is always changing
    and later figures more accurately reflect current investor
    needs,” Trunkline Gas Co., 
    90 FERC ¶ 61,017
    , at 61,117
    (2000). In other words, FERC may use post-test period data
    for purposes of the discounted cash flow analysis,
    “recognizing that updates are not permitted once the record
    has been closed and the hearing has concluded.” Opinion
    511, 
    134 FERC ¶ 61,121
    , at ¶ 208.
    9
    SFPP initially submitted return-on-equity data for the six-
    month period ending with the test period, i.e., through
    September 2008.        See Exhibit SFP-1, Prepared Direct
    Testimony of J. Peter Williamson on Behalf of SFPP, L.P.,
    No. IS08-390-002, at 3-22 (FERC June 2, 2009). However,
    the pipeline later provided two updates, one for the six-month
    period ending January 2009, see Exhibit SFP-76, No. IS08-
    390-002, at 1 (FERC June 2, 2009), and one for the six-month
    period ending April 2009, see Exhibit SFP-323, No. IS08-
    390-002, at 1 (FERC June 2, 2009). From the September
    2008 data, the nominal return on equity was 12.63 percent,
    with 7.69 percent representing the real return on equity and
    4.94 percent as inflation. 2 Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 255. From the January 2009 data, the nominal
    return on equity was 14.33 percent, distributed between 14.30
    percent real return on equity and 0.03 percent inflation.
    Exhibit SFP-76, at 1. The April 2009 data showed a nominal
    return on equity of 14.09 percent with a 14.83 percent real
    return on equity and -0.74 percent inflation. Exhibit SFP-323,
    at 1. FERC also “incorporated into the . . . record” SFPP
    cost-of-equity data for the six-month periods ending in
    February 2010 and March 2010. Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶ 209 & n.339. The nominal return on equity
    from the February 2010 data was 11.24 percent, 2.14 percent
    2
    There is some ambiguity in the record regarding the
    September 2008 return on equity data. SFPP’s initial filings
    show that the nominal return on equity for this period was
    13.01 percent with 5.37 percent inflation and 7.64 percent real
    return on equity. See Exhibit SFP-1, at 21; Exhibit SFP-5,
    No. IS08-390-002, at 9 (FERC June 2, 2009); Opinion 511-A,
    
    137 FERC ¶ 61,220
    , at ¶ 252. As the exact numbers do not
    affect our holding and the parties otherwise agree that 7.69
    percent was the real return on equity for the September 2008
    period, we refer to that figure. See SFPP’s Br. 8; FERC’s Br.
    33-34.
    10
    of which was inflation with a 9.09 percent real return on
    equity. SFPP’s Br. App. A. From the March 2010 data, the
    nominal return on equity was 11.03 percent, inflation was
    2.31 percent, and the real return on equity was 8.72 percent.
    
    Id.
    SFPP argues that FERC acted arbitrarily or capriciously
    when it relied on the September 2008 data, instead of the
    April 2009 data, in setting SFPP’s real return on equity. In
    particular, SFPP contends that FERC ignored its own “policy
    of using the most recent equity rate of return data in the
    record” and provided no explanation for its choice of the
    September 2008 data. SFPP’s Br. 22-23. In FERC’s view,
    the April 2009 data is not “representative of SFPP’s cost of
    capital during the future periods the rates proposed in this
    case may be in effect.” Opinion 511, 
    134 FERC ¶ 61,121
    , at
    ¶ 209. Specifically, that data “reflects the collapse of the
    stock market in late 2008 and early 2009” and a “minimal or
    negative inflation rate” not likely to continue into the future.
    
    Id.
    We hold that it was reasonable for FERC to conclude that
    the April 2009 data was not representative of SFPP’s long-
    term cost of capital. SFPP’s argument that FERC has a
    bright-line policy of relying on the most recently available
    data to determine the real return on equity is incorrect. As
    FERC stated in Trunkline Gas Co., the Commission “seeks to
    find the most representative figures on which to base rates.”
