# Five-Year Review of the Oil Pipeline Index

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A2022-01544

## Record

- **Collection:** Federal Register
- **Document type:** Rule
- **Published:** January 28, 2022
- **Citation:** 87 FR 4476

## Text

DEPARTMENT OF ENERGY
Federal Energy Regulatory Commission
18 CFR Part 342
[Docket No. RM20-14-001]
Five-Year Review of the Oil Pipeline Index

AGENCY:

Federal Energy Regulatory Commission.

ACTION:

Order on rehearing.

SUMMARY:

The Federal Energy Regulatory Commission (Commission) addresses arguments raised on rehearing of the December 17, 2020 Order Establishing Index Level concluding the Commission's five-year review of the index level used to determine annual changes to oil pipeline rate ceilings (December 2020 Order). The December 2020 Order established an index level of Producer Price Index for Finished Goods plus 0.78% (PPI-FG+0.78%) for the five-year period commencing July 1, 2021. In this order, the Commission grants rehearing of the December 2020 Order, in part, denies rehearing, in part, and establishes an index level of PPI-FG-0.21%.

DATES:

This order is applicable beginning January 20, 2022.

FOR FURTHER INFORMATION CONTACT:

Evan Steiner (Legal Information), Office of the General Counsel, 888 First Street NE, Washington, DC 20426, (202) 502-8792

Monil Patel (Technical Information), Office of Energy Market Regulation, 888 First Street NE, Washington, DC 20426, (202) 502-8296

SUPPLEMENTARY INFORMATION:

Order on Rehearing

(Issued January 20, 2022)

1. On December 17, 2020, the Commission issued an order establishing an oil pipeline index level of Producer Price Index for Finished Goods plus 0.78% (PPI-FG+0.78%) for the five-year period beginning July 1, 2021.
1

On January 19, 2021, Joint Commenters,
2

Liquids Shippers Group (Liquids Shippers),
3

the Canadian Association of Petroleum Producers (CAPP) (together with Joint Commenters and Liquids Shippers, Shippers), the Association of Oil Pipe Lines (AOPL), and Designated Carriers
4

(together with AOPL, Pipelines) requested rehearing or clarification of the December 2020 Order.

1

Five-Year Rev. of the Oil Pipeline Index,
86 FR 9448 (Feb. 16, 2021), 173 FERC ¶ 61,245 (2020) (December 2020 Order).

2
Joint Commenters include: The Airlines for America; Chevron Products Company; the National Propane Gas Association; and Valero Marketing and Supply Company.

3
Liquids Shippers include: Apache Corporation; Cenovus Energy Marketing Services Ltd.; ConocoPhillips Company; Devon Gas Services, L.P.; Equinor Marketing & Trading US Inc.; Fieldwood Energy LLC; Marathon Oil Company; Murphy Exploration and Production Company—USA; Ovintiv Marketing, Inc.; and Pioneer Natural Resources USA, Inc.

4
Designated Carriers include: Buckeye Partners, L.P.; Colonial Pipeline Company; Energy Transfer LP; Enterprise Products Partners L.P.; and Plains All American Pipeline, L.P.

2. As discussed below, we grant the requests for rehearing, in part, and deny the requests for rehearing, in part. As a result, we adopt an index level of PPI-FG-0.21%. This departure from the December 2020 Order results from: (a) Trimming the data set to the middle 50% of cost changes, as opposed to the middle 80%; (b) incorporating the effects of the Commission's 2018 policy change requiring Master Limited Partnership (MLP)-owned pipelines to eliminate the income tax allowance and previously accrued Accumulated Deferred Income Taxes (ADIT) balances from their page 700 summary costs of service (Income Tax Policy Change);
5

and (c) correcting the index calculation to rely upon updated page 700 cost data for 2014.

5

Inquiry Regarding the Commission's Policy for Recovery of Income Tax Costs,
162 FERC ¶ 61,227, at P 8 (2018 Income Tax Policy Statement),
reh'g denied,
164 FERC ¶ 61,030, at P 13 (2018),
request for clarification dismissed,
168 FERC ¶ 61,136 (2019),
petitions for review dismissed sub nom. Enable Miss. River Transmission, LLC
v.
FERC,
820 F. App'x 8 (2020).

3. In addition, as discussed below, we direct oil pipelines to recompute their ceiling levels for July 1, 2021 through June 30, 2022, based upon an index level of PPI-FG-0.21%. Consistent with § 342.3(e) of the Commission's regulations,
6

any oil pipeline with a filed rate that exceeds its recomputed ceiling level for July 1, 2021 through June 30, 2022 must file to reduce that rate to bring it into compliance with the pipeline's recomputed ceiling level. We direct such pipelines to submit these filings to be effective March 1, 2022.

6
18 CFR 342.3(e).

I. Background

A. The Kahn Methodology

4. The Commission reviews the oil pipeline index level
7

every five years.
8

Beginning in Order No. 561 and in each ensuing five-year review, the Commission has adjusted the index level using the Kahn Methodology, which calculates each pipeline's cost change on a per barrel-mile basis over the prior five-year period (
e.g.,
2014-2019 in this proceeding) based upon FERC Form No. 6, page 700 summary cost-of-service data. In order to remove statistical outliers and spurious data, the Kahn Methodology trims the data set by removing an equal number of pipelines at the top and bottom of the data set.
9

The Kahn Methodology then averages three measures of the trimmed data sample's central tendency (the median, mean, and weighted mean) to determine a composite central tendency and compares this average to the changing value of PPI-FG over the same five-year period. The index level is set at PPI-FG plus (or minus) this differential. Historically, the index has ranged from PPI-FG-1% to PPI-FG+2.65%, and in 2015, the Commission set the index level at PPI-FG+1.23%.

7
Pursuant to the indexing methodology, pipelines may increase their ceiling levels effective every July 1 by “multiplying the previous index year's ceiling level by the most recent index published by the Commission.” 18 CFR 342.3(d)(1). The Commission publishes an annual index figure every May in a notice issued in Docket No. RM93-11-000.

8

Revisions to Oil Pipeline Regulations Pursuant to Energy Policy Act of 1992,
Order No. 561, FERC Stats. & Regs. ¶ 30,985, at 30,941 (1993) (cross-referenced at 65 FERC ¶ 61,109),
order on reh'g,
Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 (1994) (cross-referenced at 68 FERC ¶ 61,138),
aff'd sub nom. Ass'n of Oil Pipe Lines
v.
FERC,
83 F.3d 1424 (D.C. Cir. 1996) (
AOPL I
).

9
In Order No. 561 and the 2015 and 2010 five-year reviews, the Commission relied solely upon the middle 50% of the data set.
Five-Year Rev. of the Oil Pipeline Index,
153 FERC ¶ 61,312, at PP 42-44 (2015) (2015 Index Review),
aff'd sub nom. Ass'n of Oil Pipe Lines
v.
FERC,
876 F.3d 336 (D.C. Cir. 2017) (
AOPL III
);
Five-Year Rev. of the Oil Pipeline Pricing Index,
133 FERC ¶ 61,228, at P 60 (2010) (2010 Index Review),
reh'g denied,
135 FERC ¶ 61,172 (2011) (2010 Index Rehearing Order); Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,096-097. In the 2005 and 2000 five-year reviews, the Commission averaged the middle 50% with the middle 80% but did not justify or address its consideration of the middle 80%. 2010 Index Review, 133 FERC ¶ 61,228 at P 60. In addition, in the 2000 review, considering the middle 80% did not alter the index calculation.
Id.

B. Notice of Inquiry and Comments

5. On June 18, 2020, the Commission issued a Notice of Inquiry (NOI) proposing to adopt an index level of

PPI-FG+0.09%.
10

The NOI proposed to calculate the index level by (1) trimming the data set to the middle 50% and (2) incorporating the effects of the Income Tax Policy Change upon pipeline cost changes over the 2014-2019 period.
11

The Commission explained that commenters could address issues including, but not limited to, different data trimming methodologies and whether, and if so how, the Commission should reflect the effects of cost-of-service policy changes in the index calculation.
12

10

Five-Year Rev. of the Oil Pipeline Index,
171 FERC ¶ 61,239 (2020) (NOI).

11

Id.
PP 9-10.

12

Id.

6. Ten commenters filed comments in response to the NOI.
13

Pipelines urged the Commission to use the middle 80%, as opposed to the middle 50%, and proposed to adjust the reported page 700 data for 2014 to eliminate the effects of the Income Tax Policy Change. Shippers, by contrast, argued that the Commission should continue using the middle 50% and reject Pipelines' proposed adjustments to the data set. In addition, Liquids Shippers proposed to replace the weighted mean in the Kahn Methodology's calculation of central tendency with the weighted median and to replace the returns on equity (ROE) reported on page 700 for 2014 and 2019 with standardized, industry-wide ROEs for both years. CAPP argued that negotiated rate contracts have served to reduce pipelines' risks and urged the Commission to require pipelines to provide their page 700 workpapers to investigate whether the reported page 700 ROEs reflect these effects.

13
Comments were filed by AOPL, Designated Carriers, Kinder Morgan, Inc., Colonial, Joint Commenters, Liquids Shippers, CAPP, the Energy Infrastructure Council, the Pipeline Safety Trust, and the Pipeline and Hazardous Materials Safety Administration (PHMSA).

C. December 2020 Order and Requests for Rehearing

7. The December 2020 Order established an index level of PPI-FG+0.78%.
14

The Commission adopted Pipelines' proposed adjustments to remove the effects of the Income Tax Policy Change from the index calculation
15

and to use the middle 80%,
16

and declined to adopt Liquids Shippers' and CAPP's proposals.
17

On January 19, 2021, Shippers filed requests for rehearing challenging these determinations and Pipelines requested rehearing or clarification to correct minor errors in the workpapers underlying the December 2020 Order.

14
December 2020 Order, 173 FERC ¶ 61,245 at P 2.

15

Id.
PP 16-20.

16

Id.
PP 25-32.

17

Id.
PP 36-40, 45-50, 52-53.

II. Discussion

A. 2018 MLP Income Tax Policy Change

1. December 2020 Order

8. Prior to the December 2020 Order, the Commission committed in the 2018 Income Tax Policy Statement to “incorporate the effects of [the Income Tax Policy Change] on industry-wide oil pipeline costs in the 2020 five-year review . . . .”
18

Through the Income Tax Policy Change, the Commission altered its policies so that natural gas and oil pipelines organized as MLPs could not recover the same tax costs twice in their rates.
19

Although the Commission acted immediately to address this double recovery in natural gas pipeline rates,
20

the Commission deferred action regarding oil pipeline rates and emphasized that oil pipeline rates “will be addressed in due course” during the 2020 five-year index review.
21

The Commission explained that by acting in the 2020 five-year review, the Commission would “ensure that the industry-wide reduced costs are incorporated on an industry-wide basis. . . .”
22

18
Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 8.

19
From 2005 to 2018, the Commission allowed MLP pipelines to claim a full income tax allowance in their costs of service.
Inquiry Regarding Income Tax Allowances,
111 FERC ¶ 61,139, at P 32 (2005) (2005 Income Tax Policy Statement). In a series of orders beginning in 2016, the Commission and the U.S. Court of Appeals for the District of Columbia Circuit (D.C. Circuit) found that allowing MLP pipelines to recover both an income tax allowance and an ROE determined using the Discounted Cash Flow (DCF) model results in an impermissible double recovery of tax costs. The Commission rectified the double recovery through the Income Tax Policy Change in 2018, finding that MLP pipelines could no longer recover an income tax allowance and could eliminate previously accumulated ADIT balances from their costs of service. The D.C. Circuit affirmed the Commission's decisions in 2020.
See United Airlines, Inc.
v.
FERC,
827 F.3d 122 (D.C. Cir. 2016) (
United Airlines
),
order on remand, SFPP, L.P.,
Opinion No. 511-C, 162 FERC ¶ 61,228, at P 22 (2018), (remanding the Commission's application of the 2005 policy),
reh'g denied,
Opinion No. 511-D, 166 FERC ¶ 61,142, at PP 90-95 (2019),
aff'd, SFPP, L.P.
v.
FERC,
967 F.3d 788, 793-97, 801-03 (D.C. Cir. 2020) (
SFPP
);
see also
Income Tax Policy Statement, 162 FERC ¶ 61,227,
reh'g denied,
164 FERC ¶ 61,030,
request for clarification dismissed,
168 FERC ¶ 61,136;
petitions review dismissed sub nom. Enable Miss. River Transmission, LLC
v.
FERC,
820 F. App'x 8.

20

Interstate & Intrastate Nat. Gas Pipelines,
Order No. 849, 164 FERC ¶ 61,031, at P 30 (2018),
reh'g denied,
Order No. 849-A, 167 FERC ¶ 61,051 (2019).

21
Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 46.

22

Id.
P 8.

9. However, when the 2020 five-year review arrived, the Commission reversed course. In the December 2020 Order, the Commission declined to incorporate the effects of the Income Tax Policy Change into the 2020 five-year review index calculation. Accordingly, the December 2020 Order adopted Designated Carriers' proposal to eliminate the effects of the Income Tax Policy Change from the index calculation by adjusting the reported page 700 data for all pipelines that were MLPs in 2014 to reduce the 2014 income tax allowance to zero and to revise the 2014 return on rate base to reflect the removal of ADIT.
23

23
December 2020 Order, 173 FERC ¶ 61,245 at P 16. Because the 2014 page 700 data reflected the old policy whereas the 2019 data reflected the new policy, a straightforward application of the longstanding Kahn Methodology would have incorporated the cost reductions caused by the Income Tax Policy Change. AOPL's and Designated Carriers' proposals for eliminating the effects of the Income Tax Policy Change differed. AOPL proposed to (1) eliminate the 2014 income tax allowance for all pipelines that reduced their income tax allowance from a positive number to zero in response to the 2018 Income Tax Policy Statement and continued reporting zero income tax allowance for the remainder of the 2014-2019 period, and (2) adjust these pipelines' 2014 return on rate base to reflect the elimination of their ADIT balances. Designated Carriers supported AOPL's adjustments and proposed to extend them to all pipelines that were owned by MLPs in 2014, including those that later converted to business forms eligible to recover an income tax allowance. No entity challenges on rehearing the Commission's decision not to adopt AOPL's proposal.

