United States Court of Appeals
         FOR THE DISTRICT OF COLUMBIA CIRCUIT



Argued March 13, 2020                  Decided July 24, 2020

                        No. 15-1323

         EL PASO NATURAL GAS COMPANY, L.L.C.,
                     PETITIONER

                              v.

       FEDERAL ENERGY REGULATORY COMMISSION ,
                    RESPONDENT

       EL PASO MUNICIPAL CUSTOMER GROUP, ET AL.,
                     INTERVENORS


            Consolidated with 16-1122, 18-1183


          On Petitions for Review of Orders of the
          Federal Energy Regulatory Commission


    Howard L. Nelson argued the cause for petitioner El Paso
Natural Gas Company, L.L.C. With him on the briefs were
Kenneth M. Minesinger, Francesca Ciliberti-Ayres, and J.
Curtis Moffatt.

    Richard P. Bress argued the cause for petitioners Southern
California Gas Company, et al. With him on the briefs were J.
                             2
Patrick Nevins, Charles S. Dameron, and Jonathan J.
Newlander.

    Beth G. Pacella and Lona T. Perry, Deputy Solicitors,
Federal Energy Regulatory Commission, argued the causes for
respondent. With them on the briefs were James P. Danly,
General Counsel, Robert H. Solomon, Solicitor, and Carol J.
Banta, Senior Attorney. Anand Viswanathan, Attorney,
entered an appearance.

     Howard L. Nelson, Kenneth M. Minesinger, and
Francesca Ciliberti-Ayres were on the joint brief for
intervenors El Paso Natural Gas Company, L.L.C., et al. in
support of respondent.

     Betsy Carr, Katherine B. Edwards, John Paul Floom,
Douglas M. Canter, Suzanne K. McBride, Keith A. Layton,
John P. Gregg, and Barbara S. Jost were on the brief for
intervenors BP Energy Company, et al. in support of
respondent.

    John P. Gregg, Barbara S. Jost, Christopher J. Barr, and
Keith A. Layton were on the brief for intervenors El Paso
Municipal Customer Group, et al. in support of respondent.

   Before: SRINIVASAN, Chief Judge, and GARLAND and
WILKINS, Circuit Judges.

    Opinion for the Court filed PER CURIAM.

     PER CURIAM: El Paso Natural Gas Company operates
pipelines that transport natural gas to customers across the
southwestern United States. The consolidated petitions for
review challenge a number of orders of the Federal Energy
                                3
Regulatory Commission (“FERC”) on two intertwined El Paso
rate cases.

     The proceedings at issue began in June 2008 when El Paso
filed to increase its rates under Section 4 of the Natural Gas Act
(“the 2008 Rate Case”). In March 2010, El Paso and its
customers settled that case, but reserved certain issues for
hearing, including the appropriateness of El Paso’s capital
structure. The settlement provided that resolution of that issue
would not affect the rates for the term of the settlement but
would govern future rate cases. While the 2008 Rate Case’s
reserved issues were pending before FERC, El Paso filed
another Section 4 rate case in September 2010 (“the 2011 Rate
Case”).

     FERC proceeded in both cases in parallel. In May 2012,
FERC resolved the capital structure issue in El Paso Natural
Gas Co., 139 F.E.R.C. ¶ 61,095 (2012) (“Op. 517”). In
October 2013, FERC incorporated the 2008 Rate Case’s
pending resolution of the capital structure issue and largely
resolved the 2011 Rate Case’s issues, remanding one
compliance question to an administrative law judge. El Paso
Nat. Gas Co., 145 F.E.R.C. ¶ 61,040 (2013) (“Op. 528”). In
July 2015, FERC granted partial rehearing on El Paso’s capital
structure in El Paso Natural Gas Co., 152 F.E.R.C. ¶ 61,039
(2015) (“Op. 517-A”). El Paso then petitioned for review of
Opinions 517 and 517-A. While that petition was pending, in
February 2016, FERC denied rehearing in relevant part and
addressed the compliance issue in the 2011 Rate Case in El
Paso Natural Gas Co., 154 F.E.R.C. ¶ 61,120 (2016) (“Op.
528-A”). Three California-based utilities that either ship gas
or purchase gas shipped on El Paso’s pipelines (“California
Petitioners”) then petitioned for review of Opinions 528 and
528-A. This Court held both petitions for review in abeyance
pending FERC’s disposition of requests for rehearing of
                               4
Opinion 528-A. FERC resolved those requests in El Paso
Natural Gas Co., 163 F.E.R.C. ¶ 61,079 (2018) (“Op. 528-B”).
El Paso then petitioned for review of Opinions 528, 528-A and
528-B. This Court consolidated the petitions.

     The result is that we now consider three petitions for
review of five Commission orders. In the first petition, El Paso
seeks review of Opinions 517 and 517-A. Specifically, El Paso
challenges FERC’s removal of two assets -- certain
undistributed subsidiary earnings and a loan to El Paso’s parent
-- from the equity component of El Paso’s capital structure in
the 2008 Rate Case.

     In the second petition, El Paso seeks review of Opinions
528, 528-A, and 528-B. Specifically, El Paso challenges
FERC’s determination that El Paso’s rate proposal would
violate a provision of a 1996 settlement agreement, FERC’s
exclusion of two compressor stations from El Paso’s rate base,
and FERC’s disposition of the capital structure issue as
incorporated from the 2008 Rate Case.

     In the third petition, California Petitioners seek review of
Opinions 528 and 528-A. Specifically, California Petitioners
challenge FERC’s approval of a method of allocating costs
across delivery zones based on contract-paths, and FERC’s
rejection of El Paso’s proposal to merge the three western-most
delivery zones.

     We deny the petitions for review. We hold that FERC’s
removal of both the undistributed subsidiary earnings and the
loan to El Paso’s parent from the equity component of El
Paso’s capital structure was reasoned and supported by
substantial evidence. We also hold that FERC’s conclusion
that El Paso had not demonstrated that its proposed rates would
comply with the 1996 settlement was reasonable. Further, we
                               5
hold that FERC reasonably excluded the two compressor
stations from El Paso’s rate base. Finally, we hold that FERC’s
approval of a zone-of-delivery rate design measured by
contract-paths and its rejection of equilibration for lack of
quantitative support were neither arbitrary nor contrary to law.

                               I.

     The first issue we consider in these petitions is whether
FERC arbitrarily removed two assets -- specifically, $145
million in undistributed subsidiary earnings and $615 million
in loans from El Paso to its parent -- from the equity component
of El Paso’s capital structure. Those adjustments substantially
changed El Paso’s debt-equity ratio and thus its rate of return.

     El Paso challenges those adjustments. Primarily, El Paso
contends that FERC arbitrarily departed from its supposed
practice of declining to remove an asset from equity absent
tracing the asset to an equity issuance. Because we cannot
conclude that FERC departed from its precedent, we deny the
petitions for review on this issue.

                              A.

     The Natural Gas Act (“NGA”) requires a pipeline’s rates
for the transportation or sale of natural gas to be “just and
reasonable.” 15 U.S.C. § 717c(a). “Under cost-of-service
ratemaking principles,” just and reasonable rates must “yield[]
sufficient revenue to cover all proper costs.” City of
Charlottesville v. FERC, 

, 1207 (D.C. Cir. 1985).
Those costs include not only operating expenses but also the
capital costs of the business, such as “service on the debt and
dividends on the stock.” Fed. Power Comm’n v. Hope Nat.
Gas Co., 

, 603 (1944).
                                6
      The classic cost-of-service ratemaking formula accounts
for capital costs via a return on capital invested in rate base
(i.e., in assets used to provide transportation or sale of natural
gas subject to FERC’s jurisdiction).               See City of

; Pub. Serv. Co. of N.M. v.
FERC, 

, 683 (D.C. Cir. 1981). Because different
sources of capital have different costs, a pipeline’s total cost of
capital depends both on the cost of each source and the portion
of each source in the pipeline’s total capitalization (i.e., the
pipeline’s capital structure). See Pub. Serv. 