    
    90 FERC ¶ 61,017
    , at 61,049 (emphasis added). Therefore,
    FERC “may adopt test period estimates, or it may adopt other,
    more representative figures of historical costs . . . if it
    determines that these other figures are the best, most
    representative evidence of the pipeline’s experience for the
    test period.” 
    Id.
     The real return on equity from the April
    2009 data, 14.83 percent, is the highest among each of the
    11
    periods FERC considered, and only this data includes
    negative inflation. Had FERC decided to use the April 2009
    data, SFPP would have been able to recoup essentially its
    entire nominal return on equity in its current rates, see
    Williams Pipe Line Co., 
    31 FERC ¶ 61,377
    , at 61,833-34,
    despite the fact that the February 2010 and March 2010 data
    indicated that negative inflation was a short-term
    phenomenon.       Substantial evidence therefore supported
    FERC’s finding that the April 2009 data was not the most
    representative data for assessing SFPP’s real return on equity,
    meaning that FERC did not engage in arbitrary-or-capricious
    decision-making by rejecting that data. See Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶¶ 208-09; Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶¶ 256-59.
    However, this conclusion does not end the inquiry. In lieu
    of the more recently available April 2009 data, FERC relied
    instead on the September 2008 data to fix SFPP’s real return
    on equity. See Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶ 209.
    Because we agree with SFPP that FERC provided no
    reasoned explanation for its choice of the September 2008
    data, we grant SFPP’s petition for review and vacate FERC’s
    orders with respect to this issue.
    While there may be evidence to support the conclusion
    that the nominal return on equity for September 2008 was in
    line with historical trends, this evidence does not show that
    the real return on equity for that time period was
    representative of SFPP’s costs. See Request for Rehearing of
    SFPP, L.P., No. IS08-390-002, at 11-12 (FERC Apr. 11,
    2011) (SFPP conceding that the September 2008 nominal
    return on equity is “consistent with historical periods”); see
    also Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶ 209; Opinion
    511-A, 
    137 FERC ¶ 61,220
    , at ¶¶ 252-59. To the contrary,
    FERC provides only a cursory comparison of real returns on
    12
    equity from the September 2008 through the March 2010 time
    periods, and otherwise appears to have chosen the smallest
    real return on equity from the data available. See Opinion
    511, 
    134 FERC ¶ 61,121
    , at ¶ 209. FERC was further unable
    to identify any such explanation in the record when pressed to
    do so at oral argument. See Oral Arg. Tr. 44:6-45:14. While
    “we are particularly deferential to the Commission’s expertise
    with respect to ratemaking issues,” ExxonMobil, 
    487 F.3d at 951
     (citation and internal quotation marks omitted), FERC
    cannot rely in conclusory fashion on its knowledge and
    expertise without adequate support in the record. See, e.g.,
    Int’l Union, United Mine Workers of Am. v. Mine Safety &
    Health Admin., 
    626 F.3d 84
    , 93 (D.C. Cir. 2010).
    Because we agree that FERC engaged in arbitrary-or-
    capricious decision-making by adopting the September 2008
    real return on equity without reasoned explanation, we need
    not reach SFPP’s alternative argument that FERC improperly
    rejected SFPP’s proposal to adopt an average real return on
    equity. We grant SFPP’s petition on this issue.
    B. FERC’S INDEXING ANALYSIS WAS NOT ARBITRARY
    OR CAPRICIOUS
    SFPP also argues that FERC engaged in arbitrary-or-
    capricious decision-making when it declined to apply the full
    amount of the 2009 rate index adjustment in calculating
    SFPP’s rates and refunds for the period from July 1, 2009,
    through June 30, 2010. FERC responds that it complied with
    the plain text of its regulations when it found that granting
    SFPP a full indexed rate adjustment would result in unjust
    and unreasonable rates. We agree with FERC and deny
    SFPP’s petition on this issue.