10. The Commission determined that although the index aims to reflect changes in recoverable costs, alterations to the Opinion No. 154-B methodology
24

are distinct from the annual changes to pipeline costs that are input into that methodology.
25

The Commission stated that the index is not a true-up designed to remedy over- or under-recoveries resulting from past cost-of-service policy changes, but instead simply allows for incremental rate adjustments to enable recovery of future cost changes.
26

The Commission also determined that it was not clear that the double recovery of MLP pipelines' income tax costs was ever incorporated into the index or that MLP

pipelines benefitted from the Commission's prior policy permitting them to recover an income tax allowance.
27

24
The Opinion No. 154-B methodology is the cost-of-service ratemaking methodology that the Commission uses for oil pipelines.
Williams Pipe Line Co.,
Opinion No. 154-B, 31 FERC ¶ 61,377,
order on reh'g,
Opinion No. 154-C, 33 FERC ¶ 61,327 (1985). The Opinion No. 154-B methodology is based upon trended original costs, whereby the inflationary component of the nominal return is placed in deferred earnings and recovered as a part of rate base in future years.
E.g., BP W. Coast Prods., LLC
v.
FERC,
374 F.3d 1263, 1282-83 (D.C. Cir. 2004).

25
December 2020 Order, 173 FERC ¶ 61,245 at P 17 (stating that “the purpose of indexing is to allow the indexed rate to keep pace with industry-wide cost changes, not to reflect alterations to the Commission's Opinion No. 154-B cost-of-service methodology”).

26

Id.
P 18.

27

Id.
P 19 & n.37.

2. Rehearing Requests

11. Shippers argue that the Commission's decision to adjust reported page 700 data to remove the effects of the Income Tax Policy Change contravenes established precedent and rests upon flawed reasoning. First, Shippers contend that both the D.C. Circuit and the Commission have found that the index aims to track changes in recoverable pipeline costs consistent with the Opinion No. 154-B methodology.
28

Shippers argue that the Income Tax Policy Change changed pipelines' recoverable costs by requiring MLP pipelines to remove the income tax allowance and ADIT balances from their costs of service. Thus, Shippers contend that the index should reflect this policy change.
29

28
Joint Commenters Request for Rehearing at 43-45 (quoting 2015 Index Review, 153 FERC ¶ 61,312 at P 13) (citing
AOPL III,
876 F.3d at 345-46); Liquids Shippers Request for Rehearing at 17 (quoting
AOPL III,
876 F.3d at 345; 2015 Index Review, 153 FERC ¶ 61,312 at P 13).

29
Joint Commenters Request for Rehearing at 42-46; Liquids Shippers Request for Rehearing at 16-19.

12. Second, Shippers state that the December 2020 Order contradicts the Commission's statement in the 2018 Income Tax Policy Statement that it would “incorporate the effects” of the Income Tax Policy Change in this five-year review.
30

Shippers assert that they relied upon this statement and, as a result, lost their ability to seek rehearing or judicial review of the 2018 Income Tax Policy Statement and forewent opportunities to challenge oil pipeline rates.
31

Shippers further claim that the Commission's continued inaction on eliminating the MLP income tax double recovery from oil pipeline rates, as contrasted with its actions to eliminate that double recovery from natural gas pipeline rates, raises due process concerns for oil pipeline shippers.
32

In addition, Shippers disagree with the December 2020 Order's conclusion that reflecting the Income Tax Policy Change would convert the index into a true-up designed to remedy a prior over-recovery.
33

30
2018 Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 46; Joint Commenters Request for Rehearing at 41-42, 56; Liquids Shippers Request for Rehearing at 15.

31
Joint Commenters Request for Rehearing at 57; CAPP Request for Rehearing at 11-13;
see also
Liquids Shippers Request for Rehearing at 15-16.

32
Joint Commenters Request for Rehearing at 59-60.

33

Id.
at 46-47.

13. Third, Shippers maintain that adjusting reported page 700 data is unprecedented and departs from the Commission's consistent practice of calculating the index level using unadjusted data.
34

Shippers state that the Commission has previously rejected proposals to make targeted adjustments to the data set by removing pipelines with cost changes resulting from specific factors because such proposals failed to identify other factors that could render a pipeline's data non-comparable.
35

Shippers contend that the Commission should likewise reject Pipelines' adjustments because they fail to consider other factors or policy changes.
36

34

Id.
at 46.

35

Id.
at 51 (quoting 2015 Index Review, 153 FERC ¶ 61,312 at P 34).

36

Id.

14. Fourth, Shippers state that regardless of whether prior index calculations directly incorporated the Commission's prior policies allowing MLP pipelines to recover an income tax allowance, the MLP income tax allowance became integrated into the industry's recoverable costs and thus came to be reflected in the index.
37

Shippers also argue that MLP pipelines did, in fact, benefit from these policies because they allowed MLPs to report higher costs on their page 700s, which helped to insulate their annual index rate increase filings from challenge under the Commission's Percentage Comparison Test.
38

37

Id.
at 53.

38

Id.
at 53-55. Under the Percentage Comparison Test, the Commission will investigate a protested index rate increase filing if the pipeline's page 700 revenues exceed its costs and there is a more than a 10 percentage-point differential between the index rate increase and the change in the prior two years' total cost-of-service data reported on page 700, line 9.
E.g., HollyFrontier Refin. & Mktg. LLC
v.
SFPP, L.P.,
170 FERC ¶ 61,133, at P 5 (2020).

15. Fifth, Liquids Shippers argue that the December 2020 Order further distorts the index calculation by adjusting the page 700 data of pipelines that were MLPs in 2014 and converted to C-Corporations after the 2018 Income Tax Policy Change. Liquids Shippers contend that because these pipelines were eligible as C-Corporations to report a positive income tax allowance on page 700 for 2019, reducing their 2014 income tax allowance to zero fabricates an erroneous cost increase between 2014 and 2019.
39

39
Liquids Shippers Request for Rehearing at 18-19.

3. Commission Determination

16. We grant rehearing of the December 2020 Order to calculate the index level using unadjusted page 700 data that reflects the effects of the Income Tax Policy Change upon recoverable pipeline costs.

a. The Income Tax Policy Change Should be Incorporated Into the Index Calculation

17. The index must reflect the Income Tax Policy Change in order to produce just and reasonable oil pipeline rates. Prior to the 2018 Income Tax Policy Change, MLP pipelines' rates could recover the same investor-level tax costs twice, once in an income tax allowance and again in an ROE.
40

The D.C. Circuit and the Commission both concluded that this led to an impermissible double recovery of investor-level tax costs and produced unjust and unreasonable rates.
41

The Income Tax Policy Change eliminated this double recovery by prohibiting MLP pipelines from recovering an income tax allowance. However, oil pipeline rates have yet to incorporate this policy change.
42

Thus, the impermissible double-recovery has not been eliminated from oil pipeline rates. Because indexing is the Commission's primary oil pipeline ratemaking methodology and because indexed oil pipeline rates must be just and reasonable, we conclude that the index calculation must now address the Income Tax Policy Change.

40
2005 Income Tax Policy Statement, 111 FERC ¶ 61,139 at P 32.

41

SFPP,
967 F.3d at 795-97;
United Airlines,
827 F.3d at 136; Income Tax Policy Statement, 162 FERC ¶ 61,227 at PP 8, 45. MLP pipelines do not incur income taxes at the entity level, but the Commission justified permitting them to recover an income tax allowance on the basis that their investors pay taxes on their allocated share of the MLP's taxable income. Because the D.C. Circuit and the Commission concluded that the MLP pipeline's DCF ROE already included investor-level income tax costs, a double recovery resulted from permitting an income tax allowance that recovered those same tax costs. Opinion No. 511-C, 162 FERC ¶ 61,228 at P 22.

42
Pipelines identify only one MLP oil pipeline, SFPP, L.P. (the pipeline whose rates were the subject of
United Airlines
), that has adjusted its rates in response to the Income Tax Policy Change. AOPL Initial Comments at 27-28; Designated Carriers Initial Comments at 11, 14.

18. The index was always intended to reflect changes to Opinion No. 154-B costs such as the elimination of the double recovery via the Income Tax Policy Change. The Opinion No. 154-B methodology defines the costs that oil pipelines can recover in rates and the index is the primary means for recovering those costs.
43

Accordingly, the Commission and the D.C. Circuit have long recognized that the index should reflect changes in costs recoverable under the Opinion No. 154-

B methodology,
44

and the Commission uses the Opinion No. 154-B methodology cost data reported on page 700 to calculate the index level.
45

Here, the adoption of the Income Tax Policy Change altered those costs by barring MLP pipelines from recovering in 2019 income tax costs that they were permitted to recover in 2014. By comparing the 2014 data reported on page 700 under the Commission's previous policy with the 2019 data reported under its revised policy, this index calculation will accurately capture the effects of the Income Tax Policy Change on costs recoverable under Opinion No. 154-B.
46

43
As explained above, the Opinion 154-B methodology is the Commission's cost-of-service ratemaking methodology for oil pipelines.
See supra
note 24.

44

AOPL III,
876 F.3d at 345 (finding that the Commission “has consistently treated the index as a measure of normal industry-wide cost-of-service changes”); 2015 Index Review, 153 FERC ¶ 61,312 at P 13,
aff'd, AOPL III,
876 F.3d at 345-46 (“[T]he index is meant to reflect changes to recoverable pipeline costs, and, thus, the calculation of the index should use data that is consistent with the Commission's [Opinion No. 154-B] cost-of-service methodology.”);
see also
Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,096 (lamenting that the then-existing Form No. 6 provided a “highly unsatisfactory” measure of capital cost changes because it did “not contain the information necessary to compute a trended original cost (TOC) rate base or a starting rate base” under the Opinion No. 154-B methodology).

45
2015 Index Review, 153 FERC ¶ 61,312 at PP 12-13 (adopting use of page 700 data to measure oil pipeline cost changes because, among other reasons, page 700 data is consistent with the Opinion No. 154-B methodology).

46
In contrast, adjusting the data set to remove the effects of this policy change would maintain a divergence between indexed rates and Opinion No. 154-B recoverable costs.

19. We also find that adjusting page 700 data to remove the Income Tax Policy Change's effects conflicts with the Commission's historical practice. The Commission has not previously adjusted the reported data used to derive the index level. Order Nos. 561 and 561-A “opted for a purely historical analysis”
47

for measuring pipeline cost changes based upon documented cost experience, and in each subsequent index review the Commission has calculated the index level using reported Form No. 6 data without adjustment.
48

Thus, modifying MLP pipelines' reported income tax allowances and returns on rate base would depart from the purely historical analysis on which the Commission has consistently relied since establishing the indexing regime.

47

Ass'n of Oil Pipe Lines
v.
FERC,
281 F.3d 239, 247 (D.C. Cir. 2002) (
AOPL II
) (citing
Five-Year Rev. of Oil Pipeline Pricing Index,
93 FERC ¶ 61,266, at 61,855 (2000) (2000 Index Review),
aff'd in part and remanded, AOPL II,
281 F.3d 239,
order on remand,
102 FERC ¶ 61,195 (2003); Order No. 561, FERC Stats. & Regs. ¶ 30,985 at 30,951).

48
Although the Commission has curtailed the
amount
of data it considers in calculating the index level via statistical data trimming to the middle 50%, it has never modified the specific
inputs
that pipelines have recorded in their Form No. 6 filings. Similarly, while the Commission adjusts the data set to account for pipeline mergers and divestitures that occurred during the five-year review period, these steps are distinguishable from the adjustments to omit the effects of the Income Tax Policy Change adopted in the December 2020 Order based upon Pipelines' proposals. Where pipelines filed separate page 700 data for the first year of the review period (
e.g.,
2014) and merged later in the review period, the Commission adds the separate costs that the pipelines reported for the first year and compares this sum to the newly combined company's page 700 costs reported for the last year of the data set (
e.g.,
2019). 2015 Index Review, 153 FERC ¶ 61,312 at P 38. Conversely, in the case of divestitures, the Commission adds the separate costs the pipelines reported for the last year of the data set and compares this sum to the formerly combined company's page 700 costs reported for the first year. Unlike Pipelines' proposed adjustments, which alter a specific cost item that pipelines reported on page 700, this step simply combines the total costs that the pipelines reported as separate entities at one endpoint of the review period to mirror their status as a combined entity at the other endpoint.

20. In addition, incorporating the Income Tax Policy Change into the index complies with the Energy Policy Act of 1992's (EPAct 1992) dual mandate for just and reasonable rates and for simplified and streamlined ratemaking.
49

As stated above, the D.C. Circuit and the Commission have previously held that an impermissible double recovery results from granting MLP pipelines an income tax allowance.
50

Thus, as the Commission's Opinion No. 154-B methodology evolves, oil pipeline rates adjusted via indexing must reflect those changes in order to remain just and reasonable. If the Commission omits the effects of the Income Tax Policy Change from the index calculation, the only alternative method of reflecting the elimination of the MLP income tax double recovery in rates would be through cost-of-service rate litigation.
51

We find that implementing cost-of-service policy changes in this fashion would hinder the statutory goals of efficient and simplified ratemaking embodied in EPAct 1992.
52

49
Public Law 102-486, 1801(a), 106 Stat. 2776, 3010 (1992) (codified at 42 U.S.C. 712 note).

50

See supra
note 19.