. For example, in the 2008 Rate Case, El Paso’s proposed
total capitalization ($2.9 billion) was roughly 60% equity ($1.8
billion) and 40% debt ($1.2 billion). Op. 517, 139 F.E.R.C. at
61,580. El Paso requested a 13% rate of return on its equity
and a roughly 8% return on its debt, so its proposed rate of
return would be about 11% (60% equity * 13% rate of return +
40% debt * 8% rate of return).

a pipeline’s rate of return should be based
only on the capitalization, and the corresponding capital
structure, that a pipeline devotes to rate base. See El Paso Nat.
Gas Co. v. Fed. Power Comm’n, 

, 1251 (5th Cir.
1971). But that is often infeasible. A pipeline’s rate base is
often less than the pipeline’s total capitalization because the
pipeline invests in more than just rate-base assets. For instance,
in the 2008 Rate Case, El Paso’s proposed total capitalization
was $2.9 billion, but El Paso invested only $1.9 billion in the
rate base and the remaining $1 billion in non-rate-base assets.
Op. 517, 139 F.E.R.C. at 61,580. In that situation, a pipeline’s
balance sheet, which reflects its total capitalization, does not
reflect its rate-base capitalization. And in general pipelines do
not otherwise track the source of capital used for specific
investments beyond total capitalization. Rather, fungible funds
from both debt and equity comingle in corporate accounts, and
the pipeline draws upon that undifferentiated pool of total
                               7
capitalization to make both rate-base and non-rate-base
investments.

     In light of that reality, FERC generally assumes that a
pipeline invests in rate base in the same debt-equity ratio as it
invests in everything else. See Kern River Gas Transmission
Co., 123 F.E.R.C. ¶ 61,056, at 61,459 (2008); Ark.-La. Gas
Co., 19 F.E.R.C. ¶ 63,008, at 65,057 (1982). But FERC adjusts
that assumption when circumstances allow FERC to more
accurately estimate the debt-equity ratio of capital invested in
rate base. For instance, for project-financed pipelines in which
loan agreements require that all debt be invested in rate base,
FERC assumes that the pipeline invests all of its debt in rate
base and makes up any shortfall with equity. See, e.g., Kern
River, 123 F.E.R.C. at 61,459; Wyo. Interstate Co. Ltd.,
69 F.E.R.C. ¶ 61,259, at 61,987 (1994).

     These cases involve a similarly motivated capital structure
adjustment. Specifically, if FERC can attribute a specific
non-rate-base asset solely to equity, FERC removes that asset
from the equity component of the pipeline’s total
capitalization. That adjusted capitalization and corresponding
debt-equity ratio then more accurately estimates the
debt-equity ratio of capital invested in rate base. For example,
in a prior case, FERC removed from the equity component of
El Paso’s total capitalization two non-rate-base subsidiaries
that El Paso had acquired in exchange for El Paso common
stock. El Paso Nat. Gas Co., 44 F.P.C. 73, 77 (1970), aff’d, El
Paso Nat. 

. Analogously, if FERC can
attribute a non-rate-base asset to debt financing, FERC
removes that asset from the debt component of total
capitalization. See Ark.-La. Gas, 19 F.E.R.C. at 65,057.

   To remove a non-rate-base asset solely from equity in that
manner, FERC must have a basis to attribute the asset solely to
                               8
equity. Absent such a basis, FERC adheres to its general
assumption that a pipeline invests in rate base assets in the
same debt-equity ratio as it invests in other assets. See, e.g.,
SFPP, L.P., 134 F.E.R.C. ¶ 61,121, at 61,583-84 (2011);
Mountain Fuel Res., Inc., 27 F.E.R.C. ¶ 61,171, at 61,315
(1984); S. Cal. Edison Co., 3 F.E.R.C. ¶ 63,033, at 65,203
(1978).

                              B.

     El Paso contends that FERC’s removal of certain
undistributed subsidiary earnings and a pipeline-parent loan
from the equity component of El Paso’s capitalization departed
from this line of cases. FERC did not depart from those cases.
In its decisions here, FERC reaffirmed that, “[i]n order to
remove an asset not devoted to jurisdictional service from the
equity portion of a pipeline’s capitalization, there must be a
basis to attribute that asset to equity.” Op. 517, 139 F.E.R.C.
at 61,588. FERC then went on to find sufficient bases to
attribute both the undistributed subsidiary earnings and the
pipeline-parent loan to equity before removing them.

     As for the undistributed subsidiary earnings, FERC
reasoned that the funds “represent unrealized equity in the
subsidiary, generated from pipeline operations,” which “will be
recognized . . . as retained earnings, or equity” when El Paso
appropriates them.

The undistributed earnings
“reside in a proprietary capital account, meaning they are
owing to the residual shareholder.” Op. 528-B, 163 F.E.R.C.
at 61,386. Accordingly, FERC deemed it “appropriate to
reflect the exclusion from the equity component of El Paso’s
capitalization, rather than apply the exclusion proportionately
to debt and equity.” Op. 517, 139 F.E.R.C. at 61,589.
                                9
      As for the pipeline-parent loan, FERC reasoned that El
Paso had loaned “funds generated from general revenue and
operations,” which “no debt issuance ha[d] any claim on” and
which “represent[ed] additional equity available to the pipeline
to dispose of at its discretion.”

Further, FERC
found it “more important than simple accounting,” Op. 517-A,
152 F.E.R.C. at 61,194, that the loan “represent[s] an asset that
offsets the liability that [El Paso] owes its shareholder parent
by way of common stock,” Op. 517, 139 F.E.R.C. at 61,590.
While typically a parent’s stock represents the extent of its
investment in the pipeline, El Paso’s continuous maintenance
of a large, low-interest, long-term loan to its parent changed the
“underlying financial realities.” Op. 517-A, 152 F.E.R.C. at
61,194; see also
 id. at 61,192. 
Such a loan rendered “El Paso’s
stated equity figure not representative of the amount that its
parent corporation has at stake in El Paso” or of “the risks that
the parent has undertaken through its investment.”
 Id. at
61,194.
     El Paso contends that Commission precedent precludes
that kind of attribution, and that only tracing the source of funds
for an asset to a specific equity issuance could suffice. We
disagree. FERC reasonably interpreted its precedent as
requiring attribution of an asset to equity but not necessarily
tracing the asset to a specific equity issuance. To be sure,
FERC refuses to remove investments in subsidiaries from
equity absent a basis to assume that the funds for such
investments came from equity. See, e.g., Ark.-La. Gas, 19
F.E.R.C. at 65,057. But FERC does not require that the funds
come from a stock issuance in order to attribute them to equity.
For example, in Southern Natural Gas Co., FERC attributed a
loan to a subsidiary to equity because the pipeline had recently
received a roughly equivalent amount in dividends and sales
proceeds from another subsidiary. 44 F.P.C. 567, 572-73
(1970). FERC traced the loan to those funds, which the
                               10
pipeline recorded as gains in equity accounts, not to a stock
issuance.
 Id.