    13
    As part of the Energy Policy Act of 1992, Congress
    required FERC to “issue a final rule which establishes a
    simplified and generally applicable ratemaking methodology
    for oil pipelines in accordance with section 1(5) of part I of
    the [ICA].” Pub. L. No. 102-486, § 1801(a), 
    106 Stat. 2776
    , 3010.      Congress     also     mandated     that   “the
    Commission . . . issue a final rule to streamline procedures of
    the Commission relating to oil pipeline rates in order to avoid
    unnecessary regulatory costs and delays.” 
    Id.
     § 1802(a). In
    response, FERC released a notice of proposed rulemaking on
    July 2, 1993, which set forth an indexing scheme for setting
    oil pipeline rates. See Revisions to Oil Pipeline Regulations
    Pursuant to the Energy Policy Act of 1992; Proposed
    Rulemaking, 
    58 Fed. Reg. 37,671
    , at 37,672 (1993). FERC
    then issued on November 4, 1993, its final rule implementing
    the indexing scheme. Revisions to Oil Pipeline Regulations
    Pursuant to the Energy Policy Act of 1992, 
    58 Fed. Reg. 58,753
    , at 58,754 (1993).
    Under the final rule, FERC required that oil pipelines
    utilize the indexing system for rate changes unless specified
    circumstances permit use of an alternative methodology. 
    Id. at 58,757
    . “First, a cost-of-service showing may be utilized
    to change a rate whenever a pipeline can show that it has
    experienced uncontrollable circumstances that preclude
    recoupment of its costs through the indexing system.” Id.; see
    also 
    18 C.F.R. § 342.4
    (a). “Second, whenever a pipeline can
    secure the agreement of all existing customers, it may file a
    rate change based on such a settlement.” 58 Fed. Reg. at
    58,757; see also 
    18 C.F.R. § 342.4
    (c). Finally, FERC permits
    market-based ratemaking if the pipeline can show that it
    “lacks significant market power in the markets in question
    . . . .” 58 Fed. Reg. at 58,757; see also 
    18 C.F.R. § 342.4
    (b).
    14
    At a general level, FERC’s indexing methodology directs
    pipelines to file initial rates, usually reflecting their costs-of-
    service. 58 Fed. Reg. at 58,758. Based on the initial rate
    filings, FERC then calculates rate ceilings for future years
    based on the change in the Producer Price Index for Finished
    Goods. Id. at 58,760; see also 
    18 C.F.R. § 342.3
    (d)(2).
    Importantly, “the index establishes a ceiling on rates—it does
    not establish the rate itself.” 58 Fed. Reg. at 58,759. In other
    words, “a company is not required to charge the ceiling rate,
    and if it does not, it may adjust its rates upwards to the ceiling
    at any time during the year upon filing of the requisite
    data . . . and upon giving the appropriate notice.” Id. at
    58,761. For future years, the index “is cumulative[, meaning
    that] . . . the index applies to the applicable ceiling rate, which
    is required to be calculated each year, not to the actual rate
    charged.” Id. at 58,762. The stated purpose of this regime is
    to “preserve[] the value of just and reasonable rates in real
    economic terms [by] . . . tak[ing] into account inflation, thus
    allowing the nominal level of rates to rise in order to preserve
    their real value in real terms.” Id. at 58,759.
    In this case, SFPP filed cost-of-service rates, effective
    August 1, 2008, proposing to increase the rates charged on its
    West Line “based upon the cost of providing the service
    covered by the rate . . . .” 
    18 C.F.R. § 342.4
    (a). Because this
    rate took effect during the 2008 index year—i.e., between
    July 1, 2008, and June 30, 2009—it also “constitute[d] the
    applicable ceiling level for that index year.”              
    Id.
    § 342.3(d)(5); see also id. § 342.3(c) (defining the index year
    as “the period from July 1 to June 30”). Therefore, to
    compute the ceiling level for the 2009 index year—i.e.,
    between July 1, 2009, and June 30, 2010—SFPP
    “multipl[ied] the previous index year’s [2008’s] ceiling level
    by the most recent index published by [FERC].” Id.