51
Importantly, this proceeding presents the sole opportunity for addressing the MLP income tax double recovery in indexed rates via the simplified and streamlined five-year review process. As discussed above, the Kahn Methodology calculates the index level based upon the change in industry-wide page 700 costs from the first year of the review period to the last year. Accordingly, it is only possible to reflect the Income Tax Policy Change in the instant index calculation, which measures cost changes from 2014 (when MLP pipelines reported a positive income tax allowance) to 2019 (when MLP pipelines reported zero income tax allowance). Capturing this decrease in recoverable income tax costs from 2014 to 2019 will reduce the index level to incorporate the elimination of the MLP income tax double recovery. In contrast, the 2025 five-year review will reflect no change in MLP income tax costs because MLP pipelines will report zero income tax allowance for both the first and last years of the 2019-2024 period.

52

See AOPL II,
281 F.3d at 244 (holding that an oil pipeline ratemaking regime based in large part upon cost-of-service rate proceedings “would be inconsistent with Congress's mandate under the EPAct for FERC to establish `a simplified and generally applicable ratemaking methodology.' ” (quoting EPAct 1992, at 1801(a))).

21. Finally, our holding on rehearing honors the Commission's assurances in the 2018 Income Tax Policy Statement. There, the Commission committed to “incorporate the effects of [the Income Tax Policy Change] . . . in the 2020 five-year review” so that oil pipeline rates would reflect these reduced costs.
53

Whereas the Commission acted immediately to eliminate the MLP income tax double recovery from natural gas pipeline rates,
54

the Commission deferred adjusting oil pipeline rates until the 2020 five-year index review. Failure to act here would leave oil pipeline rates unaddressed indefinitely. While Pipelines urge the Commission to disregard our assurances from the 2018 Income Tax Policy Statement, they offer no alternative remedy.
55

Moreover, we recognize that

53
Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 8;
see also Inquiry Regarding the Effect of the Tax Cuts and Jobs Act on Commission-Jurisdictional Rates,
162 FERC ¶ 61,223, at P 4 (2018) (“The Commission must ensure that the rates, terms, and conditions of jurisdictional services under the Federal Power Act (FPA), the Natural Gas Act (NGA), and the Interstate Commerce Act are just, reasonable, and not unduly discriminatory or preferential.”);
id.
P 8 (directing oil pipelines to report on page 700 an income tax allowance consistent with the Income Tax Policy Change and the Tax Cuts and Jobs Act. As opposed to initiating cost-of-service complaints against oil pipelines, deferring action until the 2020 five-year index review best fulfilled EPAct 1992's dual mandate for simplified oil pipeline ratemaking and just and reasonable rates.
See supra
P 20 & note 51.

54
Specifically, the Commission required natural gas pipelines to submit a one-time filing for the purpose of evaluating the impact of the Income Tax Policy Change and the Tax Cuts and Jobs Act upon the pipeline's revenue requirement. Order No. 849, 164 FERC ¶ 61,031 at P 30. This process allowed for MLP natural gas pipelines to voluntarily reduce their rates in response to the Income Tax Policy Change and for the Commission to initiate rate investigations pursuant to section 5 of the Natural Gas Act where the pipeline appeared to be over-recovering its cost of service as a result of the policy change.
E.g., Stagecoach Pipeline & Storage Co.,
166 FERC ¶ 61,199 (2019);
N. Nat. Gas Co.,
166 FERC ¶ 61,033 (2019). In contrast to MLP natural gas pipelines, Pipelines identify only one MLP oil pipeline, SFPP, L.P. (the pipeline whose rates were the subject of
United Airlines
), that has adjusted its rates in response to the Income Tax Policy Change.
See supra
note 42.

55
We recognize that the 2018 Income Tax Policy Statement provided non-binding guidance regarding the Commission's future intentions. Accordingly, in the NOI initiating this proceeding, the Commission invited the commenters to address this issue. NOI, 171 FERC ¶ 61,239 at P 10. Our determination here is based upon the full consideration of the extensive record developed in this proceeding.

shippers relied upon our assurances in considering whether to bring challenges to oil pipeline rates following the Income Tax Policy Change.
56

56
Joint Commenters Request for Rehearing at 57; CAPP Request for Rehearing at 11-13;
see also
Liquids Shippers Request for Rehearing at 15-16.

b. Reconsidering the December 2020 Order

22. As discussed below, we reject the reasons provided by the December 2020 Order for excluding the Income Tax Policy Change from the index calculation.

23. First, there is no meaningful distinction between changes to the Opinion No. 154-B methodology and changes to the costs that pipelines input into that methodology.
57

Rather, changes to the Opinion No. 154-B methodology produce corresponding changes to the costs that pipelines can recover. Thus, for purposes of determining the index, any meaningful measure of changes to recoverable costs between 2014 and 2019 must reflect the Income Tax Policy Change. The December 2020 Order's adjustments to the page 700 data omit the effects of the Income Tax Policy Change—as though MLP pipelines did not receive an income tax allowance in 2014.
58

Given the purpose of the indexing regime to adjust rates for changes to Opinion No. 154-B recoverable costs, a true “apples-to-apples” comparison involves comparing the recoverable costs in 2014 with the recoverable costs in 2019—if companies received an income tax allowance in 2014 but did not in 2019, the index must reflect that reality.

57
December 2020 Order, 173 FERC ¶ 61,245 at P 17 (stating that “the purpose of indexing is to allow the indexed rate to keep pace with industry-wide cost changes, not to reflect alterations to the Commission's Opinion No. 154-B cost-of-service methodology”).

58
In the December 2020 Order, the Commission stated that “[j]ust as a business must account for changes to its accounting policies when comparing costs over two different periods, we must make a similar adjustment to the reported page 700 data here to derive an `apples-to-apples' comparison of pipeline cost changes.”
Id.
This analogy to accounting methods is misplaced. Whereas an accounting methodology simply involves the method of recording costs, as explained above, the Income Tax Policy Change directly affected the costs that MLP pipelines can recover under the Opinion No. 154-B methodology.

24. Second, contrary to the statements in the December 2020 Order, we find that reflecting the Income Tax Policy Change does not effectuate a true-up for prior-period over-recoveries.
59

Consistent with the purposes of the five-year review, incorporating the effects of the Income Tax Policy Change in the index calculation will align pipelines'
future
rates with their
future
costs recoverable under Opinion No. 154-B. By failing to reflect the Income Tax Policy Change in the calculation of the prospective index, the approach adopted in the December 2020 Order would cause future indexed rates to become estranged from future recoverable costs.

59

Id.
P 18.

25. Third, we disagree with the December 2020 Order's reasoning that “[b]ecause no prior index calculation incorporated the [2005 policy] allowing MLP pipelines to recover an income tax allowance, it is not necessary to reflect the policy change denying those pipelines an income tax allowance in the calculation here.”
60

This statement disregards indexing's purpose and oversimplifies the Commission's historical practice. Indexed rates have always served as a means for recovering pipeline income tax costs. Accordingly, the five-year review index calculation was always intended to incorporate changes in pipeline income tax costs, even if the Commission previously measured those costs using an imperfect estimate.
61

Now, the Commission uses page 700 data that directly measures income tax costs. The Commission should not disregard this data when calculating the index level.

60

Id.
P 19.

61
Before the 2015 Index Review when the Commission began using page 700 data, the Commission estimated pipeline cost changes using a rough proxy based upon Form No. 6 accounting data. This accounting data did not directly measure changes in the income tax costs recoverable under Opinion No. 154-B. December 2020 Order, 173 FERC ¶ 61,245 at P 19;
see also
2015 Index Review, 153 FERC ¶ 61,312 at PP 14-15 (describing this proxy and its deficiencies). The Commission relied upon this proxy because direct measures of capital costs and income tax costs were not available when the index was first established. 2015 Index Review, 153 FERC ¶ 61,312 at P 14. Before page 700 was created, the Commission lamented that “the measure of the capital cost component of the cost of service is highly unsatisfactory” because Form No. 6 did “not contain the information necessary to compute a trended original cost . . . rate base or a starting rate base as allowed for in Order No. 154-B.” Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,096.

26. Moreover, the facts here undercut Pipelines' claim that MLP income taxes have not been incorporated into pipeline rates.
62

Prior to the 2005 income tax policy change, MLP pipelines were eligible to include at least a partial income tax allowance in their costs of service.
63

To the extent that prior index calculations did not incorporate the 2005 income tax policy directly, pipeline rates did substantially come to reflect that policy over time.
64

To explain further, as the number of pipelines in the Commission's data set expanded,
65

all initial rates and non-indexing rate changes would have reflected MLPs pipelines' ability to recover a full income tax allowance under the previous 2005 policy. Although we recognize that prior index reviews imperfectly captured the 2005 income tax policy change, we know that the 2005 policy change plainly affected oil pipeline rates over the last 15 years.
66

Furthermore, Pipelines' argument ignores how MLP pipelines' ability to claim an income tax allowance under the previous 2005 policy shielded those pipelines' rates from challenge.
67

Therefore, we are not persuaded by arguments based upon the 2005 policy change that the Commission must remove the 2018 Income Tax Policy Change from this index calculation.

62
Designated Carriers Initial Comments at 17-20;
see also
December Order, 173 FERC ¶ 61,245 at P 19.

63

Lakehead Pipe Line Co., L.P.,
Opinion No. 397, 71 FERC ¶ 61,338, at 62,314-15 (1995),
reh'g denied,
Opinion No. 397-A, 75 FERC ¶ 61,181 (1996) (permitting partnership entities like MLP pipelines to recover an income tax allowance for income attributable for corporate partners, but not for income attributable to individuals or other non-corporate partners);
see also Riverside Pipeline Co., L.P.,
48 FERC ¶ 61,309, at 62,018 (1989) (applying pre-
Lakehead
policy permitting partnership pipelines to recover a full income tax allowance as if they were corporations).

64
Consistent with EPAct 1992's mandate for a simplified and streamlined ratemaking regime, the Commission does not scrutinize the costs underlying each individual pipeline's rates when developing the industry-wide index. Rather, the Commission reaches its determinations based upon what is appropriate on balance for the industry as a whole.

65
Notably, 164 of the 277 total oil pipelines in the Commission's data set, or 59%, have been added since the 2005 five-year review.

66
In urging the Commission to adopt the adjustment to the reported page 700 data to eliminate the effects of the Income Tax Policy Change, neither AOPL nor Designated Carriers account for the extent to which the Commission's prior income tax policies permitting MLPs to recover an income tax allowance were incorporated into pipelines' existing rates.

67
Specifically, the Commission evaluates cost-of-service complaints and challenges to annual index rate increases based upon the differential between costs and revenues on page 700. To the extent that an MLP pipeline's page 700 revenues exceeded its costs, the ability to report an income tax allowance as a cost on page 700 would have reduced the gap between revenues and costs. This lower cost-revenue differential would have reduced the pipeline's exposure to cost-of-service rate complaints and challenges to index rate changes.

27. Fourth, the adjustments adopted in the December 2020 Order lead to incongruous and unreasonable results because they enable pipelines, including those with an existing double recovery, to increase their rates above the levels that would have resulted absent the D.C. Circuit's and the Commission's double-recovery findings. The Commission adopted the Income Tax Policy Change in response to findings by the D.C. Circuit and the Commission that MLP pipeline rates

were double recovering those pipelines' income tax costs.
68

Absent the D.C. Circuit's and the Commission's holdings prohibiting MLP pipelines from recovering an income tax allowance in their costs of service, MLP pipelines, like corporate pipelines, would have reported a reduction in their income tax allowances as a result of the Tax Cuts and Jobs Act. However, by treating MLP pipelines' income tax liability as zero for both 2014 and 2019, Pipelines' adjustments eliminate the downward effect the Tax Cuts and Jobs Act would have exerted upon MLP pipelines' recoverable income tax costs during the 2014-2019 period.
69

Thus, not only do Pipelines' adjustments eliminate the reduction in industry-wide recoverable costs resulting from the Income Tax Policy Change, but they also diminish the separate reduction in MLP pipelines' recoverable costs that would have resulted from the Tax Cuts and Jobs Act had that policy change not occurred. As a result, incorporating Pipelines' adjustments in the cost-change analysis would produce a
higher
index level than what would have resulted absent the Income Tax Policy Change eliminating the MLP income tax double recovery. Therefore, we decline to adopt Pipelines' adjustments given this incongruous and unreasonable result and instead calculate the index level using unadjusted page 700 data.

68
2018 Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 8.

69
All commenters agree that the index should reflect the decrease resulting from the Tax Cuts and Jobs Act to the income tax allowance recoverable by pipelines organized as corporations. December 2020 Order, 173 FERC ¶ 61,245 at P 10 n.20.

c. Pipelines' Remaining Arguments Are Unpersuasive

28. We are unpersuaded by Pipelines' remaining arguments for removing the effects of the Income Tax Policy Change from the index calculation. Regarding their claim that the policy change should be excluded because it did not affect pipelines' actual income tax costs,
70

we find that this argument misconstrues the cost changes that the index is designed to measure. As discussed above, “the index is meant to reflect changes to recoverable pipeline costs” measured under the Opinion No. 154-B methodology.
71

Thus, the index is designed to track changes in the income tax costs that pipelines can recover under the Commission's cost-of-service ratemaking methodology. In arguing that the Income Tax Policy Change only modified the ratemaking treatment of MLP income tax costs without affecting actual costs,
72

Pipelines overlook that changes in ratemaking treatment produce the very Opinion No. 154-B cost-of-service changes that the index calculation seeks to measure.
73

70

E.g.,
AOPL Initial Comments at 29-31; AOPL Reply Comments at 11; Designated Carriers Initial Comments at 7, 9-12;
see also
Kinder Morgan Initial Comments at 3-4.