     FERC proceeded 
similarly below, attributing both assets
at issue to equity derived from operations. FERC noted that
the undistributed subsidiary earnings “reside in a proprietary
capital account, meaning they are owing to the residual
shareholder.” Op. 528-B, 163 F.E.R.C. at 61,386. When El
Paso’s parent appropriates them, they “will be recognized . . .
as retained earnings, or equity.” Op. 517, 139 F.E.R.C. at
61,589. Likewise, FERC traced the pipeline-parent loan to
“general revenue and operations,” which “no debt issuance
ha[d] any claim on” and which “represent[ed] additional equity
available to the pipeline to dispose of at its discretion.”
 

Moreover, the loan offset El Paso’s parent’s outlay of
common stock. Op. 517-A, 152 F.E.R.C. at 61,195.

     FERC’s disposition is also consistent with numerous
FERC precedents that have removed similar assets from the
equity component of capital structure. See, e.g., Holyoke Water
Power Co. & Holyoke Power & Elec. Co., 28 F.E.R.C.
¶ 61,361, at 61,651 (1984) (undistributed subsidiary earnings);
Ark.-La. Gas, 19 F.E.R.C. at 65,057 (same); United Gas Pipe
Line Co., 13 F.E.R.C. ¶ 61,044, at 61,096 (1980) (same); S.
Cal. Edison Co., 3 F.E.R.C. at 65,203 (same); Distrigas of
Mass. Corp., 18 F.E.R.C. ¶ 63,036, at 65,121 (1982) (a
pipeline-parent loan). Especially in light of the “deference . . .
due to the Commission’s interpretation of its own precedent,”
Mo. Pub. Serv. Comm’n v. FERC, 
783 F.3d 310
, 316 (D.C. Cir.
2015), we cannot conclude that the Commission departed from
its context-sensitive approach to capital structure adjustments.

     In addition, FERC’s disposition was reasonable. At
bottom, FERC seeks to estimate the debt-equity ratio invested
in rate base to the extent feasible. See El Paso Nat. Gas,
                               
11
449 F.2d at 1251
. Retained earnings represent equity just like
the proceeds of an equity issuance. It would make little sense
for FERC to remove from equity assets traced to proceeds of
an equity issuance (one form of equity), but not assets
attributed to retained earnings (another form of equity).
Moreover, FERC’s consideration of the underlying financial
realities, and specifically the pipeline-parent loan’s impact on
them, was reasoned and entitled to the “great deference” we
afford FERC’s enforcement of the just and reasonable
standard. Morgan Stanley Capital Grp. Inc. v. Pub. Util. Dist.
No. 1, 
554 U.S. 527
, 532 (2008).

     FERC’s attribution of internally generated funds to equity
in this case did not exceed the bounds of the just and reasonable
standard. As discussed, while the rate of return should be based
only on the capitalization that a pipeline devotes to public
service, that may be infeasible when “non-public segments of
such capital” cannot be “distinctly identified and surely
isolated.” El Paso Nat. 
Gas, 449 F.2d at 1251
. In such
situations, “a potential shareholder or lender-investor” may be
unable to “determine the value of the regulated versus the non-
regulated operations and calculate the sureness of his regulated
return on the one and the commercial risk he assumes on the
other.”
 

50. 
Here, however, FERC found that “El
Paso’s debtors are able to independently weigh the risks” of El
Paso’s rate base assets and the outstanding pipeline-parent loan
balance. Op. 517, 139 F.E.R.C. at 61,590-91.

     El Paso provides no reason to disturb that finding. That
the funds loaned were internally generated does not mean that
investors cannot distinctly identify the loan itself as non-public
and evaluate it independently from El Paso’s public
capitalization. The same is true for undistributed subsidiary
earnings. Accordingly, we cannot conclude that FERC’s
                               12
removal of either asset exceeded the bounds of the just and
reasonable standard.

     For these reasons, we deny the petitions for review on this
issue.

                               II.

     Next, El Paso challenges FERC’s determination that it
charged for costs prohibited by a 1996 settlement agreement.
That year, California customers returned their rights to about
35% of El Paso’s total capacity in response to state efforts to
deregulate the electricity industry. Freeport-McMoRan Corp.
v. FERC, 
669 F.3d 302
, 306 (D.C. Cir. 2012). For pipeline
customers, unsubscribed capacity poses a problem. Pipeline
rates are based on costs, so fewer customers means fewer
people to split those costs. El Paso’s remaining customers
therefore faced the potential for major rate hikes if El Paso
could not resell this unsubscribed capacity, and El Paso risked
a further drop in demand brought on by higher prices.

     In order to spread the risk, El Paso and its customers (now
called “rate-protected shippers”) struck a deal: current
customers would shoulder some costs in the short term (until
2004) in exchange for, as relevant here, a long-term promise
that they would not thereafter pay for costs related to El Paso’s
1995 capacity if that capacity became unsubscribed or was
discounted. Specifically, Article 11.2(b) of the settlement
provides:

         El Paso agrees that the firm rates applicable to service
         to any [rate-protected shipper] will exclude any cost,
         charge, surcharge, component, or add-on in any way
         related to the capacity of its system on December 31,
         1995 . . . that becomes unsubscribed or is subscribed
                                13
         at less than the maximum applicable tariff rate as
         [adjusted for inflation].

Op. 528, 145 F.E.R.C. at 61,183 n.5.

     In essence, Article 11.2(b) modifies El Paso’s ability to
charge “discount adjustments” to rate-protected shippers.
Normally, when a pipeline gives some customers a discount
due to competitive conditions, it can require other customers to
help bear the cost. See Ala. Mun. Distributors Grp. v. FERC,
312 F.3d 470
, 472-73 (D.C. Cir. 2002). Similarly, when a
pipeline has unsubscribed capacity, each customer’s share of
fixed costs will increase. Article 11.2(b) prevents El Paso from
charging rate-protected shippers for those costs insofar as they
are “in any way related” to unsubscribed or discounted 1995
capacity.

     El Paso has spent the intervening years trying to get out of
this bargain.1 Today though, El Paso accepts that Article
11.2(b) applies. It simply argues that it has complied.

     FERC has developed a two-step process, unchallenged
here, for testing El Paso’s compliance with Article 11.2(b):
First, it calculates “whether El Paso’s firm contracts at or above
the rate cap exceed 4,000 MMcf/d.” And second, it determines
“whether El Paso proposes to shift the costs of unsubscribed or
discounted capacity to the rates of Article 11.2(b) shippers.”

     1
       See, e.g., Freeport-McMoRan 
Corp., 669 F.3d at 308
; Op.
528-A, 154 F.E.R.C. at 61,705; Op. 517-A, 152 F.E.R.C. at 61,223;
Op. 528, 145 F.E.R.C. at 61,253; Op. 517, 139 F.E.R.C. at 61,606-
07; El Paso Nat. Gas Co., 133 F.E.R.C. ¶ 61,129, at 61,622 (2010);
El Paso Nat. Gas Co., 132 F.E.R.C. ¶ 61,155, at 61,785 (2010); El
Paso Nat. Gas Co., 124 F.E.R.C. ¶ 61,227, at 62,168 (2008); El Paso
Nat. Gas Co., 124 F.E.R.C. ¶ 61,124, at 61,662 (2008); El Paso Nat.
Gas Co., 114 F.E.R.C. ¶ 61,290, at 61,014 (2006).
                                14
Op. 528, 145 F.E.R.C. at 61,267 (quoting Op. 517, 139
F.E.R.C. at 61,624).