    § 342.3(d)(1). The index for 2009 was 7.6025 percent.
    15
    Revisions to Oil Pipeline Regulations Pursuant to the Energy
    Policy Act of 1992, 
    127 FERC ¶ 61,184
     (2009). SFPP
    therefore contends that it has the right to apply this full index
    when calculating its 2009 rates. FERC argues that, because
    SFPP’s cost-of-service rates for 2008 already partially
    “accounted for the changes in costs associated with the index
    increase,” Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 407,
    SFPP can only apply that portion of the 2009 index “not
    reflected in the cost of service adopted by Opinion No. 511 or
    the rates SFPP must establish [in Opinion No. 511-A],” 
    id. ¶ 405
    . In particular, FERC permitted SFPP to use an index of
    1.9006 percent, “correspond[ing] to the three months of 2008
    cost changes that are outside” the period of costs already
    covered by SFPP’s proposed rates. 
    Id.
     In other words, FERC
    limited SFPP’s 2009 index to twenty-five percent of the
    published value for that index year.
    Were this information all that the Court had to consider,
    SFPP’s argument that FERC “ignore[d] its regulations, which
    have the force of law,” SFPP’s Br. 35, might be plausible in
    light of the plain text of FERC’s indexing regulations, see 
    18 C.F.R. § 342.3
    . But the analysis is only half-complete.
    “[M]erely because the Commission regulations permit SFPP
    to request the index increase does not mean that the
    Commission is bound to accept the indexed rate increase.”
    Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 407. In particular,
    “persons with a substantial economic interest in the tariff
    filing may file a protest to a tariff filing pursuant to the
    Interstate Commerce Act.” 
    18 C.F.R. § 343.2
    (b). A protest
    to a proposed rate under 
    18 C.F.R. § 343.2
     must allege
    “reasonable grounds for asserting that the rate violates the
    applicable ceiling level, or that the rate increase is so
    substantially in excess of the actual cost increases incurred by
    the carrier that the rate is unjust and unreasonable, or that the
    rate decrease is so substantially less than the actual cost
    16
    decrease incurred by the carrier that the rate is unjust and
    unreasonable.” 
    Id.
     § 343.2(c)(1). In this case, the Shippers
    did file protests to SFPP’s indexed rates for the 2009 index
    year. See Protest and Comments of Chevron Products
    Company, ConocoPhillips Company, Continental Airlines,
    Inc., Northwest Airlines, Inc., Southwest Airlines Co., US
    Airways, Inc., and Valero Marketing and Supply Company on
    SFPP, L.P. Compliance Filing (“Protest I”), Nos. IS08-390-
    002, IS08-390-006, IS11-338-000 (FERC June 15, 2011);
    Protest of ExxonMobil Oil Corporation and BP West Coast
    Products LLC of Compliance Filing Implementing Opinion
    No. 511 (“Protest II”), No. IS08-390-006 (FERC June 15,
    2011). Therein, they argued that because “[t]he 2009 index is
    based on [FERC’s] computation of industry-wide cost
    increases between 2007 and 2008[,]” SFPP should not be
    permitted to double-recover its costs by combining its 2008
    cost-of-service rates with proposed 2009 indexed rates.
    Protest II, at 12. Equivalently, the Shippers alleged that
    SFPP’s 2009 indexed rate increase was “substantially in
    excess of the actual cost increases incurred by [SFPP]” during
    2008. 
    18 C.F.R. § 343.2
    (c)(1). FERC agreed. See Opinion
    511-A, 
    137 FERC ¶ 61,220
    , at ¶ 411; Opinion 511-B, 
    150 FERC ¶ 61,096
    , at ¶¶ 27-33. “Because the subject of our
    scrutiny is a ratemaking—and thus an agency decision
    involving complex industry analyses and difficult policy
    choices—the court will be particularly deferential to the
    Commission’s expertise.” Ass’n of Oil Pipe Lines, 
    83 F.3d at 1431
    . With this principle in mind, we discern no error in
    FERC’s decision-making.