71
2015 Index Review, 153 FERC ¶ 61,312 at P 13 (citing Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,096). In fact, the Commission updated its calculation of the index level to rely upon page 700 because it includes actual total cost-of-service data consistent with Opinion No. 154-B.
Id.
PP 13-14.

72
AOPL Initial Comments at 30 (quoting Shehadeh Initial Decl. at 14); Designated Carriers Initial Comments at 3, 7-8, 10-11.

73

AOPL III,
876 F.3d at 345 (“[N]either Order No. 561 nor the subsequent index review orders indicate that the index was intended to measure something distinct from the costs measured under its cost-of-service methodology. Rather, the Commission has consistently treated the index as a measure of normal industry-wide cost-of-service changes . . . .”).

29. Moreover, the income tax costs that pipelines can recover under Opinion No. 154-B are distinct from the actual tax costs that pipelines have paid to the taxing authority. Instead, as the D.C. Circuit has recognized, income tax costs recoverable under the Commission's cost-of-service methodology are not equivalent to “actual taxes paid.”
74

Accordingly, because recoverable income tax costs do not correspond to taxes paid, we reject Pipelines' claim that the index should only reflect changes in actual income tax costs.

74

City of Charlottesville
v.
FERC,
774 F.2d 1205, 1213-15 (D.C. Cir. 1985) (Scalia, J.). As then-Judge Scalia explained:

[T]he imprecision of the “actual taxes paid” formulation is exceeded only by the name of the Holy Roman Empire: two out of the three words are wrong. Taxes, yes. But not necessarily
actual
taxes, since inexact estimations are often allowed,
e.g.,
a nationwide tax allowance applied to all individual utilities . . . . And not necessarily taxes
paid,
since tax liability incurred by current activities but in fact not paid currently can be charged to present ratepayers,
e.g.,
taxes deferred by reason of accelerated depreciation but passed on to current ratepayers through normalization. So the principle should be expressed `actual or estimated taxes paid or incurred'—whereupon it ceases to constrain the Commission with regard to taxes any more than the Commission is constrained with regard to its treatment of other expenses. Which is as it should be.

Id.
at 1215 (emphasis in original) (citing
Pub. Sys.
v.
FERC,
709 F.2d 73, 81-82 (D.C. Cir. 1983);
Tenneco Oil Co.
v.
FERC,
571 F.2d 834, 844 (5th Cir. 1978)).

30. We also reject AOPL's assertion that the Income Tax Policy Change should be excluded because it represents an extraordinary, one-time event that is not representative of likely future cost experience.
75

As discussed above, the Kahn Methodology calculates the index level based upon historical cost changes, and does not address speculative assertions about future developments.
76

Consistent with this approach, the Commission has previously rejected similar requests to adjust the data set for one-time cost changes resulting from events that were unlikely to reoccur in the future. For example, in the 2000 Index Review, the Commission rejected a proposed adjustment to address one-time cost savings resulting from the establishment of the indexing methodology and its associated cost efficiency incentives.
77

The D.C. Circuit affirmed this decision, finding that the Commission reasonably adhered to its purely historical analysis and “declined to embroil itself in the complexity and iffiness of” a forward-looking methodology.
78

Similarly, in the 2010 Index Review, the Commission rejected shipper proposals to manually trim the data set to remove pipelines that reported one-time cost increases attributable to expansions or major rate base changes.
79

Just as the Commission declined to adjust the data sets in those proceedings to eliminate the effects of one-time events, we likewise decline to adjust the data set here to eliminate the effects of the Income Tax Policy Change.
80

75
AOPL Reply Comments at 13-14.

76

See
2000 Index Review, 93 FERC at 61,855 (“The purpose of our indexing methodology is to permit adjustment to ceiling rates based on historical not anticipated cost changes over some future period.”).

77

Id.; see also id.
(rejecting proposed adjustment based upon anticipated future cost increases due to increased environmental and safety regulations).

78

AOPL II,
281 F.3d at 247.

79
2010 Index Review, 133 FERC ¶ 61,228 at PP 48-55.

80
Furthermore, we find that AOPL's arguments are internally inconsistent. AOPL's reasoning for excluding the Income Tax Policy Change because it is an extraordinary, one-time policy change would apply equally to the Tax Cuts and Jobs Act, yet AOPL does not oppose reflecting the Tax Cuts and Jobs Act's effects in the index calculation. AOPL Initial Comments at 25-26; AOPL Reply Comments at 10.

31. We disagree with AOPL's contention that the Income Tax Policy Change renders the page 700 data not “consistent enough,” and, therefore, that the page 700 data must be adjusted to remove the effects of the Income Tax Policy Change.
81

This argument relies upon a passage in
AOPL III
stating that the Commission, in adopting the use of page 700 data to measure pipeline cost changes, determined in the 2015 Index Review that “the assumptions [required by page 700] should reflect established ratemaking practices and

81
AOPL Reply Comments at 13-14.

thus should be consistent enough to accurately calculate the index.”
82

The D.C. Circuit's use of “consistent” refers to pipelines' consistent compliance with

the Commission's prevailing policies in their page 700 filings, not, as AOPL argues, that the index level cannot reflect policy changes that occur during the five-year review period.
83

Moreover, as discussed above, the index should reflect industry-wide changes to recoverable costs such as those caused by the Income Tax Policy Change—thus, it is appropriate for the 2014 page 700 data to include income tax allowances for MLPs while the 2019 page 700 data does not.

82

AOPL III,
876 F.3d at 345 (citing 2015 Index Review, 153 FERC ¶ 61,312 at P 18).

83

AOPL III,
876 F.3d 345;
see also
2015 Index Review, 153 FERC ¶ 61,312 at P 18 (“The allocation methodologies used by pipelines on page 700 should reflect established ratemaking practices, and thus these allocation methodologies should be sufficiently robust to calculate the index. . . . [T]o the extent a pipeline's page 700 ratemaking assumptions change over a period of time, pipelines are obligated to note them on their page 700.”). Pipelines that were MLPs consistently claimed an income tax allowance in 2014 and consistently did not claim an income tax allowance in 2019.

32. Finally, we reject Designated Carriers' remaining claims as irrelevant, unsupported, or without merit. Designated Carriers incorrectly claim that income tax allowance costs should be removed from the 2014 page 700 data for MLP pipelines because the Commission has previously found that partnership investors' income tax costs are not properly considered costs in a partnership pipeline's regulated cost of service.
84

To the contrary, the Commission and the D.C. Circuit have concluded that MLP pipelines incur investor-level income tax costs that are already reflected in the pipeline's DCF ROE, such that including an income tax allowance in the pipeline's cost of service alongside the ROE results in an impermissible double recovery.
85

Accordingly, the issue in this proceeding is whether the index level and the resulting pipeline rates should reflect the elimination of that double recovery. As discussed above, we find that by adjusting the data set to eliminate MLP pipelines' 2014 income tax allowances, Designated Carriers' proposal would allow the income tax double recovery to persist in pipeline rates.

84
Designated Carriers Initial Comments at 10-11 (citing Opinion No. 511-C, 162 FERC ¶ 61,228 at P 28; Webb Initial Aff. P 8).

85
Opinion No. 511-C, 162 FERC ¶ 61,228 at P 22,
aff'd, SFPP,
967 F.3d at 795-97;
see also United Airlines,
827 F.3d at 136. Moreover, Designated Carriers misconstrue the applicable law. Neither the D.C. Circuit nor the Commission have held that these costs are not properly included in a partnership pipeline's cost of service. Rather, both
United Airlines
and the 2018 Income Tax Policy Statement concluded that partnership investors' income tax costs are already recovered by the ROE and that allowing partnership pipelines to recover an income tax allowance in addition to that ROE would impermissibly double recover those costs.
United Airlines,
827 F.3d at 135-37; 2018 Income Tax Policy Statement, 162 FERC ¶ 61,227 at PP 8-9, 45.

33. Designated Carriers misconstrue Commission precedent in arguing that Pipelines' proposed adjustments accord with the Commission's actions applying the Income Tax Policy Change retroactively in the 2018 Income Tax Policy Statement and in Docket Nos. IS08-390 and IS09-437.
86

In the 2018 Income Tax Policy Statement, issued on March 15, 2018, the Commission applied the policy change
prospectively
by directing pipelines to report their income tax costs in accordance with its revised policy in their upcoming Form No. 6 filings due for submission on April 18, 2018, which would include cost-of-service data for 2017 and 2016.
87

The Commission did not apply the new policy retroactively to periods before the years encompassed by those impending filings. In Docket Nos. IS08-390 and IS09-437, the Commission applied its revised income tax allowance policy in pending cost-of-service rate proceedings to the time periods at issue,
88

which predated the 2018 Income Tax Policy Change.
89

Contrary to Designated Carriers' claim, applying the Commission's new policy to a pipeline whose rates were the subject of pending proceedings involving earlier time periods does not support applying that policy retroactively to revise the reported cost-of-service data of pipelines whose rates were not the subject of ongoing litigation.

86
Designated Carriers Initial Comments at 9, 11-12, 14-15 (citing
SFPP, L.P.,
162 FERC ¶ 61,229, at P 8 (2018); Opinion No. 511-C, 162 FERC ¶ 61,228 at PP 28, 54-57; 2018 Income Tax Policy Statement, 162 FERC ¶ 61,227 at PP 8, 46, n.83; Webb Initial Aff. PP 9, 11). AOPL echoes this argument in its reply comments. AOPL Reply Comments at 18-19.

87
2018 Income Tax Policy Statement, 162 FERC ¶ 61,227 at P 46 n.83.

88

See SFPP, L.P.,
Opinion No. 435, 86 FERC ¶ 61,022, at 61,093-94 (1999) (“Commission practice is to base its decision on the policy in effect in the year a regulatory decision is made, and then apply that decision to the time frame to which the case applies.”);
see also Consol. Edison Co. of N.Y.
v.
FERC,
315 F.3d 316, 323-24 (D.C. Cir. 2003) (explaining that an agency may apply a new substantive rule to decide a pending proceeding).

89
The Docket No. IS08-390 proceeding addressed SFPP's West Line rates to be effective August 1, 2008. Opinion No. 511-C, 162 FERC ¶ 61,228 at P 4. The Docket No. IS09-437 proceeding addressed SFPP's East Line rates to be effective January 1, 2010.
SFPP, L.P.,
Opinion No. 522-B, 162 FERC ¶ 61,229 at P 8.

34. Designated Carriers' claim that reflecting the Income Tax Policy Change in the index calculation would constitute retroactive ratemaking likewise lacks merit.
90

The rule against retroactive ratemaking “prohibits the Commission from adjusting current rates to make up for a utility's over- or under-collection in prior periods.”
91

By contrast, the five-year review uses past cost changes to calculate the index adjustment that pipelines can use to adjust their
future
rates. Accounting for reduced recoverable costs in calculating the prospective index adjustment does not modify current rates to account for prior period over- or under-recoveries and therefore does not contravene the bar against retroactive ratemaking.

90
Designated Carriers Initial Comments at 16.

91

SFPP,
967 F.3d at 801 (quoting
Old Dominion Elec. Coop.
v.
FERC,
892 F.3d 1223, 1227 (D.C. Cir. 2018)).

35. Designated Carriers also do not provide support for their contention that incorporating the Income Tax Policy Change in the index would negatively impact MLP pipelines twice, such as SFPP, whose cost-of-service rates have already been revised to remove the income tax allowance and ADIT balances.
92

As discussed above, Designated Carriers have only identified one pipeline (out of 240 pipelines filing page 700 with the Commission) whose rates have been lowered to reflect the Income Tax Policy Change and thus have not shown that this alleged harm would affect any pipeline besides SFPP.
93

More generally, the Commission calculates the index level based upon normal industry-wide cost changes, without regard to the particular experiences of individual pipelines. To do otherwise would produce nonsensical results, as indexing would cease to function as a generally applicable ratemaking methodology if the index was adjusted to account for

the particular cost changes of each individual pipeline.

92
Designated Carriers Initial Comments at 15 (citing Webb. Aff. P 14).

93
Moreover, even as to SFPP, it is unclear that incorporating the Income Tax Policy Change in the index calculation would produce the adverse effects that Designated Carriers describe. First, after the Commission adopted the Income Tax Policy Change, SFPP converted to a Schedule-C Corporation eligible to recover an income tax allowance and defended their rates on that basis in their East Line rate case in Docket No. OR16-6-000. Second, SFPP's implementation of the Income Tax Policy Change (before its conversion to a C-Corporation) actually produced an
increase
to its rates on its West Line system. In response to Opinion No. 511-C, SFPP removed the income tax allowance and previously accumulated ADIT balances from its West Line cost-of-service rates. Opinion No. 511-D, 166 FERC ¶ 61,142 at P 59. As a result, SFPP's West Line rates increased to levels above the rates established following Opinion No. 511-B, which included an income tax allowance and ADIT balances.
Compare
SFPP, Compliance Filing, Docket No. IS08-390-011, Tab A, COS Summary at 2 (filed May 14, 2018) (rates filed in response to Opinion No. 511-C),
with
SFPP, Compliance Filing, Docket No. IS08-390-008, Tab A, COS Summary at 2 (filed Apr. 6, 2015) (rates filed in response to Opinion No. 511-B). Because reflecting the Income Tax Policy Change in SFPP's West Line rates resulted in a rate increase, we are unconvinced that incorporating this policy change in the index calculation would somehow adversely impact SFPP for a second time as Designated Carriers allege.