     The first step is used to determine whether El Paso has any
unsubscribed or discounted 1995 capacity. Because El Paso
operates as an integrated whole, it is difficult to assign a unit of
capacity to a particular year. Instead, FERC treats the first
4,000 million cubic feet per day (MMcf/d) -- El Paso’s
approximate 1995 capacity -- as the 1995 capacity. See
Freeport-McMoRan, 669 F.3d at 312-13 
(upholding this
presumption). If El Paso subscribes 4,000 MMcf/d of capacity
at its maximum rate, FERC will presume that there is no
unsubscribed or discounted 1995 capacity and thus, there is no
“cost, charge, surcharge, component, or add-on in any way
related to” that capacity that El Paso can pass on. But if El Paso
has not subscribed 4,000 MMcf/d at the maximum rate, FERC
will proceed to the second step and examine El Paso’s rates
more closely to see whether El Paso is charging those costs to
rate-protected shippers.

     El Paso accepts that it has not met the 4,000 MMcf/d
threshold, but disagrees with FERC regarding how to
determine whether the costs of discounted or unsubscribed
capacity are being charged to rate-protected shippers. Because
El Paso does not dispute that it bears the burden of proving
Article 11.2(b) compliance, see Op. 528-A, 154 F.E.R.C. at
61,742, we ask only whether FERC reasonably rejected El
Paso’s approach, not whether FERC’s approach is right. We
review FERC’s decision under the familiar arbitrary-and-
capricious standard. 5 U.S.C. § 706(2)(A); see FERC v. Elec.
Power Supply Ass’n, 
136 S. Ct. 760
, 782 (2016).

     El Paso contends that FERC must look at the facilities
(e.g., the physical pipeline) that existed in 1995 and see what
costs those add to El Paso’s current rates. Since any facilities
                                15
that existed in 1995 have significantly depreciated in value, the
revenues attributable to those facilities now exceed their costs,
which, in El Paso’s view, means that there are no costs “in any
way related to” discounted or unsubscribed 1995 capacity.
Hence, there is no impermissible cost-shifting.

     FERC, by contrast, points out that Article 11.2(b) does not
mention “facilities,” but only refers to “the capacity of [El
Paso’s] system on December 31, 1995.” Op. 528-B, 163
F.E.R.C. at 61,377 (emphasis omitted). In FERC’s view, El
Paso takes a wrong turn at the outset by looking at facility costs.
Instead, to simplify slightly, FERC compares the rates El Paso
wants to charge rate-protected shippers with the rates it would
have charged them had it managed to subscribe 4,000 MMcf/d
at the maximum rate. Op. 528-A, 154 F.E.R.C. at 61,742.
Since the proposed rates are higher, El Paso has not complied
with Article 11.2(b).

     We agree with FERC. Its reading is consistent with the
text and purpose of the 1996 settlement. The settlement refers
to capacity, not facilities, and FERC reasonably concluded that
these are distinct concepts. See Op. 528-A, 154 F.E.R.C. at
61,740-41. While El Paso maintains that the “cost of capacity
can be measured only by reference to the cost of facilities that
create that capacity,” El Paso Opening Br. 23 (emphasis
added), FERC’s alternative methodology shows otherwise.
And given the capacious language of the settlement -- covering
costs “in any way related to” discounted capacity -- FERC’s
approach also avoids unduly narrowing which costs are
prohibited. Cf. Morales v. Trans World Airlines, Inc., 
504 U.S.
374
, 383 (1992) (noting that the “ordinary meaning” of
“relating to” is a “broad one”).

    FERC’s reading also effectuates the settlement’s purpose
by providing shippers with long-term protection. That promise
                               16
would mean little if the prohibited costs quickly depreciated
away. Especially in light of the “high degree of deference” we
give to FERC’s interpretation of settlement agreements,
Freeport-McMoRan, 669 F.3d at 308
, we conclude that FERC
reasonably rejected El Paso’s facilities-based approach.

     From there, El Paso’s remaining arguments fall away. El
Paso barely contests FERC’s bottom-line conclusion that it
shifted costs, simply reiterating that FERC should have looked
at depreciated facility costs -- the argument we just rejected.
Instead, El Paso focuses on FERC’s rejection of two studies
that purport to show El Paso’s compliance with Article 11.2(b).
But both of those studies “erroneously identify the cost of 1995
capacity as the cost of the facilities comprising El Paso’s 1995
system.” Op. 528-A, 154 F.E.R.C. at 61,741; see Op. 528-B,
163 F.E.R.C. at 61,374. By putting all of its eggs in one basket,
El Paso made FERC’s task -- and ours -- straightforward.
Having reasonably rejected El Paso’s premise, FERC
reasonably rejected studies that insist on that premise.

     El Paso makes one other argument, but it too is readily
dispatched. El Paso claims that FERC improperly treated El
Paso’s failure to subscribe 4,000 MMcf/d at the maximum rate
as dispositive, ignoring whether El Paso charged rate-protected
shippers for discounted capacity. But that is not what FERC
did. Instead, FERC first noted that, because El Paso had not
met the 4,000 MMcf/d threshold, discounted or unsubscribed
1995 capacity existed. At the second step, it then analyzed El
Paso’s proposed rates and determined that they passed the cost
of those discounts on to rate-protected shippers. Op. 528-A,
154 F.E.R.C. at 61,742; Op. 528-B, 163 F.E.R.C. at 61,378,
61,380. That two-step analysis is fully consistent with FERC’s
long-standing approach to Article 11.2(b).
                                17
                                III.

    Finally, El Paso challenges FERC’s decision to exclude
two compressor stations from its cost calculation. 2 As part of
its mandate to ensure “just and reasonable” rates, FERC
generally looks to a pipeline’s cost of service. N. Nat. Gas Co.
v. FERC, 
700 F.3d 11
, 13 (D.C. Cir. 2012) (quoting 15 U.S.C.
§ 717c(a)). By regulation, FERC considers a pipeline’s costs
during a test period, including a twelve-month base period and
an up-to-nine-month adjustment period.               18 C.F.R.
§ 154.303(a). But FERC may permit “reasonable deviation”
from the test period,
 id. § 154.303(d), 
which it does when test-
period estimates are “substantially in error or would yield
unreasonable results,” Nat’l Fuel Gas Supply Corp., 51
F.E.R.C. ¶ 61,122, at 61,334 (1990).

      In September 2010, two days before it initiated its 2011
rate case and during the adjustment period for that same case,
El Paso applied to abandon its Deming and Tucson compressor
stations. See Abandonment Appl. 1 (Sept. 28, 2010) (2 J.A.
1161);3 El Paso Rate Filing 5 (Sept. 30, 2010) (2 J.A. 575)
(initiating the 2011 rate case with an adjustment period running
from July 1, 2010, to March 31, 2011). It noted that those
compressors “have become functionally obsolete and are no
longer required to provide natural gas transportation service.”

     2
      Natural gas is transported at high pressure. Pipelines use
compressor stations at strategic locations to maintain that pressure
and pump gas along the pipeline. See FERC, AN INTERSTATE
NATURAL GAS FACILITY ON MY LAND? WHAT DO I NEED TO
KNOW? 20 (Aug. 2015).
     3
      The parties filed two separately paginated appendices in these
consolidated cases. For ease of reference, we cite the appendix in
No. 15-1323 as Volume 1 and the appendix in Nos. 16-1122 and 18-
1183 (including the supplement) as Volume 2.
                              18
Abandonment Appl. 1 (2 J.A. 1161). Both had been used only
in reserve since at least 2004 and “recently” were only
“intermittently run and tested to maintain compliance” with
Department of Transportation and Environmental Protection
Agency requirements.
 Id. at 6-7 
(2 J.A. 1166-67). El Paso also
noted the benefit to its customers: “Any appropriate rate
impact to customers resulting from a timely approval of this
abandonment application should be reflected in [El Paso’s
2011] rate case filing.”
 Id. at 12 
(2 J.A. 1172). FERC approved
the abandonment in September 2011, after the end of the test
period. El Paso Nat. Gas Co., 136 F.E.R.C. ¶ 61,180 (2011).