    SFPP’s principal retort to this otherwise straightforward
    application of FERC’s regulations is that the alleged purpose
    of FERC’s indexing procedures is to permit a pipeline to
    capture future inflation-based cost adjustments, not prior-year
    cost-of-service changes.        FERC responds, somewhat
    17
    cryptically, that indexing “allows rates to track inflation in the
    general economy, essentially preserving pipelines’ existing
    rates in real economic terms.” FERC’s Br. 43.
    SFPP’s argument is irrelevant to this case. Admittedly,
    whether FERC’s indexing mechanism is retrospective or
    prospective is unclear. For example, FERC has previously
    described the purpose of indexing as “preserv[ing] the value
    of just and reasonable rates in real economic terms . . . [by]
    tak[ing] into account inflation, thus allowing the nominal
    level of rates to rise in order to preserve their real value in
    real terms.” Revisions to Oil Pipeline Regulations Pursuant
    to the Energy Policy Act of 1992, 58 Fed. Reg. at 58,759; see
    also SFPP, L.P., 
    117 FERC ¶ 61,271
    , at 62,337 (2006). By
    contrast, we have stated that indexing “enable[s] pipelines to
    recover costs by allowing pipelines to raise rates at the same
    pace as they are predicted to experience cost increases.”
    Ass’n of Oil Pipe Lines, 
    83 F.3d at 1430
    . However, once a
    party files a protest to a pipeline’s proposed rates, FERC’s
    regulations state that the Commission will compare the
    “actual cost increases incurred by the carrier” with the
    proposed rate increase. 
    18 C.F.R. § 343.2
    (c)(1) (emphasis
    added). FERC made this comparison when it noted that SFPP
    would effectively double-recover its 2008 costs were it to
    receive the full 2009 index. See Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶¶ 409-11; Opinion 511-B, 
    150 FERC ¶ 61,096
    ,
    at ¶¶ 27-33. While admittedly FERC’s analysis was less
    quantitative than in prior rate proceedings, we hold that FERC
    provided sufficient justification for its decision to reduce
    SFPP’s 2009 index to one-quarter of the published value. See
    Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 411 n.687; SFPP,
    L.P., 
    135 FERC ¶ 61,274
    , at 62,513 ¶¶ 11-12 (2011)
    (describing the so-called “percentage comparison test”); see
    also SFPP, L.P., 
    117 FERC ¶ 61,271
    , at 62,337 ¶ 5 (denying
    indexed rate increase to SFPP’s East Line rates where base
    18
    rates already “recover[ed] all the relevant operating and
    capital costs”).
    SFPP’s reliance on prior FERC proceedings involving
    indexing, see, e.g., Opinion 435, 
    86 FERC ¶ 61,022
    , at 61,085
    (2000); Opinion 435-A, 
    91 FERC ¶ 61,135
    , at 61,516 (2000),
    is inapposite. As SFPP admitted during oral argument, those
    proceedings at most permitted FERC to apply the full index to
    SFPP’s rates but did not compel it. See Oral Arg. Tr. 22:12-
    :15. Notably, FERC did not address in those cases whether
    the indexed rates were “so substantially in excess of the actual
    cost increases incurred by the carrier,” 
    18 C.F.R. § 343.2
    (c)(1), which it has done here. We otherwise agree
    with FERC that SFPP “has failed to demonstrate that
    [FERC’s] determination . . . is inconsistent with precedent.”
    FERC’s Br. 48.
    We therefore deny SFPP’s petition on this issue.
    C. FERC MUST DEMONSTRATE THAT THERE IS NO
    DOUBLE RECOVERY OF TAXES FOR PARTNERSHIP
    PIPELINES
    The Shippers argue that FERC engaged in arbitrary-or-
    capricious decision-making when it granted an income tax
    allowance to SFPP. Specifically, the Shippers note that, as a
    partnership pipeline, SFPP is not taxed at the pipeline level.