36. Finally, to the extent that Designated Carriers argue that the Commission should have “trued up” prior index levels in the 2015 Index Review to account for the impact of the 2005 income tax policy change upon recoverable costs, this argument is unsupported.
94

Designated Carriers do not specify the type of analysis they believe the Commission should have performed in the 2015 Index Review
95

and fail to quantify the impact of this analysis upon pipelines' recoverable costs. Furthermore, any arguments concerning the Commission's actions in previous index reviews are outside the scope of this proceeding.

94
Designated Carriers Initial Comments at 17-20 (citing Webb Aff. PP 19-22).

95
To the extent that Designated Carriers argue the Commission should have retroactively revised previously established index levels to allow pipelines to recover for prior under collections in excess of their then-effective rates, this would conflict with both indexing's purpose and the filed rate doctrine.
E.g., Ark. La. Gas Co.
v.
Hall,
453 U.S. 571, 577 (1981) (explaining that the filed rate doctrine “forbids a regulated entity to charge rates for its services other than those properly filed with the appropriate federal regulatory authority” (citation omitted)). Alternatively, if they argue that the Commission should adjust the going-forward index level upward because prior index calculations did not incorporate the 2005 policy change, they have not demonstrated that the multiple income tax policy changes the Commission has adopted since it established the indexing regime, including
Lakehead
and the 2005 policy change, caused pipelines to under-recover their costs on a systematic basis.

B. Statistical Data Trimming

1. December 2020 Order

37. In the December 2020 Order, the Commission departed from its prior practice established in the 2010 and 2015 Index Reviews of using the middle 50%.
96

Instead, for the first time, the Commission relied solely upon the middle 80%. The Commission decided that it would consider more data in measuring industry-wide cost changes because using a broader sample should enhance the Commission's calculation of the central tendency of industry cost experience.
97

The Commission further stated that “normal” cost changes are best defined by using the inclusive data sample embodied in the middle 80% in order to accurately identify the central tendency of industry-wide cost changes.
98

96

See supra
note 9.

97
December Order, 173 FERC ¶ 61,245 at P 26.

98

Id.
P 27.

38. Additionally, the Commission held that “mere generalized concerns” about outlying data do not justify excluding the experiences of pipelines included in the middle 80% but not the middle 50% (
i.e.,
the incremental 30%) from the Commission's review of industry cost changes.
99

The Commission stated that unlike in prior index reviews, the record here does not contain “detailed analyses” showing that pipelines in the incremental 30% experienced anomalous cost changes that would skew the index.
100

99

Id.
P 28.

100

Id.

2. Rehearing Requests

39. Shippers argue that the December 2020 Order conflicts with precedent and fails to justify departing from the Commission's established practice of trimming the data set to the middle 50%. They first contend that using the middle 80% contravenes the Commission's findings in the 2015 and 2010 Index Reviews that the index aims to reflect normal cost changes and that the middle 50% more effectively excludes anomalous cost data than the middle 80%, which includes pipelines further removed from the median whose cost changes may result from idiosyncratic circumstances rather than ordinary pipeline operations.
101

According to Shippers, the December 2020 Order fails to distinguish those findings and instead attempts to redefine “normal” cost changes to encompass the widest possible range of data, regardless of whether that data reflects typical experience. Shippers argue that the middle 80% in this proceeding includes pipelines with anomalous cost changes and that the central tendency of a data sample that includes such unrepresentative data fails to reflect normal industry-wide cost changes.
102

In addition, Shippers dispute the December 2020 Order's conclusion that the presence of anomalous data in the middle 50% in prior reviews supports using the middle 80% in this proceeding. Shippers argue that the December 2020 Order does not demonstrate that the middle 50% includes unrepresentative data and, even if it did, this would not justify using a larger sample that likely includes more idiosyncratic data.
103

101
Joint Commenters Request for Rehearing at 23-26 (citing 2015 Index Review, 153 FERC ¶ 61,312 at PP 23, 42-44; 2010 Index Review, 133 FERC ¶ 61,228 at P 61); Liquids Shippers Request for Rehearing at 37-38 (citing 2015 Index Review, 153 FERC ¶ 61,312 at PP 42-44; 2010 Index Review, 133 FERC ¶ 61,228 at PP 60-63 & n.36).

102
Joint Commenters Request for Rehearing at 26-27; Liquids Shippers Request for Rehearing at 44-45.

103
Joint Commenters Request for Rehearing at 32.

40. Similarly, Shippers state that the December 2020 Order ignores the Commission's findings in 2015 and 2010 that trimming to the middle 50% provides a simplified and objective method for removing unrepresentative data that minimizes the need to scrutinize individual pipeline data or engage in manual data trimming.
104

Shippers assert that expanding the data sample to the middle 80% discards this simplified and effective tool for removing outliers without an adequate replacement.
105

104

Id.
at 34 (citing 2015 Index Review, 153 FERC ¶ 61,312 at P 42). In both the 2015 and 2010 Index Reviews, shipper commenters proposed manual data trimming methodologies in which they carefully reviewed the costs for each of the 150-200 pipelines in the data set to remove those pipelines with cost changes resulting from specific factors not broadly shared across the industry, such as large rate base expansions.
See
2015 Index Review, 153 FERC ¶ 61,312 at PP 19-21 (describing manual data trimming proposals); 2010 Index Review, 133 FERC ¶ 61,228 at PP 34-47 (same).

105
Joint Commenters Request for Rehearing at 33 (citing
AOPL II,
281 F.3d at 245 (vacating and remanding the Commission's determination in the 2000 Index Review to decline to engage in statistical data trimming as unjustified departure from prior practice of trimming to the middle 50%)).

41. Shippers next argue that the record in this proceeding does not support this departure from established practice and in fact provides a stronger basis for using the middle 50% than in prior index reviews. In particular, Shippers state that the middle 50% represents a greater percentage of barrel-miles subject to the index (82.2% in the NOI data set) than in 2015 (56%) or 2010 (76%),
106

whereas the middle 80% is more widely dispersed than in 2015 or 2010 and includes outlying cost increases that are not offset by comparable cost decreases.
107

Moreover, Shippers assert that the December 2020 Order acknowledged that “the record contains no evidence addressing whether the more dispersed cost changes in the incremental 30% resulted from pipeline-specific factors rather than from broadly shared circumstances representative of ordinary pipeline operations.”
108

Given the Commission's previous findings that the middle 80% more likely includes pipelines with idiosyncratic and outlying data, Shippers argue that this lack of evidence supports continued use of the middle 50%.
109

106

Id.
at 30; Liquids Shippers Request for Rehearing at 41-42.

107
Joint Commenters Request for Rehearing at 30; Liquids Shippers Request for Rehearing at 43, 46-47.

108
Joint Commenters Request for Rehearing at 38 (quoting December 2020 Order, 173 FERC ¶ 61,245 at P 29).

109

Id.
at 38-39.

42. Shippers further contend that the December 2020 Order erroneously places the burden upon shipper commenters to justify continued use of the middle 50% by faulting them for

failing to present detailed analyses of the incremental 30%.
110

Shippers state that the Commission discouraged commenters from submitting such evidence by declining to consider similar analyses in the 2015 and 2010 Index Reviews.
111

Moreover, Shippers assert that it is not incumbent upon commenters to justify continued application of the Commission's existing policy. Rather, they argue that the agency attempting to depart from a well-established practice bears the burden of explaining why the reasoning underlying that practice should no longer control.
112

Similarly, Shippers claim that it was incumbent upon Pipelines, as the proponents of a change in Commission policy, to justify the change by demonstrating that the incremental 30% does not contain outlying data. Shippers argue that Pipelines failed to make this showing and that the limited evidence in the record analyzing the incremental 30% indicates that it contains anomalous data that skews the index calculation.
113

According to Shippers, this evidence was sufficient to justify using the middle 50% consistent with established practice.
114

110

Id.
at 35-39; Liquids Shippers Request for Rehearing at 51-52.

111
Joint Commenters Request for Rehearing at 35, 37. Liquids Shippers observe, moreover, that the Commission did not rely upon such analyses when it declined to use the middle 80% in the 2015 and 2010 Index Reviews. Liquids Shippers Request for Rehearing at 51 (citing 2015 Index Review, 153 FERC ¶ 61,312 at P 43; 2010 Index Review, 133 FERC ¶ 61,228 at P 61).

112
Joint Commenters Request for Rehearing at 35-36 (citing
Encino Motorcars, LLC
v.
Navarro,
136 S. Ct. 2117, 2126 (2016);
FCC
v.
Fox Television Stations, Inc.,
556 U.S. 502, 515 (2009);
Balt. Gas & Elec. Co.
v.
FERC,
954 F.3d 279, 286 (D.C. Cir. 2020);
Air All. Houston
v.
EPA,
906 F.3d 1049, 1066 (D.C. Cir. 2018)); Liquids Shippers Request for Rehearing at 52 (citing
FCC
v.
Fox Television Stations, Inc.,
556 U.S. at 515-16).

113
Joint Commenters Request for Rehearing at 38; Liquids Shippers Request for Rehearing at 45-51.

114
Joint Commenters Request for Rehearing at 38; Liquids Shippers Request for Rehearing at 52-53.

3. Commission Determination

43. We are persuaded by Shippers' arguments on rehearing and grant rehearing of the December 2020 Order to calculate the index level based upon the middle 50%, consistent with the Commission's practice in the 2015 and 2010 Index Reviews.
115

We conclude that the record in this proceeding does not justify departing from the Commission's established practice of calculating the index level based solely upon the middle 50%.

115
2015 Index Review, 153 FERC ¶ 61,312 at PP 42-44,
aff'd, AOPL III,
876 F.3d at 342-44; 2010 Index Review, 133 FERC ¶ 61,228 at PP 60-63. Although the Commission averaged the middle 50% with the middle 80% in the 2000 and 2005 five-year reviews, it did not justify or address its consideration of the middle 80%. 2010 Index Review, 133 FERC ¶ 61,228 at P 60. Moreover, the Commission has never relied upon the middle 80% alone and provided a detailed explanation in the 2015 and 2010 Index Reviews why it would not consider the middle 80%. As the D.C. Circuit explained, “[n]othing in any of the Commission's past index review orders bound the agency to use the middle 80% of pipelines' cost-change data in any later proceeding.”
AOPL III,
876 F.3d at 353.

a. The Record in This Proceeding Supports Using the Middle 50% To Calculate the Index Level

44. As an initial matter, the objective of the index is to reflect the cost experience of a typical pipeline during ordinary pipeline operations.
116

The index is not designed to recover extraordinary cost changes, including those resulting from atypical or idiosyncratic circumstances.
117

These extraordinary cost changes are recovered using the Commission's alternate ratemaking methodologies rather than through indexing.
118

In addition, the presence of such extraordinary cost changes in the data set can inflate the index level.
119

116

E.g.,
2010 Index Review, 133 FERC ¶ 61,228 at P 61; Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,097 (“The role of an index is to accommodate normal cost changes.”).

117
The Commission has held, and the D.C. Circuit has affirmed, that use of an index sufficiently high to encompass extraordinary costs “would provide windfalls to many oil pipelines by allowing rate changes substantially above cost changes” and “effectively abdicate [the Commission's] responsibilities for rate regulation under the ICA.” Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,097,
aff'd, AOPL I,
83 F.3d at 1434;
see also
2010 Index Review, 133 FERC ¶ 61,228 at P 54 (interpreting the use of “extraordinary” in Order Nos. 561 and 561-A as referring to “pipelines experiencing changed per barrel-mile costs that were greater than the changing costs experienced by other pipelines regardless of the causes underlying any particular pipeline's cost changes.”).

118
Order No. 561-A, FERC Stats. & Regs. ¶ 31,000 at 31,097 (“Extraordinary costs can be recovered through either of the alternate rate change means—cost of service or settlement rates—as provided in [Order No. 561].”).

119
Such cost changes would impact the composite central tendency of the data sample through the weighted mean and unweighted mean, which, unlike the median, reflect the cost experiences of all pipelines in the sample, including those at the upper and lower bounds.

45. To avoid inflating the index, the Commission excludes pipelines with extraordinary or idiosyncratic cost changes from the analysis. Along these lines, in the 2010 and 2015 Index Reviews, the Commission found that the middle 50% more appropriately adjusts the index level for normal cost changes than the middle 80%, which, by definition, includes pipelines relatively far removed from the median.
120

The Commission also concluded that pipelines in the incremental 30% are more likely to have cost changes resulting from idiosyncratic factors, such as a rate base expansion, plant retirement, or localized changes in supply and demand, that do not reflect normal industry-wide experience.
121

Thus, the Commission found that the middle 50%, more effectively than the middle 80%, trims pipelines with anomalous cost changes from the data set while avoiding the complexities and distorting effects of laborious and subjective manual data trimming methodologies.
122

Following the 2015 Index Review, the D.C. Circuit affirmed the Commission's decision to trim the data set to the middle 50% instead of the middle 80%.
123

120
2010 Index Review, 133 FERC ¶ 61,228 at P 61; 2015 Index Review, 153 FERC ¶ 61,312 at P 43 (“[B]y definition, costs at the top (or bottom) of the middle 80 percent deviate significantly from the cost experience of other pipelines”);
id.
P 44 (“Pipelines in the middle 80 percent, as opposed to the middle 50 percent, are more likely to have outlying cost changes which could result from idiosyncratic factors particular to that pipeline.”).