     Despite those statements, El Paso sought to include the
compressor costs in its rates. FERC’s Administrative Law
Judge (“ALJ”) found this to be “entirely unjust and
unreasonable,” noting that although FERC did not approve the
abandonment until after the end of the test period, El Paso’s
application represented that the stations were no longer useful
during the test period. El Paso Nat. Gas Co., 139 F.E.R.C.
¶ 63,020, at 66,214 (2012) (“ALJ Op.”). FERC affirmed and
reaffirmed the ALJ’s conclusion. Op. 528, 145 F.E.R.C. at
61,214-15; Op. 528-A, 154 F.E.R.C. at 61,677. In doing so,
FERC noted that El Paso “does not dispute that these
compressor stations have not served any real function related
to the transportation of natural gas for a number of years.”
Op. 528, 145 F.E.R.C. at 61,214.

    El Paso presses its case to us, arguing that excluding the
compressor costs is inconsistent with two strands of FERC
precedent. In Tennessee Gas Pipeline Co., 73 F.E.R.C.
¶ 61,368 (1995), and Panhandle Eastern Pipe Line Co., 71
F.E.R.C. ¶ 61,228 (1995), FERC permitted pipelines to charge
for facilities in service during the test period but later
abandoned. And in Wyoming Interstate Co., 76 F.E.R.C.
¶ 61,252 (1996), and Eastern Shore Natural Gas Co., 76
                               19
F.E.R.C. ¶ 61,358 (1996), FERC permitted pipelines to charge
for new backup compressor stations. All of these cases simply
reflect FERC’s longstanding practice of permitting charges for
facilities that are “used and useful.” See La. Pub. Serv.
Comm’n v. FERC, 
174 F.3d 218
, 228-29 (D.C. Cir. 1999).

     Although El Paso points to its own expert’s testimony that
the stations were used and useful, see Prepared Rebuttal
Testimony of Mark A. Westhoff 26-27 (2 J.A. 756-57),
FERC’s factual finding to the contrary is “conclusive” if
supported by “substantial evidence,” 15 U.S.C. § 717r(b). Our
review on this score is “highly deferential.” PJM Power
Providers Grp. v. FERC, 
880 F.3d 559
, 562 (D.C. Cir. 2018)
(internal quotation marks omitted). Here, FERC reasonably
found, based on El Paso’s own statements, that the Deming and
Tucson compressor stations served no purpose for backup or
otherwise during the test period. In light of that finding, FERC
properly held El Paso to its promise not to charge customers
for those costs.

     Taking a different tack, El Paso argues that, since FERC
excluded the compressor costs, it should also have considered
post-test-period changes that favored El Paso. But making one
post-test-period adjustment does not obligate FERC to make all
such adjustments. FERC considers subsequent developments
when test-period estimates are “substantially in error or would
yield unreasonable results.” Nat’l Fuel, 51 F.E.R.C. at 61,334.
As El Paso makes no effort to show how the changes it favors
meet that standard, this argument fails as well.

                              IV.

    For their part, California Petitioners -- Southern California
Gas Company, San Diego Gas & Electric Company, and
Southern California Edison Company -- challenge two FERC
                               20
orders -- Opinion 528 and Opinion 528-A -- on several issues:
(1) FERC’s approval of a “zone-of-delivery” rate
methodology; (2) FERC’s approval of the measurement of
those zones with reference to “contract paths”; and (3) FERC’s
rejection of El Paso’s “equilibration” proposal and its
determination that the proceeding was properly conducted
under Section 4 of the NGA, 15 U.S.C. § 717c, rather than
under Section 5,
 id. § 717d.
     For the reasons detailed herein, we deny the petition on all
appealed issues. There is substantial evidence supporting
FERC’s finding that El Paso’s continued use of a
zone-of-delivery design now calculated by reference to
contract paths is just and reasonable under Section 4; El Paso’s
uncontested dekatherm-mileage study supports a rate design
reflecting moderate, distance-based differences in rates
increasing from east to west, and contract paths are a
reasonable measurement tool in this case. FERC also
reasonably affirmed the ALJ’s determination that El Paso
failed to prove its equilibration proposal was just and
reasonable, finding equilibration would significantly modify
the results of the dekatherm-mileage study without sufficient
empirical support. Finally, approving the zone-of-delivery
design and rejecting equilibration did not result in a rate of
FERC’s own making such that Section 5 of the NGA is
triggered. FERC properly proceeded under Section 4 of the
NGA and as such was not required to consider California
Petitioners’ alternative rate proposal.

     In a rate-setting proceeding, a pipeline may seek to recover
its mileage-based fixed costs, or costs associated with
maintaining sufficient capacity to serve peak needs on the
system. Because of the interconnected, multi-path nature of El
Paso’s system, most gas being delivered has more than one
flow option, and higher demand can require gas to flow through
                               21
more indirect routes. These fluctuations make the actual route
gas will flow to each delivery point impossible to calculate,
complicating the allocation of fixed costs. In El Paso’s last
fully litigated rate case in 1959 (before the pipeline was as
complex as it is today), FERC approved a zone-of-delivery rate
methodology, under which shippers pay the same rate to
deliver gas to any point within each of the five state zones --
Texas, New Mexico, Arizona, Nevada, and California --
increasing in modest increments from east to west. See El Paso
Nat. Gas Co., 22 F.P.C. 260, 280-82 (1959). In El Paso’s two
most recent rate settlements, it has continued to use these
state-wide zones.

     In the 1990s as pipeline demand grew, El Paso had
insufficient capacity to serve all customers, prompting FERC
to order the creation and assignment of “contract paths.” See
Op. 528-A, 154 F.E.R.C. at 61,682. These contract paths were
not created to establish the path that gas would actually flow
from receipt and delivery points, but rather the path on El
Paso’s system that a shipper had rights to under its contract.
Contract paths were then assigned as a mechanism to ensure
that El Paso had capacity to meet the demands of all shippers’
contracts on peak days. See Op. 528, 145 F.E.R.C. at 61,223;
Op. 528-A, 154 F.E.R.C. at 61,686.

     In this rate case, El Paso proposed to continue its 50-plus-
year-old zone-of-delivery rate design. See ALJ Op., 139
F.E.R.C. at 66,217. To support its proposal, El Paso conducted
a dekatherm-mileage study, which calculated the average
distance gas is transported to each of the five rate zones based
on assigned contract paths. The study determined the mileage
associated with every firm shipper’s contract path in the state,
then the mileages were added together and weighted by the
total contract delivery volume for each zone. See
 id. at 66,222;
see also El Paso Ex. No. 224, at 41 (2 J.A. 680) (El Paso
                                22
witness Richard Derryberry stating, “[b]ecause a shipper is
able to rely on its contract paths when they are needed most, at
the peak, I believe such paths provide a more accurate measure
of the facilities need[ed] to serve the shipper, and the associated
distance of haul, than the ‘typical’ flows” on El Paso’s system).
After calculating the average miles of haul for each of the five
state zones, El Paso further proposed to “equilibrate” the
distances, or equalize the rates, for the three western states --
Arizona, Nevada, and California -- into a single zone. See El
Paso Ex. No. 107, at 29-30 (2 J.A. 729-30). After equilibration,
the proposed California rates would have been slightly lower
and the Arizona rates slightly higher than under a pure
five-zone approach. Op. 528, 145 F.E.R.C. at 61,207.4 After
a hearing, the ALJ found El Paso’s zone-of-delivery design as
measured by contract paths just and reasonable but rejected the
additional step of equilibration. See ALJ Op., 139 F.E.R.C. at
66,220-29.