    Because FERC’s discounted cash flow return on equity
    already ensures a sufficient after-tax return to attract
    investment to the pipeline, they argue, the tax allowance
    results in “double recovery” of taxes to SFPP’s partners. In
    FERC’s view, we already decided this issue in ExxonMobil,
    where we held that FERC’s policy of permitting partnership
    pipelines to receive a tax allowance was “not unreasonable”
    in light of “FERC’s expert judgment about the best way to
    19
    equalize after-tax returns for partnerships and corporations.”
    
    487 F.3d at 953
    . FERC therefore posits that the Shippers’
    petition in this case is an impermissible collateral attack on
    our decision in ExxonMobil. Further, FERC denies that
    granting a tax allowance to SFPP results in a double-recovery
    of taxes and avers that any disparity in after-tax returns to
    partners or shareholders arises from the Internal Revenue
    Code, not from FERC’s tax allowance policy. Because we
    reserved the issue of whether the combination of the
    discounted cash flow return on equity and the tax allowance
    results in double recovery of taxes for partnership pipelines,
    we disagree with FERC’s collateral attack argument.
    Nonetheless, we acknowledge that our opinion in ExxonMobil
    stated that it may be reasonable for FERC to grant a tax
    allowance to partnership pipelines. However, because FERC
    failed to demonstrate that there is no double-recovery of taxes
    for partnership, as opposed to corporate, pipelines, we hold
    that FERC acted arbitrarily or capriciously. We therefore
    grant the Shippers’ petition.
    As all parties acknowledge, this case is not the first time
    that we have considered FERC’s tax allowance policy for oil
    pipelines. Until our decision in BP West Coast, 
    374 F.3d at 1293
    , FERC relied on the so-called Lakehead policy when
    granting tax allowances. Named for the FERC decision in
    which the Commission formalized the policy, see Lakehead
    Pipe Line Co., 
    71 FERC ¶ 61,338
    , at 62,314-15 (1995), the
    Lakehead policy addressed the situation in which a
    partnership pipeline has both corporate-partners and
    individual-partners. FERC therein concluded:
    20
    When partnership interests are held by corporations,
    the partnership is entitled to a tax allowance in its cost-
    of-service for those corporate interests because the tax
    cost will be passed on to the corporate owners who
    must pay corporate income taxes on their allocated
    share     of     income     directly     on   their     tax
    returns. . . . However, the Commission concludes that
    Lakehead should not receive an income tax allowance
    with respect to income attributable to the limited
    partnership interests held by individuals. This is
    because those individuals do not pay a corporate
    income tax.
    
    Id.
    We reviewed the Lakehead policy in BP West Coast and
    held that “[w]e cannot conclude that FERC’s inclusion of the
    income tax allowance in SFPP’s rates is the product of
    reasoned decisionmaking.” 
    374 F.3d at 1288
    . In that case,
    we started from the principle “that the regulating commission
    is to set rates in such a fashion that the regulated entity yields
    returns for its investors commensurate with returns expected
    from an enterprise of like risks.” 
    Id. at 1290
    . Consistent with
    this principle, we rejected FERC’s justifications for its
    Lakehead policy and held that “where there is no tax
    generated by the regulated entity, either standing alone or as
    part of a consolidated corporate group, the regulator cannot
    create a phantom tax in order to create an allowance to pass
    through to the rate payer.” 
    Id. at 1291
    .
    Concededly, our use of the term “phantom tax” in BP
    West Coast lacked precision. This was made apparent in
    ExxonMobil, as several shipper-petitioners challenged
    FERC’s revised tax allowance policy, which granted a full
    income tax allowance to both partnership pipelines and
    21
    corporate pipelines, regardless of the identities of the partners
    or shareholders. 
    487 F.3d at 950
    . We rejected the
    petitioners’ arguments in that case, stating that because
    “investors in a limited partnership are required to pay tax on
    their distributive shares of the partnership income, even if
    they do not receive a cash distribution[,] . . . the income
    received from a limited partnership should be allocated to the
    pipeline and included in the regulated entity’s cost-of-
    service.” 