121
2010 Index Review, 133 FERC ¶ 61,228 at P 61.

122
2015 Index Review, 153 FERC ¶ 61,312 at P 42 (citing 2010 Index Review, 133 FERC ¶ 61,228 at PP 60-63).

123

AOPL III,
876 F.3d at 342 (explaining that the court had “little difficulty in finding that the Commission adequately and reasonably justified its decision not to consider the middle 80 percent of pipelines' cost-change data” in that proceeding).

46. Upon reconsideration of the December 2020 Order, we find that the record in the instant proceeding does not justify a different result. The scatter plot below
124

demonstrates that the middle 80% in this data set includes several pipelines near its upper bound that are considerably removed from the other pipelines in the sample.

124
This scatter plot modifies a similar chart submitted by Joint Commenters. Joint Commenters Reply Comments, Brattle Group Report at 19, Figure 3 (scatter plot illustrating dispersion of the middle 50% and middle 80% in the unadjusted 2020 data set). The modifications reflect the adjustments adopted herein to the page 700 data set.

ER28JA22.000

47. Furthermore, the pipelines at the upper bound of the middle 80% exert an outsized influence that inflates the index calculation.
125

The difference between the middle 50% and the middle 80% results primarily from 8 pipelines at the upper bound of the middle 80%. Namely, expanding the data sample from the middle 50% to the middle 70%, which omits the top and bottom 8 pipelines in the middle 80%, only increases the composite central tendency by 3 basis points, from −0.21% to −0.18%.
126

By contrast, expanding the sample to include these 16 pipelines increases the composite central tendency by an additional 29 basis points, from −0.18% to 0.11%.
127

In contrast to their outsized effect on the index, the 8 pipelines at the upper bound of the middle 80% account for only 2.10% of total barrel-miles.

125
AOPL's calculations demonstrate that using the middle 80% would increase the cost change calculation by 41 basis points while only expanding the number of barrel-miles in the analysis by approximately 14%. Shehadeh Initial Decl., Exhibit A11 (calculating that the composite central tendency of the cost change data, when incorporating AOPL's proposed adjustments to remove the effects of the Income Tax Policy Change, increases from 0.90% to 1.31% when expanding from the middle 50% to the middle 80%); Shehadeh Initial Decl., Exhibit A12 (stating that the middle 50% and middle 80% contain 81.9% and 95.8%, respectively, of total barrel-miles in 2014 subject to the index).

126
As discussed above, the Kahn Methodology calculates a composite central tendency by averaging the data sample's median, weighted mean, and unweighted mean.
See supra
P 4.

127
Attach. A, Exhibit 14. The outsized impact these pipelines exert upon the index calculation undermines the conclusion in the December 2020 Order that the dispersion of the middle 80% is not relevant because it results from “just a few pipelines at the top of the middle 80%.” December 2020 Order, 173 FERC ¶ 61,245 at P 29. Furthermore, this analysis rebuts the statement in the December 2020 Order that the record did not contain a “detailed showing” that using the additional data in the middle 80% would distort the index calculation.
Id.

Not only does the middle 80% include pipelines at its upper bound that diverge considerably from the other pipelines in the sample, but the record further establishes that the middle 80% as a whole is even more dispersed than in 2015 or 2010,
128

as illustrated in the bar chart below.
129

128
When the data sample is highly dispersed, data at the outer bounds of the middle 80% are further removed from the remaining data and thus can have an outsized and distorting effect if used to measure the central tendency.

129
The bar chart modifies a similar chart submitted by Joint Commenters. Joint Commenters Reply Comments, Brattle Group Report at 18, Figure 2 (bar chart illustrating dispersion of middle 50% and middle 80% in 2010, 2015, and the unadjusted 2020 data sets). The modifications reflect the adjustments adopted herein to the page 700 data set.

ER28JA22.001

48. Additionally, AOPL has presented no evidence that the middle 80% in this proceeding lacks the type of atypical and idiosyncratic cost changes observed in the middle 80% in the 2015 and 2010 Index Reviews.
130

To the contrary, the record demonstrates that the additional data included in the incremental 30% contains pipelines with idiosyncratic cost changes resulting from circumstances that are not broadly shared across the industry. For example, Joint Commenters identify 7 pipelines in the incremental 30% whose reported cost changes resulted from irregular circumstances or specific factors not broadly shared across the industry, such as temporary shutdowns or pipeline ruptures.
131

130
AOPL, the proponent of changing the Commission's policy to use the middle 80% instead of the middle 50%, had the opportunity to provide such evidence in its initial comments and reply comments.
See
5 U.S.C. 556(d) (“Except as otherwise provided by statute, the proponent of a rule or order has the burden of proof.”);
P.R.
v.
Fed. Mar. Comm'n,
468 F.2d 872, 881 (D.C. Cir. 1972) (“Ultimately, the rule requiring the proponent of an order to sustain the burden of its justification rests on the policy of requiring a person seeking a change from the status quo to take on the burden of justifying the change.”);
see also S. Ga. Nat. Gas Co.,
73 FERC ¶ 61,354, at 62,106 (1995) (“[W]here there is a `settled practice,' the proponent of a change to that practice has the burden of proof.”).

131
Joint Commenters Reply Comments, Brattle Group Report at 13-17. For example, PMI Services North America, Inc., reported an inflated 2019 cost of service per barrel-mile due to a temporary shutdown of one of its pipeline segments and Mobil Pipe Line Company experienced a pipeline rupture in 2013 that distorted its 2014 cost-of-service data.
Id.
at 15-17.

49. In sum, the record demonstrates that the middle 80% in this proceeding includes pipelines with extraordinary cost changes that are not reflective of ordinary pipeline operations. Accordingly, we find that for purposes of calculating the index level in this proceeding, using the more tailored data sample embodied by the middle 50% produces a more accurate measure of “normal” cost changes and minimizes the risk that the index will be distorted by pipelines with unrepresentative cost experiences. Pipelines have not demonstrated why the instant record is distinguishable from the 2015 and 2010 Index Reviews such that the Commission should depart from the data trimming methodology it employed in those proceedings.

b. Reconsidering the December 2020 Order

50. We believe the reasons given in the December 2020 Order for using the middle 80% in this proceeding to be in error.

51. First, the mere fact that the middle 80% includes additional data does not support departing from the middle 50%.
132

The middle 50% already includes 81% of industry-wide oil pipeline barrel-miles,
133

which is significantly more than the barrel-miles used in prior index reviews.
134

Moreover, the middle 80% only incorporates an additional 15% more of the industry's barrel-miles. Thus, although using the middle 50% excludes 48 pipelines from the cost-change analysis,
135

omitting these pipelines does not deprive the Commission of a robust data sample. Moreover, any benefits of considering the larger data sample do not outweigh the risk, discussed above, that this additional data will distort the measurement of normal cost changes.

132

See AOPL III,
876 F.3d at 343 (noting the Commission has “rejected the precise principle” that the middle 80% is preferable because it includes a larger number of pipelines) (citing 2010 Index Review, 133 FERC ¶ 61,228 at PP 57, 61); 2015 Index Review, 153 FERC ¶ 61,312 at P 44 (rejecting argument that “the middle 80 percent should be used merely because it contains more barrel-miles”).

133
Attach. A, Exhibit 1.

134
In the 2015 and 2010 Index Reviews, the Commission concluded that it was “unnecessary to include the middle 80 percent to obtain a representative sample of the data” where the middle 50% included 56% and 76%, respectively, of total barrel-miles subject to the index. 2010 Index Review, 133 FERC ¶ 61,228 at P 63;
see also
2015 Index Review, 153 FERC ¶ 61,312 at P 44 n.85 (concluding that the fact that the middle 50% included a lower percentage of barrel-miles than in 2010 “is not a sufficient basis to risk including more outlying data”),
aff'd, AOPL III,
876 F.3d at 344.

135
December Order, 173 FERC ¶ 61,245 at P 26.

52. Second, we disagree with the December 2020 Order and find that for purposes of this proceeding, “normal” cost changes are best measured using a more tailored data sample that excludes the anomalous and idiosyncratic data in the middle 80%.
136

For the reasons discussed above,
137

this record demonstrates that “including data from the middle 80% distorts our measurement of the industry-wide central tendency [used to calculate the index level].”
138

Rather, using the middle 50% is more consistent with the index's purpose of allowing recovery for normal cost changes, not extraordinary costs.

136
We disagree with the statement in the December 2020 Order that using the middle 80% is appropriate because the index average will be significantly below the relatively high cost changes at the upper bound.
Id.
PP 27, 32. Even if the index average is not set at the upper bound of the data sample, including the upper bound of the middle 80% nonetheless produces an index average inflated by anomalous cost experience.
See
2010 Index Review, 133 FERC ¶ 61,228 at P 61 (“Using the middle 50[%] ensures that pipelines with relatively large cost increases or decreases do not distort the index.”).

137

See supra
PP 46-50.

138
December 2020 Order, 173 FERC ¶ 61,245 at P 27. The December 2020 Order erroneously implied that entities supporting continued use of the middle 50% must provide a “compelling showing” that using the middle 80% would distort the calculation of the index level.
Id.
Although the record here provides such a compelling showing, we clarify that entities advocating for a departure from the Commission's practice of using the middle 50% bear the burden of justifying that change.
See supra
note 129.

53. Third, in the December 2020 Order, the Commission sought to distinguish the 2010 and 2015 Index Reviews on the basis that, unlike in the instant review, commenters in those proceedings “presented detailed analyses demonstrating that the incremental 30% contained anomalous cost changes . . . .”
139

However, we no longer find this reasoning persuasive because, as in prior reviews, the present record demonstrates the middle 80% includes outlying cost increases, reflects significant dispersion, and includes pipelines with idiosyncratic cost changes. Although shippers submitted more detailed analyses in 2010 and 2015, they presented this evidence to support manual data trimming proposals that required a labor-intensive pipeline-by-pipeline analysis of page 700 data. Finding manual data trimming to be highly subjective, the Commission rejected this approach because “[a]ny potential improvement from manual data trimming is outweighed by the increase in the potential for error or manipulation.”
140

Rather, the Commission concluded that, instead of manual data trimming, using the middle 50% more effectively addressed those same issues in a manner that was more consistent with simplified and streamlined ratemaking.
141

We conclude that it would be incongruous to reject such manual data trimming while at the same time requiring commenters to present similar analyses to justify continued use of the middle 50%.
142

139
December 2020 Order, 173 FERC ¶ 61,245 at P 28.

140
2015 Index Review, 153 FERC ¶ 61,312 at P 34;
see also id.
PP 36, 42; 2010 Index Review, 133 FERC ¶ 61,228 at P 62.

141
2015 Index Review, 153 FERC ¶ 61,312 at PP 36, 42; 2010 Index Review, 133 FERC ¶ 61,228 at P 62.

142
In any case, the December 2020 Order overstates the absence of evidence regarding anomalous data among the 48 pipelines in the incremental 30%. Acknowledging that Shippers identified 7 pipelines, the December 2020 Order stated that for the remaining 41 there is no evidence of anomalous data. December 2020 Order, 173 FERC ¶ 61,245 at P 28. However, this ignores the chart above that examined the entire middle 80% and showed how those pipelines at the top of the middle 80% were inflating the index level.

c. AOPL's Remaining Arguments Are Not Persuasive

54. We reject AOPL's remaining arguments in support of using the

middle 80% as unpersuasive. First, AOPL erroneously claims that the Commission should use the middle 80% based upon its previous recognition that “it is preferable to apply the larger data set when the additional data is available using the Kahn Methodology.”
143

However, the D.C. Circuit rejected this exact argument following the 2015 Index Review, finding that the quoted language “addressed FERC's approach to selecting the pool of pipelines whose costs should be measured at all—not the portion of the resulting data to trim before calculating the normal industry change in costs.”
144

Further, the court explained that the Commission had in fact rejected the argument that it is preferable to use a larger data sample merely because additional data is available. Instead, the Commission concluded that the middle 50% more appropriately adjusts the index level for normal cost changes, notwithstanding the fact that it contains less data than the middle 80%.
145

We reject AOPL's argument for the same reasons here.

143
AOPL Initial Comments at 19-20 (quoting 2010 Index Rehearing Order, 135 FERC ¶ 61,172 at P 41).

144

AOPL III,
876 F.3d at 343 (citing 2010 Index Rehearing Order, 135 FERC ¶ 61,172 at P 41 & n.38).

145

Id.
(citing 2010 Index Review, 133 FERC ¶ 61,228 at PP 57, 61).

55. Second, we dismiss AOPL's claim that the middle 80% provides a more accurate measure of industry cost changes merely because it resembles a lognormal distribution.
146

As the Commission found in the 2015 Index Review and as the D.C. Circuit affirmed, to the extent that the middle 80% data conforms to a lognormal distribution, outlying cost increases per barrel-mile will not be offset by similarly outlying cost decreases.
147

This concern is illustrated in the instant record, where the middle 80% includes multiple pipelines with cost increases above 100% and no pipelines with cost decreases of negative 100%.
148

Thus, using the middle 80% would skew the index upward based upon these outlying cost increases, which is contrary to the index's objective of reflecting normal cost changes.
149

146
AOPL Initial Comments at 20-21; AOPL Reply Comments at 8-9 (citing Shehadeh Initial Decl. at 24). A lognormal distribution is a continuous probability distribution of a random variable whose logarithm is normally distributed.