     FERC affirmed these findings in Opinion 528, 145
F.E.R.C. at 61,222-23, 61,225-26, 61,227, and reaffirmed them
in relevant part in Opinion 528-A, 154 F.E.R.C. at 61,679-86,
61,690-93.      Below we address California Petitioners’
objections to the challenged portions of FERC’s orders in turn.

                                A.

    The question of how to allocate costs among a pipeline’s
customers is “a difficult issue of fact, and one on which [FERC]
enjoys broad discretion.” Midcoast Interstate Transmission,
Inc. v. FERC, 
198 F.3d 960
, 971 (D.C. Cir. 2000) (citation
omitted). And since the question involves “both technical
understanding and policy judgment,” this Court’s “important

     4
        The impact on Nevada was hypothetical given the lack of
contracts there. See El Paso Ex. No. 177, at 37 (2 J.A. 625).
                              23
but limited role is to ensure that [FERC] engaged in reasoned
decisionmaking -- that it weighed competing views, selected [a
result] with adequate support in the record, and intelligibly
explained the reasons for making that choice.” Elec. Power
Supply 
Ass’n, 136 S. Ct. at 784
.

    In a Section 4 rate case, the pipeline bears the burden to
prove the justness and reasonableness of any changes it
proposes to its previously approved (and presumptively
reasonable) rate design. See 15 U.S.C. § 717c(e); see also 5
U.S.C. § 556(d) (proponent of order bears the burden of proof);
18 C.F.R. § 154.301(c) (stating a “natural gas company filing
for a change in rates or charges . . . [bears] the burden of
proving that the proposed changes [in rates] are just and
reasonable”). In order for distance-based rates to be just and
reasonable in a Section 4 proceeding, FERC must reasonably
conclude that the cost of transmission on the system varies
materially with the distance from the nominated point of
receipt to the point of delivery.            See 18 C.F.R.
§ 284.10(c)(3)(ii). And FERC’s findings must be supported by
substantial evidence in the record.           Algonquin Gas
Transmission Co. v. FERC, 
948 F.2d 1305
, 1311 (D.C. Cir.
1991).

    California Petitioners first argue that FERC’s approval of
El Paso’s proposed zone-of-delivery method was not just and
reasonable or supported by substantial evidence. Cal. Pet’rs
Opening Br. 25. Because the modern-day El Paso pipeline is
an integrated and reticulated system, it relies extensively on
displacement, or the “substitution of gas at one point for gas
received at another point.” Interstate Nat. Gas Ass’n of Am. v.
FERC, 
285 F.3d 18
, 42 (D.C. Cir. 2002). Pointing to the ALJ’s
concession that, due to displacement and other pipeline
features, it is impossible to “accurately calculate distance [of
product flows],” ALJ Op., 139 F.E.R.C. at 66,222 , California
                               24
Petitioners argue that distance-based rates are “untenable.”
Cal. Pet’rs Opening Br. 26. That is, since “it is impossible to
tell how far any particular shipment of gas will actually travel
to reach a delivery point[,] . . . the impact of distance on the
cost of transportation is unknowable.”
 Id. at 25-26. 
They point
out that, as El Paso’s own expert acknowledged, “[t]he changes
that have occurred on the [El Paso] system in recent years have
almost all been in the direction of deemphasizing the
importance of distance as a cost causation factor[.]” El Paso
Ex. No. 224, at 4 (2 J.A. 643). For example, the predominant
source of gas has shifted to the San Juan Basin, which now
supplies two-thirds of all deliveries and “is more or less
equidistant from all major delivery centers.”
 Id. at 16 
(2 J.A.
655). As such, California Petitioners argue there is “no rational
basis for saying that it costs El Paso more to make deliveries to
California than to make deliveries to Texas.” Cal. Pet’rs
Opening Br. 29.

    In Opinions 528 and 528-A, FERC affirmed and reaffirmed
the ALJ’s approval of El Paso’s proposed zone-of-delivery
design, mostly on the basis of El Paso’s “thorough and
detailed” dekatherms-mileage study. Op. 528, 145 F.E.R.C. at
61,222. FERC found El Paso’s unchallenged dekatherm-
mileage study “demonstrated somewhat shorter average
transportation mileages, and thus less cost responsibility, for
zones moving from east (Texas) to west (California).” FERC
Br. 62-63 (citing ALJ Op., 139 F.E.R.C. at 66,218, 66,222,
66,228; Op. 528, 145 F.E.R.C. at 61,222-23; Op. 528-A, 154
F.E.R.C. at 61,679). Finding “El Paso’s mileage studies were
meticulously prepared, and the assumptions underlying the
studies [were] reasonable and the differences in mileages
between the same receipt point/delivery point combinations
reflect[ed] operational limitations on El Paso’s system,” Op.
528-A, 154 F.E.R.C. at 61,685, FERC held the studies
provided “substantial evidence to support” El Paso’s proposed
                                25
rate design of “moderate, but reasonable, differences in rates
due to distance sensitivity,” Op. 528, 145 F.E.R.C. at 61,222-
23. Although FERC agreed that some factors (i.e., contra-
flows, displacement, and the integrated nature of the pipeline)
complicate a pure distance-based calculation, it found the study
properly accounted for these realities. See Op. 528-A, 154
F.E.R.C. at 61,683.

    Given the level of deference we grant FERC’s ratemaking
decisions and the comprehensive nature of El Paso’s
dekatherm-mileage study, which illustrates that “distance still
has at least a modest effect on system cost responsibility,” ALJ
Op., 139 F.E.R.C. at 66,228, we find there is substantial
evidence to support El Paso’s proposed zone-of-delivery
methodology. A distance-sensitive rate -- reflecting modest
increases moving east to west through the five state zones --
has been in place on the El Paso pipeline for over fifty years.
See
 id. at 66,217 
(discussing the 1959 litigation and 1990
Settlement); El Paso Nat. Gas Co., 54 F.E.R.C. ¶ 61,316, at
61,934 (1991) (approving the continued use of historic zone
rates because “[t]he zones do reflect differences in the distance
of haul”), on reh’g, 56 F.E.R.C. ¶ 61,290, at 62,156 (1991).
And although California Petitioners correctly point to the
increasing complexity of the system over time, FERC
reasonably found the unchallenged study provided a
“reasonable method to account” for these realities. Op. 528-A,
154 F.E.R.C. at 61,683. The study proactively addressed
contraflows, displacement, and other phenomena California
Petitioners point to, and illustrated that these characteristics did
not offset the finding that “distance remains a significant factor
in determining the cost of transporting gas on El Paso’s
system.” Op. 528, 145 F.E.R.C. at 61,223. El Paso’s state-
defined rate zones, previously approved by FERC, remain
presumptively just and reasonable, see Morgan Stanley Capital
Grp., 554 U.S. at 530-31
; see also ALJ Op., 139 F.E.R.C. at
                                26
66,221, and given the evidence in the record, FERC did not act
arbitrarily and capriciously in reaffirming them here.