    Id. at 954
    . FERC did not create a “phantom tax”
    because it did not arbitrarily distinguish between corporate
    and individual partners in a partnership pipeline, and the
    Commission adequately explained why partner taxes could be
    considered a pipeline cost.
    In this case, the Shippers challenge the same tax
    allowance policy at issue in ExxonMobil. Given that nothing
    has changed with regard to this policy, FERC’s argument that
    the Shippers present an impermissible collateral attack to our
    ExxonMobil decision is, on first consideration, conceivable.
    However, as the Shippers mention in their reply brief, FERC
    averred during briefing in ExxonMobil that it was addressing
    the double recovery issue in a separate proceeding. See Br. of
    Resp’t at 30-31, ExxonMobil Oil Corp. v. FERC, 
    487 F.3d 945
     (D.C. Cir. 2007) (Nos. 04-1102 et al.). While we did not
    expressly reserve the issue in our ExxonMobil opinion, the
    fact that FERC took this position both in ExxonMobil and in
    an accompanying case, see Br. of Resp’t at 29-30, Canadian
    Ass’n of Petroleum Producers v. FERC, 
    487 F.3d 973
     (D.C.
    Cir. 2007) (No. 05-1382), reflects our implicit reservation of
    the question. To clarify, we held in ExxonMobil that, to the
    extent FERC has a reasoned basis for granting a tax
    allowance to partnership pipelines, it may do so. 487 F.3d at
    955. The Shippers now challenge whether such a reasoned
    basis exists based on grounds that FERC agreed were not at
    22
    issue in the prior case. We therefore hold that the Shippers’
    petition is not a collateral attack on that decision.
    As to the merits, we hold that FERC has not provided
    sufficient justification for its conclusion that there is no
    double recovery of taxes for partnership pipelines receiving a
    tax allowance in addition to the discounted cash flow return
    on equity. Despite their attempts to inundate the record with
    competing mathematical analyses of whether a double
    recovery of taxes for partnership pipelines exists, the parties
    do not disagree on the essential facts. First, unlike a
    corporate pipeline, a partnership pipeline incurs no taxes,
    except those imputed from its partners, at the entity level. See
    
    26 U.S.C. § 7704
    (d)(1)(E). Second, the discounted cash flow
    return on equity determines the pre-tax investor return
    required to attract investment, irrespective of whether the
    regulated entity is a partnership or a corporate pipeline. See
    Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶¶ 243-44; Shippers’ Br.
    6; see also supra Part II.A (discussing the mechanics of the
    discounted cash flow methodology). Third, with a tax
    allowance, a partner in a partnership pipeline will receive a
    higher after-tax return than a shareholder in a corporate
    pipeline, at least in the short term before adjustments can
    occur in the investment market. See FERC’s Br. 29;
    Shippers’ Br. 34-35; Oral Arg. Tr. 32:17-33:2.
    These facts support the conclusion that granting a tax
    allowance to partnership pipelines results in inequitable
    returns for partners in those pipelines as compared to
    shareholders in corporate pipelines. Because the Supreme
    Court has instructed that “the return to the equity owner
    should be commensurate with returns on investments in other
    enterprises having corresponding risks,” FERC has not shown
    that the resulting rates under FERC’s current policy are “just
    and reasonable.” Hope Nat. Gas Co., 
    320 U.S. at 603
    . 
    FERC 23
    attempts to circumvent this deduction by arguing, first, that
    there is no “gross-up” in the return rate for partnership
    pipelines to account for income taxes, and, second, that any
    disparate treatment between partners in partnership pipelines
    and shareholders in corporate pipelines is the result of the
    Internal Revenue Code, not FERC’s tax allowance policy.
    These arguments, which are two sides of the same
    metaphorical coin, are not persuasive.