147
2015 Index Review, 153 FERC ¶ 61,312 at P 43 (“using the middle 80 percent would skew the index upward based upon these outlying cost increases, which is contrary to the objective of the index to reflect normal industry-wide cost changes”),
aff'd, AOPL III,
876 F.3d at 344.

148

See
Liquids Shippers Reply Comments at 24 (citing Crowe Reply Aff. at 4-5).

149
We also question the mathematical reasoning underlying AOPL's argument. Specifically, a lognormal distribution occurs when performing a natural logarithm transformation of a data set produces a normal distribution. However, it is not possible to take the natural logarithm of negative numbers.
Id.
at 24-25. Because the data set here contains negative numbers, it cannot be lognormally distributed.

56. Third, AOPL misconstrues Commission precedent in claiming that reliance on the middle 50% is only appropriate where there are concerns of erroneous data.
150

Although use of the middle 50% in Order No. 561 was based in part upon concerns about erroneous data, the Commission has relied upon the middle 50% to exclude not only inaccurate data, but also extraordinary data that is unrepresentative of normal cost experience.
151

As the D.C. Circuit explained when upholding the Commission's continued use of the middle 50% in the 2015 Index Review, the Commission provided extensive justification for its ongoing reliance on the middle 50% in both the 2010 and 2015 Index Reviews.
152

Thus, even where the reported page 700 data is accurate, it remains necessary to use the middle 50% to avoid including outlying data that exerts a disproportionate impact on the index calculation.

150
AOPL Initial Comments at 21-22 (citing Shehadeh Initial Decl. at 21-22).

151

See
2010 Index Review, 133 FERC ¶ 61,228 at P 61 (“Even when accurate data is reported, pipelines in the middle 80, as opposed to the middle 50, are more likely to have cost changes resulting from factors particular to that pipeline, such as a rate base expansion, plant retirement, or localized changes in supply and demand.”).

152

See AOPL III,
876 F.3d at 343 (rejecting AOPL's argument that the Commission was precluded from excluding the middle 80% when “that data is available and accurate”);
id.
at 339 (“[C]ontrary to AOPL's assertion, nothing in any of FERC's past index review orders bound the agency to use the middle 80 percent of pipelines' cost-change data.”).

57. In sum, we conclude that the evidence does not support departing from the Commission's established practice of trimming the data set to the middle 50%. Pipelines have presented the same arguments that the Commission rejected in the 2010 Index Review and that the Commission and the D.C. Circuit rejected in the 2015 Index Review. Pipelines also presented no evidence demonstrating that the middle 80% contains fewer pipelines with idiosyncratic cost changes than in 2010 and 2015. Moreover, as articulated above, the record in this proceeding provides less support for using the middle 80% than in 2015 or 2010 because the middle 50% includes a considerably higher percentage of industry-wide barrel-miles (81% in 2020 versus 76% in 2010 and 56% in 2015) and the middle 80% of this data set is more dispersed. We therefore grant Shippers' requests for rehearing to calculate the index level using the middle 50%.
153

153
Consistent with the Commission's historical practice, nothing in this order precludes commenters from proposing modifications to the Kahn Methodology, including different data trimming methodologies, in future five-year reviews based upon the records in those proceedings.
See
NOI, 171 FERC ¶ 61,239 at P 8 (“We invite interested persons to submit comments regarding . . . any alternative methodologies for calculating the index level for the five-year period commencing July 1, 2021. Commenters may address issues that include, but are not limited to, different data trimming methodologies . . . .”).

C. Liquids Shippers' Proposal To Calculate the Composite Measure of Central Tendency Using the Weighted Median

58. Liquids Shippers argued in their comments that the weighted mean of the data set in this proceeding accords undue weight to two pipelines, Colonial and Enbridge Energy, L.P. (Enbridge). Liquids Shippers asserted that these pipelines are substantial outliers in terms of barrel-miles and cost changes
154

and that both reported inaccurate page 700 data for 2014 and 2019.
155

Because the weighted mean affords these pipelines significant weight, Liquids Shippers argued that using it to calculate the composite measure of central tendency will skew the index upwards and fail to track normal industry-wide cost changes.
156

To remedy this issue, Liquids Shippers proposed to replace the weighted mean in the index calculation with the median of the barrel-mile weighted cost changes in the middle 50% (weighted median), as calculated by their witness Elizabeth H. Crowe. Alternatively, if the Commission decides not to replace the weighted mean with the weighted median, Liquids Shippers proposed reducing the weighting afforded to the weighted mean in the Kahn Methodology from 33.3% to 20% or 10%.
157

154
Liquids Shippers Initial Comments at 13-15.

155

Id.
at 17-19.

156

Id.
at 16-19.

157

Id.
at 20 n.45; Crowe Initial Aff. at 8-9.

1. December 2020 Order

59. The December 2020 Order declined to adopt Liquids Shippers' proposals. First, the Commission found that removing the weighted mean from the index calculation would conflict with longstanding Commission precedent relying upon the weighted mean and with Dr. Kahn's testimony in the Order No. 561 rulemaking proceeding endorsing its use.
158

Second, the Commission explained that the index aims to track cost changes among

pipelines of all sizes and that discarding the weighted mean or reducing the weighting it receives in the analysis would upset the balance between large and small pipelines that the Kahn Methodology achieves.
159

Third, the Commission determined that Liquids Shippers' calculation of the weighted median was methodologically flawed and did not provide a useful measure of central tendency for purposes of calculating the index.
160

Fourth, the Commission concluded that Liquids Shippers' challenges to Colonial's and Enbridge's page 700 data are misplaced and unavailing on the merits.
161

158
December 2020 Order, 173 FERC ¶ 61,245 at P 36.

159

Id.
P 37.

160

Id.
PP 38-39.

161

Id.
P 40.

2. Rehearing Request

60. Liquids Shippers renew their arguments that Colonial and Enbridge are outliers in terms of cost changes
162

and barrel-miles
163

and that these pipelines reported inaccurate page 700 data for 2014 and 2019.
164

As a result, Liquids Shippers argue that using the weighted mean in this proceeding skews the index level upwards, fails to reflect industry-wide cost changes, and increases the likelihood that inaccurate or erroneous page 700 data will distort the index calculation.
165

Liquids Shippers argue that the December 2020 Order failed to address their evidence that Colonial and Enbridge are outliers in terms of barrel-miles or acknowledge the errors in those pipelines' page 700 data. Although the Commission has previously declined to consider challenges to individual pipelines' page 700 inputs, Liquids Shippers state that this proceeding is distinct because of the substantial weight the weighted mean accords to Colonial and Enbridge.
166

162
Liquids Shippers Request for Rehearing at 56-57. Liquids Shippers assert that Enbridge and Colonial reported annual cost changes of 3.1% and 4.3%, respectively, both of which exceed the median of the data set (0.05%), the unweighted mean of the middle 80% (1.45%), and the unweighted mean of the middle 50% (0.29%).
Id.
(citing December 2020 Order, 173 FERC ¶ 61,245 at Workpapers, Exhibit 5 Tab, Column P, Lines 21 and 35;
id.
at Workpapers, Exhibit 1 Tab, Column F, Lines 11-12;
id.
at Workpapers, Exhibit 5 Tab, Column Q, Line 184).

163
Specifically, Liquids Shippers state that Colonial and Enbridge represent 40% of the total barrel-miles in the untrimmed data set of 160 pipelines and 48% of the total barrel-miles in the middle 50% sample used in the NOI. Liquids Shippers Request for Rehearing at 54-55.

164

Id.
at 65-66.

165

Id.
at 58, 65-67. Liquids Shippers state that removing Enbridge and Colonial from the data set would cause the index level proposed in the NOI to decrease from PPI-FG+0.09% to PPI-FG-0.34%.
Id.
at 57 (citing Liquids Shippers Initial Comments at 15-16; Crowe Initial Aff. at 6-7). Given this effect, Liquids Shippers argue that affording these pipelines significant weight will skew the index upward.
Id.
at 58.

166

Id.
at 67.

61. Liquids Shippers further argue that the December 2020 Order erred in relying upon earlier index proceedings to justify using the weighted mean in this case. Liquids Shippers contend that this proceeding is distinguishable from prior five-year reviews because the weighted mean is heavily influenced by just two pipelines and a commenter has demonstrated that two outlying pipelines skew the weighted mean.
167

Furthermore, Liquids Shippers state that there is limited judicial and Commission precedent addressing use of the weighted mean and that existing precedent supports only the use of some weighted measure of central tendency.
168

Liquids Shippers maintain that they do not object to the Commission taking pipeline size into account or according additional weight to larger pipelines when calculating the index level, so long as two pipelines like Colonial and Enbridge are not permitted to skew the result.
169

167

Id.
at 64, 67.

168

Id.
at 62-63 (quoting
AOPL II,
281 F.3d at 241).

169

Id.
at 63.

62. In addition, Liquids Shippers object to the December 2020 Order's suggestion that shippers should challenge the inputs in a particular pipeline's page 700 by filing a complaint.
170

Liquids Shippers state that a cost-of-service complaint against a pipeline's base rates is unlikely to result in changes to its page 700 and that there would be no commercial benefit for a shipper to file a complaint for the sole purpose of challenging the pipeline's page 700 inputs.
171

Liquids Shippers argue that by requiring shippers to challenge page 700 inputs in a complaint or litigated rate proceeding, the Commission is insulating pipelines' page 700 data from meaningful review.
172

170

Id.
at 68 (citing December 2020 Order, 173 FERC ¶ 61,245 at P 40 n.87).

171

Id.
at 68-69.

172

Id.
at 69.

3. Commission Determination

63. We are unpersuaded by Liquids Shippers' arguments and deny rehearing. As the December 2020 Order explains, replacing the weighted mean in the calculation of the composite central tendency would contravene longstanding Commission practice dating to the rulemaking proceeding that established the indexing regime.
173

As discussed below, although no commenter has previously challenged the use of the weighted mean in the Kahn Methodology, we find that Liquids Shippers have not justified departing from the Commission's well-established policy.
174

173
December 2020 Order, 173 FERC ¶ 61,245 at P 36.

174

See supra
note 129.

64. As an initial matter, Liquids Shippers acknowledge that the Kahn Methodology appropriately relies upon a weighted measure of central tendency
175

but fail to propose a credible alternative to the weighted mean. As discussed above, the December 2020 Order rejected Liquids Shippers' proposed weighted median calculation as methodologically flawed. The Commission explained that the established statistically appropriate method for calculating the weighted median, as applied to pipeline cost changes, is to order the pipelines by cost-change percentage, compute each pipeline's share of total barrel-miles, and identify the pipeline whose share of total barrel-miles causes the cumulative share to reach 50%.
176

However, rather than identify the pipeline that causes the cumulative share of total barrel-miles represented in the sample to reach 50%, Ms. Crowe derives the median value of the weighted cost-change percentages for 2019 without regard to the barrel-miles represented above and below that cost change.
177

Unlike the correct calculation of the weighted median, Ms. Crowe does not order pipelines by cost changes, and instead orders them by cost change times barrel-miles.
178

The Commission found that under this approach, it is unclear whether the median pipeline of a given sample reported (a) relatively high cost changes and low barrel-miles or (b) relatively low cost changes and high barrel-miles.
179

The Commission also observed that a small shift in the data sample's median would produce significant and multidirectional changes in the calculation's result.
180

Thus, the

Commission determined that this calculation produces “haphazard results” that “do not reflect a convergence towards a central tendency of industry-wide cost changes.”
181

The Commission further explained that Ms. Crowe's methodology would “nullify the influence of larger pipelines upon the index calculation and thereby defeat the purpose of relying upon a weighted measure of central tendency.”
182

On rehearing, Liquids Shippers do not address these findings or attempt to rectify the identified flaws in Ms. Crowe's weighted median calculation. Thus, even if we were inclined to replace the weighted mean with a different weighted measure of central tendency, Liquids Shippers present no credible alternative.

175
Liquids Shippers Request for Rehearing at 63.

176
December 2020 Order, 173 FERC ¶ 61,245 at P 38 (citing Shehadeh Reply Decl. at 11 & App. B, Ex. 1). In fact, as explained in the December 2020 Order, the pipeline reflecting the weighted median using such a calculation would be Enbridge (which as discussed below, Liquids Shippers allege should be removed as an outlier from the data set).
Id.
P 40.

177

Id.
P 39.

178

Id.

179

Id.
n.84.

180
For example, a median reflecting the pipeline with the next lowest weighted percentage change (Wildcat Liquids Caddo LLC) would reduce Ms. Crowe's result from 0.57% to −1.74% (a decrease of over 200%), whereas a median reflecting the next highest weighted percentage change (reported by Wesco Pipeline, LLC) would reduce the result by an even greater amount, from −0.57% to −2.28% (a decrease of 400%).
Id.
n.85.

181

Id.

182

Id.
n.86 (citing
AOPL II,
281 F.3d at 241). Specifically, the Commission explained that because Ms. Crowe orders the pipelines by barrel-mile cost change times barrel-miles, a pipeline with high barrel-miles would likely only lie near the median of the data sample if it reported extremely low cost changes.
Id.