                                B.

    Next, California Petitioners argue that even if there is a
rational basis for distance-sensitive rates, that contract paths
are not a rational tool for measuring costs associated with such
distance. Cal. Pet’rs Opening Br. 31-39. As FERC conceded,
El Paso’s contract paths “were never developed or approved
for the purpose of cost allocation,” Op. 528-A, 154 F.E.R.C. at
61,686, but were created to help with capacity allocation
problems. See Op. 528, 145 F.E.R.C. at 61,223 (noting that
“contract paths reflect the level of service El Paso is obligated
to provide on any day”); Tr. of Hearing 1472 (Nov. 4, 2011) (2
J.A. 613) (El Paso witness Derryberry conceding that on an
average day there is “no relationship . . . between the contract
path that a shipper holds in its contract and the [actual] flow [of
gas] on that day”). California Petitioners do not contest the
dekatherm-mileage study or its findings, see Cal. Pet’rs
Opening Br. at 13, 39; see also ALJ Op., 139 F.E.R.C. at
66,221 n.156 (noting participants do not dispute the distances
El Paso assigned to contract paths); they simply argue that
contract paths are “fundamentally not a rational measure of
cost incurrence,” so the study’s quality is “irrelevant,” Cal.
Pet’rs Opening Br. 39. They argue that the length of a contract
path bears “no such rational relationship” to the distance gas
actually travels and thus to the costs actually incurred to
provide that transportation, especially given aspects of the
pipeline such as contraflows.
 

In support, California
Petitioners provide a map of the complex pipeline,
 id., and
point 
to statements by an El Paso expert conceding that gas
often flows along shorter, more efficient routes than the
assigned contract paths, Tr. of Hearing 2183, 2186 (Nov. 14,
2011) (2 J.A. 764, 767). They analogize using the contract path
                               27
methodology to mapping a drive from Bethesda to Baltimore
via Alexandria, arguing it would not be rational to use this
circuitous and rarely sensible detour to Virginia as a substantial
factor in calculating the average drive time between two
Maryland cities. Cal. Pet’rs Opening Br. 35.

    Despite California Petitioners’ claims, we find there is
substantial evidence supporting FERC’s finding that El Paso’s
proposed contract-path methodology is just and reasonable in
this case. As California Petitioners point out, “[t]he relevant
question is whether [contract paths] are a rational tool for the
purpose of measuring cost incurrence, a purpose for which they
concededly were not developed.”
 

However, we
disagree that just because contract paths do not reflect the path
gas actually flows on El Paso’s system (which California
Petitioners admit is impossible to determine,
 id. at 2), 
they
cannot be used to calculate the costs incurred by El Paso to
provide that transportation. California Petitioners correctly
point out the dekatherm-mileage study shows that, even under
peak conditions, actual flows replicate contract paths only “60
to 70 percent” of the time and only in the northern parts of the
system, Tr. of Hearing 2617 (Nov. 16, 2011) (2 J.A. 742-43),
arguing this is a “far cry” from resembling actual flows, Cal.
Pet’rs Opening Br. 37. However, FERC reasonably found that
even though contract paths do not consistently reflect actual
flows, they can still be appropriate measuring tools for
ratemaking purposes.

     Even though a shipper’s gas may not actually travel along
its assigned contract path, those paths still “reflect a shipper’s
right to capacity along a specified path, not subject to [a] prior
claim by any other shipper, on all days[.]” FERC Br. 53; see
also Op. 528-A, 154 F.E.R.C. at 61,680, 61,685-86; Op. 528,
145 F.E.R.C. at 61,223. As the ALJ noted, allocating fixed
costs based on capacity rights as established by contract paths
                                28
“acknowledges that installed capacity is the pipeline’s major
fixed cost driver.” ALJ Op., 139 F.E.R.C. at 66,224 . And
“[b]ecause a shipper is able to rely on its contract paths when
they are needed most, at the peak, . . . such paths provide a more
accurate measure of the facilities need[ed] to serve the shipper,
and the associated distance of haul, than the ‘typical’ flows[.]”
El Paso Ex. No. 224, at 41 (2 J.A. 680). Furthermore, FERC
credited testimony that “reliance on typical or average, flows
may well understate the capacity -- and therefore the related
mileage -- needed to serve a particular shipper.”
 Id. El Paso’s
dekatherm-mileage study clearly established (and California
Petitioners do not challenge) the relative length of the average
contract paths in each zone, which supported moderate, but
reasonable, distance-based differences in rates. See Op. 528,
145 F.E.R.C. at 61,222. Although it’s clear why California
Petitioners would desire that shorter routes be used to calculate
their rates, it was reasonable for FERC to find that, in these
circumstances, allocating fixed costs based on capacity rights
reasonably reflects the costs required to provide services to
customers on a complex and integrated pipeline. Although we
might not drive from Bethesda to Baltimore via Alexandria
every day, if the highway authority must maintain a dedicated
lane for us to take that route on a high-traffic day, the associated
expenses seem a reasonable measure of the fixed costs
expended to serve our needs.

                                C.

    California Petitioners’ next argue that even if the contract
path and zone-of-delivery methodologies are reasonable,
FERC’s decision to reject El Paso’s proposed “equilibration”
of the western-zone rates was arbitrary and capricious. Cal.
Pet’rs Opening Br. 23. In its ratemaking proposal, after
calculating rates for each of the five state zones based on
contract paths, El Paso proposed to “equilibrate” the California,
                               29
Arizona, and Nevada zones by averaging their rates into a
single rate, maintaining separate zonal rates only for New
Mexico and Texas. Op. 528-A, 154 F.E.R.C. at 61,680; see
also ALJ Op., 139 F.E.R.C. at 66,187 n.27 (noting
equilibration is not a previously approved practice). El Paso
asserted equilibration was justified because any cost
differences between the California and Arizona zones due to
distance of haul were minimal and offset by other factors that
made transportation to Arizona more expensive. Op. 528, 145
F.E.R.C. at 61,226. The ALJ found El Paso had not shown its
equilibration proposal would result in just and reasonable rates
since the concept was inconsistent with the distance-sensitive
nature of El Paso’s contract path methodology and its
dekatherm-mileage study. ALJ Op., 139 F.E.R.C. at 66,228-
29. FERC affirmed this finding, stating equilibration would
“significant[ly] modif[y]” the results of the detailed
dekatherm-mileage study -- which did reflect differences in
average mileages between the California and Arizona zones --
without offering any comparable empirical support. Op. 528,
145 F.E.R.C. at 61,227-28.

     California Petitioners disagree, pointing to El Paso’s
assertion below that without equilibration, “the resulting zone
of delivery rates would overstate the importance of distance in
allocating costs.” El Paso Br. Opp. Exceptions 47 (Sept. 19,
2012) (2 J.A. 866). They argue the rejection of equilibration
places undue importance on state boundaries, which are
themselves inherently arbitrary, and that “there is no more
‘empirical support’ for treating Arizona as its own zone than
there is for . . . dividing Arizona into two (or more) zones or []
combining Arizona with California . . . especially [] given that
the only California delivery points on the El Paso system are
literally on the Arizona border.” Cal. Pet’rs Opening Br. 40-
41. They argue El Paso presented sufficient evidence showing
that higher system costs in Arizona -- including the use of
                               30
smaller diameter delivery laterals which have a higher per-unit
cost -- offset the slightly higher distance to California,
justifying western-zone equilibration. See Op. 528, 145
F.E.R.C. at 61,227.