    The crux of FERC’s “gross-up” theory is that “in the
    context of Commission rate design[,]” Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 290, “the Commission does not gross up
    a jurisdictional entity’s operating revenues or return to cover
    the income taxes that must be paid to obtain its after-tax
    return,” 
    id. ¶ 280
    . What the Commission apparently means
    by this rather obscure statement is that it imputes the income
    taxes paid by partners in a partnership pipeline to the pipeline
    itself, meaning that an income tax allowance is then necessary
    to equalize the after-tax “entity-level” rates of return for
    partnership and corporate pipelines. See Opinion 511, 
    134 FERC ¶ 61,121
    , at ¶¶ 241-50; see also Opinion 511-A, 
    137 FERC ¶ 61,220
    , at ¶ 319. Of course, when one then considers
    the after-tax returns to partners or shareholders, the necessary
    conclusion is that partners in a partnership pipeline receive a
    windfall compared to shareholders in a corporate pipeline, a
    point which FERC concedes. See FERC’s Br. at 29; Oral
    Arg. Tr. 32:17-33:2. FERC, in a form of Orwellian
    doublethink, attributes this disparity in returns to the Internal
    Revenue Code while simultaneously denying that double-
    recovery exists. See Opinion 511-A, 
    137 FERC ¶ 61,220
    , at
    ¶ 315.
    True, FERC has a justifiable basis for its attribution of
    partner taxes to the partnership pipeline. In ExxonMobil, we
    acknowledged that “investors in a limited partnership are
    24
    required to pay tax on their distributive shares of the
    partnership income, even if they do not receive a cash
    distribution.” 487 F.3d at 954. By contrast, “a shareholder of
    a corporation is generally taxed on the amount of the cash
    dividend actually received.” Id. For this reason, allocation of
    partner-level taxes to a partnership pipeline may not result in
    a “phantom tax” of the type we rejected in BP West Coast.
    However, our holding in ExxonMobil did not absolve FERC
    of its obligation to ensure “commensurate . . . returns on
    investments” for “equity owner[s]” as required under Hope
    Natural Gas, 
    320 U.S. at 603
    . Even if FERC elects to impute
    partner taxes to the partnership pipeline entity, it must still
    ensure parity between equity owners in partnership and
    corporate pipelines. FERC’s failure to do so in this case is
    therefore arbitrary or capricious.
    The remaining issue is the appropriate remedy. The
    Shippers do not request that we overturn our decision in
    ExxonMobil, which we are unable to do in any case absent an
    en banc decision from the Court. See LaShawn A. v. Barry,
    
    87 F.3d 1389
    , 1395 (D.C. Cir. 1996). But we also believe
    such action is unnecessary. When questioned at oral
    argument, FERC conceded that it might be able to remove
    any duplicative tax recovery for partnership pipelines directly
    from the discounted cash flow return on equity. See Oral
    Arg. Tr. 36:3-:10. We note also that, prior to ExxonMobil,
    FERC considered the possibility of eliminating all income tax
    allowances and setting rates based on pre-tax returns. See
    Policy Statement on Income Tax Allowances, 
    111 FERC ¶ 61,139
    , at 61,741 (2005). To the extent that FERC can
    provide a reasoned basis for such a policy, we do not read our
    decision in ExxonMobil as foreclosing that option. See 487
    F.3d at 955 (“Arguably, a fair return on equity might have
    been afforded if FERC had chosen the fourth alternative of
    computing return on pretax income and providing no tax
    25
    allowance at all for the pipeline owners.”). We therefore
    grant the Shippers’ petition, vacate FERC’s orders with
    respect to this issue, and remand for FERC to consider these
    or other mechanisms for which the Commission can
    demonstrate that there is no double recovery.
    III.   CONCLUSION
    For the reasons stated herein, the Court: (i) grants-in-part
    SFPP’s petition with respect to the choice of data for
    assessing SFPP’s real return on equity, vacates FERC’s
    orders accordingly, and remands for further proceedings
    consistent with this opinion; (ii) denies-in-part SFPP’s
    petition with respect to the indexing issue; and (iii) grants the
    Shippers’ petition, vacates FERC’s orders with respect to the
    double recovery issue, and remands to FERC for further
    proceedings consistent with this opinion.
    So ordered.