65. In addition, we remain unpersuaded by Liquids Shippers' claims that the weighted mean needs to be modified or replaced because two large pipelines, Colonial and Enbridge, allegedly skew the index calculation. First, the December 2020 Order found that the record indicates that neither Colonial nor Enbridge reported outlying cost changes,
183

and Liquids Shippers do not refute these findings on rehearing.
184

Although both Colonial and Enbridge reported barrel-mile cost changes above the median in the middle 50%, this does not make them outliers in terms of cost changes.
185

Second, the fact that the weighted mean in this proceeding ascribes additional weight to two pipelines with high barrel-miles does not support removing this measure of central tendency or reducing its weighting in the Kahn Methodology. Rather, the Kahn Methodology includes the weighted mean in the calculation of central tendency specifically to provide appropriate weight to large pipelines like Colonial and Enbridge whose cost changes are highly reflective of industry cost experience.
186

This additional weighting is necessary to ensure that “minor firms do not skew the result.”
187

Because unweighted measures of central tendency weight all cost changes equally without regard to pipeline size, failing to incorporate a weighted measure would allow the cost experiences of small pipelines to obscure the experiences of pipelines that represent a much larger share of the industry's barrel-miles. In this proceeding, for instance, three small pipelines representing 0.00073% of the barrel-miles in the middle 50% influence the sample's unweighted mean by the same degree as Colonial and Enbridge, which represent 50.04% of the barrel-miles in the middle 50%.
188

Thus, the fact that the weighted mean accords significant weight to Colonial and Enbridge is fully consistent with its role in the index calculation and does not skew the index calculation as Liquids Shippers allege.
189

To the extent that Liquids Shippers oppose use of the weighted mean in this proceeding because it provides significant weighting to the two largest pipelines,
190

we find that this concern does not justify eliminating the weighted mean from the index calculation in the absence of a credible alternative.
191

183
The Commission observed that both Colonial and Enbridge are included in the middle 50% of cost changes, which indicates that their cost experiences did not diverge significantly from industry norms. December 2020 Order, 173 FERC ¶ 61,245 at P 40.

184

See
Liquids Shippers Request for Rehearing at 65-66 (acknowledging the Commission's findings but arguing that they do “not respond to [Liquids Shippers'] evidence or [their] concerns that Enbridge Energy and Colonial skew the index due to being extreme outliers in terms of barrel-miles . . . .”).

185
The 2014-2019 cost changes in the middle 50% ranged from −32.23% to 28.97%. Colonial's cost change of 23.72% lies well within the middle 50%'s upper bound, while Enbridge's cost change of 3.43% lies close to the median of the sample.

186
December 2020 Order, 173 FERC ¶ 61,245 at P 37.

187

AOPL II,
281 F.3d at 241.

188
Whereas removing the cost changes of Colonial and Enbridge would reduce the unweighted mean by 7 basis points (from −0.20% to −0.27%), removing the cost changes of Wesco Pipeline LLC, Hilcorp Pipeline Company, LLC, and Black Bear Liquids LLC increases the unweighted mean by the same magnitude of 7 basis points (from −0.20% to −0.13%). Attach. A, Exhibit 13.

189
December 2020 Order, 173 FERC ¶ 61,245 at P 37.

190
Liquids Shippers Request for Rehearing at 63, 65-66.

191
As discussed above, although Liquids Shippers contend that another approach to weighting pipeline cost changes may achieve a better balance between large and small pipelines, they have not justified an alternative to the weighted mean.

66. Moreover, we continue to find that Liquids Shippers' challenges to the reported page 700 data of Colonial and Enbridge are outside the scope of this proceeding. As the December 2020 Order explains, indexing proceedings are not an appropriate forum for challenging specific pipelines' page 700 inputs.
192

In the five-year review, the Commission must review pipeline cost changes on an industry-wide basis to establish the generic index that pipelines may use to adjust their rates going forward. Allowing commenters to litigate individual pipelines' page 700 inputs would risk expanding this review into a wide-ranging rate proceeding involving complex cost-of-service issues that would require significant time to resolve. Given that the Commission must consider industry-wide cost changes based upon data for over 160 pipelines and must complete each five-year review in order to establish the index level for use in index filings to be effective on July 1 of the following year,
193

it would be unworkable to permit challenges to individual pipeline page 700 inputs in this proceeding.

192
December 2020 Order, 173 FERC ¶ 61,245 at P 40;
see also AOPL I,
83 F.3d at 1437 (holding that the Commission did not err in Order No. 561 by declining to periodically review individual pipeline costs and instead requiring shippers to challenge individual pipeline rates via protests or complaints);
Calnev Pipe Line L.L.C.,
127 FERC ¶ 61,304, at P 5 (2009) (“[T]he Commission has made quite clear that it will not review allegations regarding the appropriateness of a pipeline's cost of service or the accuracy of its accounting in an index proceeding. Such allegations must be included in a complaint once the index-based filing becomes effective.” (citing
SFPP, L.P.,
123 FERC ¶ 61,317 (2008);
BP W. Coast Prods. LLC
v.
SFPP, L.P.,
121 FERC ¶ 61,243 (2007))).

193
NOI, 171 FERC ¶ 61,239 at P 11.

67. Furthermore, we are not persuaded by Liquids Shippers' claim that reporting errors by Colonial and Enbridge are skewing the index level upwards by 43 basis points.
194

Regarding Enbridge, this argument is particularly unpersuasive. First, removing Enbridge from the middle 50%, while retaining Colonial in that sample, actually
increases
the index level rather than decreasing it as Liquids Shippers imply.
195

Second, correcting Enbridge's alleged reporting errors only marginally influences the index calculation. Liquids Shippers claim that the 12.71% ROE that Enbridge reported on page 700 for 2019 exceeds both the 9.84% ROE that it reported for 2014 and the 10.85% ROE that many pipelines reported on page 700 for 2019. However, adjusting Enbridge's 2019 page 700 ROE from 12.71% to 9.84% or 10.85% would only impact the index level by 2 basis points.
196

194
Liquids Shippers allege that when using the data set underlying the NOI proposal, removing Enbridge and Colonial from the middle 50% reduces the index level by 43 basis points (from PPI-FG+0.09% to PPI-FG-0.34%). Liquids Shippers Request for Rehearing at 57 (citing Liquids Shippers Initial Comments; Crowe Initial Aff. at 6-7). Similarly, removing those pipelines from the middle 50% of the data set adopted in the instant order would reduce the index level by 44 basis points, from PPI-FG-0.21% to PPI-FG-0.65%. Attach. A, Exhibit 8.

195
Removing Enbridge from the middle 50% but not Colonial, increases the index level from PPI-FG-0.21% to PPI-FG-0.14%. Attach. A, Exhibit 9.

196
Lowering Enbridge's page 700 ROE from 12.71% to either 9.84% or 10.85% would reduce

the index level from PPI-FG-0.21% to PPI-FG-0.23%. Attach. A, Exhibit 10.

68. Similarly, although Colonial accounts for most of the 44 basis-point shift in the index calculation that results from removing Colonial and Enbridge from the middle 50%, correcting Colonial's alleged reporting errors produces only a
de minimis
change in the index level. Liquids Shippers argue that Colonial reported in its 2014 and 2019 page 700 filings that it is 92% financed by equity, but reported on its balance sheet and in an ongoing rate proceeding that it is 100% financed by debt.
197

However, adjusting Colonial's capital structure to 50% equity and 50% debt produces a mere one-basis-point change to the index level.
198

Accordingly, given these relatively minor effects, we are unpersuaded by Liquids Shippers' claim that using the weighted mean in this proceeding increases the likelihood that page 700 reporting errors will skew the index calculation.

197
Liquids Shippers Request for Rehearing at 59-60 (citing Crowe Initial Aff. at 5-6).

198
Using the data set adopted in this proceeding, adjusting Colonial's capital structure to 50% equity and 50% debt while preserving the composition of the middle 50% increases the index level by one basis point, from PPI-FG-0.21% to PPI-FG-0.20%. Attach. A, Exhibit 11.

69. Furthermore, we find that requiring shippers to challenge page 700 inputs outside of the five-year review process does not present an infeasible approach. First, Liquids Shippers' argument that a cost-of-service complaint is unlikely to result in a change to the pipeline's page 700 reporting is without merit. For example, if the Commission determines in a cost-of-service rate proceeding that a pipeline set its rates based upon an inaccurate capital structure, the pipeline would be required to implement this determination in its subsequent page 700 reporting.
199

Second, we are unpersuaded by Liquids Shippers' claim that a complaint challenging a pipeline's page 700 inputs would bring shippers “no commercial benefits.”
200

Where a shipper believes that a pipeline may have reported inaccurate or erroneous information on its page 700, initiating a complaint proceeding provides the parties and the Commission with a full opportunity to develop an evidentiary record that would allow for a meaningful review of the challenged page 700 inputs. If the complaint is successful, the Commission would direct the pipeline to revise its page 700 to correct any errors or inaccuracies. These revisions, in turn, could alter the cost and revenue data on which shippers and the Commission rely in evaluating cost-of-service complaints against the pipeline's rates and challenges to the pipeline's annual index rate changes. Thus, although we recognize the burden and expense associated with filing a complaint, we disagree with Liquids Shippers' claim that there would be no commercial benefits to filing a complaint against a pipeline's page 700 inputs.

199
The instructions on page 700 require pipelines to determine their page 700 inputs consistent with the Opinion No. 154-B cost-of-service methodology. To comply with this instruction, a pipeline must adhere to the Commission's application of the Opinion No. 154-B methodology in proceedings involving the pipeline's rates.

200
Liquids Shippers Request for Rehearing at 84.

D. Liquids Shippers' Proposal To Adopt Standardized ROEs for 2014 and 2019

70. Liquids Shippers argued in their comments that the reported page 700 ROEs conflict with the Commission's cost-of-service ratemaking methodology because they are self-reported and vary substantially.
201

In addition, Liquids Shippers maintained that uncertainty surrounding the Commission's oil pipeline ROE policy at the time pipelines submitted their page 700 filings for 2019 undermined the reliability of the reported ROEs for 2019.
202

Thus, Liquids Shippers urged the Commission to replace pipelines' reported page 700 ROEs for 2014 and 2019 with standardized ROEs for purposes of calculating the index level. For 2014, Liquids Shippers proposed a standardized ROE of 10.29%, which 54 pipelines reported in their 2014 page 700 filings.
203

For 2019, Liquids Shippers proposed to use the 10.02% ROE that Trial Staff proposed in an ongoing Colonial rate proceeding based upon data for the six-month period ending in November 2019.
204

201
Liquids Shippers Initial Comments at 21-23.

202

Id.
at 25-28. In support of this argument, Liquids Shippers contend that two pipelines submitted updated Form No. 6 filings in July 2020 indicating that the page 700 ROEs they reported in April 2020 did not comply with the Commission's then-applicable policy relying solely upon the DCF model. Liquids Shippers Request for Rehearing at 76-77 (citing Liquids Shippers Initial Comments at 25-28) (referring to updated Form No. 6 filings of Plains Pipeline, LP, and Rocky Mountain Pipeline System LLC).

203
Ms. Crowe stated that 45 pipelines reported a 10.29% ROE on their page 700s for 2014. Crowe Initial Aff. at 11-12. However, based upon a review of Form No. 6 filings submitted in 2016, the Commission found in the December 2020 Order that 54 pipelines reported this ROE for 2014 in the column on page 700 for previous year data. December 2020 Order, 173 FERC ¶ 61,245 at P 43 n.97.

204
Liquids Shippers Initial Comments at 30-31; Crowe Initial Aff. at 11 (citing Trial Staff, Exhibit S-00057 (Direct and Answering Cost-Based Rate Testimony of Commission Trial Staff Witness Robert J. Keyton), Docket Nos. OR18-7-002 et al. (filed Jan. 14, 2020)).

1. December 2020 Order

71. The December 2020 Order declined to adopt standardized ROEs for 2014 and 2019 and concluded that Liquids Shippers have not demonstrated that the reported page 700 ROEs are unreliable or inconsistent with Commission policy.
205

First, the Commission rejected Liquids Shippers' argument that page 700 ROEs are unreliable simply because they are self-reported, reasoning that the instructions on page 700 required pipelines to determine ROE consistent with the Commission's then-applicable policy of relying solely upon the DCF model.
206

Second, the Commission found that variation among page 700 ROEs does not indicate that this data is unreliable and that such variation may result from differences in proxy group composition and relative risk.
207

Third, the Commission rejected Liquids Shippers' contention that pipelines were uncertain as to the Commission's oil pipeline ROE policy when they submitted their 2019 Form No. 6 filings. The Commission found that pipelines had adequate notice of the prevailing policy through the page 700 instruction requiring pipelines to determine ROE consistent with the then-current Opinion No. 154-B methodology.
208

Fourth, the Commission found that Liquids Shippers have not supported their proposed standardized ROEs.
209

Finally, the Commission concluded that determining standardized ROEs would complicate the five-year review process and undermine indexing's purpose as a simplified and streamlined ratemaking regime.
210

205
December 2020 Order, 173 FERC ¶ 61,245 at P 45.

206

Id.
P 46.

207

Id.
P 47.

208

Id.
P 48.

209

Id.
P 49.

210

Id.
P 50.

2. Rehearing Request

72. Liquids Shippers contend that the December 2020 Order erred by failing to replace the reported 2014 and 2019 page 700 ROEs with Liquids Shippers' proposed standardized ROEs. They repeat their argument that the reported page 700 ROEs cannot be consistent with the Commission's cost-of-service methodology because they vary substantially.
211

Liquids Shippers

emphasize that these ROEs were selected by the pipelines themselves. Furthermore, Liquids Shippers contend that the page 700 ROEs fail to accurately capture changing market conditions between 2014 and 2019 because some pipelines reported a 2019

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2022-01544. Public record. Not legal advice.