    El Paso argued below and California Petitioners argue now
that FERC’s 1962 decision in Tennessee Gas Transmission
Company also supports its equilibration proposal. 27 F.P.C.
202 (1962). There, as here, FERC found it impossible to
identify the portion of the mainline facilities installed or
operated for the benefit of any individual customer.
 Id. at 208.
In Tennessee Gas, the New England zone of that system
featured lateral pipelines not present elsewhere in the system
requiring a special type of service, which FERC found justified
a differentiated cost zone.
 Id. at 212-13. 
California Petitioners
argue this precedent clearly supports El Paso’s equilibration
proposal given the laterals and other distinct costs on the
Arizona portion of the pipeline. See Op. 528, 145 F.E.R.C. at
61,228; see also Cal. Pet’rs Opening Br. 42-43.

    Given the level of deference we grant FERC, the
“indisputabl[e]” differences between Arizona and California’s
rates as illustrated by the dekatherm-mileage study, ALJ Op.,
139 F.E.R.C. at 66,228 , El Paso’s failure to provide substantial
evidence illustrating any uniqueness of Arizona’s laterals as
compared to laterals all over the southern portion of the system,
and the lack of other empirical evidence supporting
equilibration, we find FERC did not act arbitrarily or
capriciously in finding El Paso failed to meet its burden. In
evaluating equilibration, FERC was faced with a new,
unapproved practice for the pipeline. FERC pointed to clear
findings by the ALJ that El Paso’s dekatherm-mileage study --
which California Petitioners do not attempt to discredit --
“indisputably generates different average mileages for the
California and Arizona zones,” FERC Br. 65 (quoting ALJ Op.,
                               31
139 F.E.R.C. at 66,228), and El Paso proffered no comparably
detailed study, empirical cost comparison, or other analysis
undermining that finding, see Op. 528, 145 F.E.R.C. at 61,227.
FERC, limited to the record before it, reasonably found that
“based on the evidence in this proceeding, El Paso did not show
that its equilibration proposal would result in just and
reasonable rates.”
 Id. at 61,228. 
FERC acknowledged the cost
of Arizona’s laterals and other state-specific programs, but
reasonably found these expenses were “offset by other factors
including the integrated manner in which El Paso operates its
system, the mechanisms El Paso has to address costs associated
with the non-ratable deliveries of gas, and significantly
discounted rates for deliveries to California.”
 Id. at 61,227.
Perhaps, as FERC noted,
 id. at 61,227-28, 
a zone-of-delivery
rate design with only two east-west zones could have been
developed with the appropriate evidentiary support. However,
it was reasonable to find that an equilibration of this nature was
not supported by the dekatherm-mileage study or other equally
substantive evidence in the record, and thus could not be
deemed just and reasonable in this proceeding. Additionally,
FERC reasonably found Tennessee Gas distinguishable
because, unlike in that case where the laterals were limited to
the New England zone, El Paso’s laterals run across at least
three southern states, not just through Arizona, and El Paso
provided no empirical cost comparison or analysis justifying
recalibrating solely the Arizona rates in this way. See
 id. at
61,228.
    Next, California Petitioners argue that by approving a
zone-of-delivery design but rejecting equilibration, FERC
adopted a rate of its own making that was “substantially
different from El Paso’s proposal and that overstates the effect
of distance on rates.” Cal. Pet’rs Opening Br. 44. They assert
that although the zone-of-delivery method has been in effect on
the pipeline for over 50 years, FERC has never approved the
                                32
contract path methodology for measuring those zones, so its
partial approval was not an approval of the status quo but a rate
of its own design.
 Id. at 44. 
Such action, they argue, converted
this from a proceeding under Section 4, 15 U.S.C. § 717c, to a
proceeding under Section 5,
 id. § 717d, 
compelling FERC to
consider alternative rate proposals, including California’s
proposed “postage-stamp rate” methodology (i.e., a non-
distance sensitive rate). Cal. Pet’rs Opening Br. 44-50.

    Section 4 of the NGA “limits [FERC] to two courses of
action [in ruling on a ratemaking proposal], ‘acceptance (in
whole or part) or rejection of the pipeline’s proposed rates.’”
W. Res., Inc. v. FERC, 
9 F.3d 1568
, 1574 (D.C. Cir. 1993)
(quoting Sea Robin Pipeline Co. v. FERC, 
795 F.2d 182
, 183
(D.C. Cir. 1986)) (emphasis added). If the rate imposed by
FERC “differs significantly” from the rate proposed by the
pipeline, it can no longer be attributed to the pipeline or qualify
for Section 4 treatment, and the proceeding must be conducted
pursuant to Section 5.
 Id. at 1579. 
Section 5 requires a
showing that: (1) the pipeline failed to show its proposed rate
was just and reasonable under Section 4; (2) the default
position, the prior rate, is no longer just and reasonable; and (3)
FERC’s substitute rate is itself just and reasonable.
 Id. Under
this 
Court’s precedents, FERC can alter a proposed rate and
remain in a Section 4 proceeding as long as its change
represents “at least partial approval of the change” for which
the pipeline itself petitioned. Pub. Serv. Comm’n of N.Y. v.
FERC, 
642 F.2d 1335
, 1345 (D.C. Cir. 1980). But this Court
has rejected FERC’s argument that Section 4 permits it to
approve any rate, no matter how materially different from that
proposed by the pipeline, so long as it can be viewed as a “part”
of the original request. W. Res., 
Inc., 9 F.3d at 1579 
(finding
FERC proposed a rate that “differed substantially” from its old
rates by “employ[ing] a completely different strategy in
quantifying distinctions between the two kinds of service” and
                               33
adding a “50% backhaul rate”).

    Despite California Petitioners’ assertions, it seems clear
FERC approved “part” of the proposed rate design without
“differ[ing] substantially” from El Paso’s proposal.
 Id. Unlike
in 
Western Reserve where a novel methodology and rate
calculation schema were imposed by FERC, in approving the
zone-of-delivery design but rejecting the additional step of
equilibration, FERC simply left a version of El Paso’s
preexisting methodology in place and rejected slight changes
to rates in California and Arizona. See Op. 528-A, 154
F.E.R.C. at 61,682. Although this is the first time FERC has
approved the use of contract paths as a measurement tool for
the cost of transporting gas to the zones (El Paso proposed the
contract path methodology in its prior two rate cases, but both
resulted in settlements, see id.), this is merely a measurement
tool further supporting the five zones that have long been part
of El Paso’s rate design. The approved rate design simply does
not differ so substantially from El Paso’s original proposal that
the proceeding must now stand scrutiny under Section 5.

    Thus, FERC was correct in finding it unnecessary to
consider alternative rate proposals, including California
Petitioners’ preferred postage-stamp methodology. Under
Section 4, if a pipeline’s proposal is just and reasonable, FERC
must accept it (in whole or in part), regardless of whether other
just and reasonable rates might exist. See W. Res., 
Inc., 9 F.3d
at 1578
. As discussed above, California Petitioners made no
showing that El Paso’s proposed rates were unjust or
unreasonable, and as such there was no basis to consider
alternative rate designs under Section 5. See 15 U.S.C. § 717d;
see also Elec. Power Supply 
Ass’n, 136 S. Ct. at 784 
(stating it
is “not [the court’s] job” to supplant FERC’s reasoned,
explained choice of rate).
                          34
                          V.

For the foregoing reasons, the petitions for review are

                                                   Denied.