Opinion Per Curiam.
PER CURIAM:Several customers of Great Lakes Gas Transmission Company (Great Lakes), an interstate natural gas pipeline, petition this court to review dispositions of the Federal Energy Regulatory Commission (Commission or FERC) relating to Great Lakes’ sales and transportation rates. For the reasons stated in this opinion, we affirm the Commission’s rulings in part, and vacate and remand the agency’s orders in part.
I. Facts and Procedural Background
The Great Lakes pipeline system extends 973 miles eastward from the Canadian border near Emerson, Manitoba, to the Canadian border near Detroit. Great Lakes provides sales services, year-round transportation services, and seasonal transportation services. Four Great Lakes customers who use one or more of these services are participants in this review proceeding: petitioner TransCanada PipeLines Limited (TransCanada)1 is the primary user of Great Lakes’ year-round transportation service; petitioner Natural Gas Pipeline of America (Natural) uses the year-round transportation service and the sales service; intervenor Michigan Consolidated Gas Company (Mich Con) uses the sales service; and petitioner ANR Pipeline Company (ANR)2 uses several seasonal transportation services.
In November 1978, Great Lakes filed proposed rates under section 4 of the Natural Gas Act, 15 U.S.C. § 717c (1982). This filing included one of the seasonal transportation services used by ANR, called the T-6 service; however, Great Lakes proposed no increase in the T-6 rate. The Commission suspended the proposed rates and established hearing procedures. In August 1980, Great Lakes made another rate filing; this filing included, and proposed to increase, the rates for two other seasonal transportation services used by ANR, called the T-8 and T-10 rates. FERC also suspended these rates, and consolidated for hearings before an adminisT trative law judge (ALJ) certain issues from the August 1980 and the November 1978 filings. See Great Lakes Gas Transmission Co., 17 F.E.R.C. 11 63,029, at 65,-085-86 (Nov. 10, 1981) [hereinafter Initial Decision]. The Commission’s eventual rulings on these consolidated issues affirmed in part and modified in part the initial decision of the ALJ. Great Lakes Gas Transmission Co., 24 F.E.R.C. 1161,014 (July 8, 1983) [hereinafter Commission Decision]; Great Lakes Gas Transmission Co., 25 F.E.R.C. ¶ 61,319 (Nov. 30, 1983) [hereinafter Commission Clarification]. Portions of the Commission’s adjudication are challenged in this review proceeding.
At issue before this court are three Commission dispositions, each challenged by one or more petitioning Great Lakes customers. In addition, FERC’s stance requires us to address again the allocation of the burden of proof under section 5 of the Natural Gas Act, 15 U.S.C. § 717d (1982), as it differs from and interacts with the proof burden under section 4. Assignment of the burden of proof under these sections is not a novel question in this court; we ruled on the issue definitively in Public Service Commission v. FERC, 642 F.2d 1335 (D.C.Cir.1980) (Transco), cert. denied, 454 U.S. 879, 102 S.Ct. 360, 70 L.Ed.2d 189 (1981).
The essential features of the three substantive Commission determinations that petitioners challenge are summarized below.
Change in Zone Boundaries — For the purpose of setting transportation components of the pipeline’s rates, the Great Lakes pipeline is divided into three zones— Western, Central, and Eastern. Delivery points within each zone are charged the *318same rate for transportation. Because gas moves from west to east in the pipeline, delivery points in .the Western Zone pay the lowest transportation charge, those in the Central Zone the next lowest, and those in the Eastern Zone the highest. The three zones were originally established in 1971, and from that time until the FERC decision here under review the zones were of roughly equal length. Initial Decision, 17 F.E. R.C. at 65,102.
At the consolidated rate filing hearings, Mich Con, which purchases gas in both the Central and Eastern Zones, challenged the location of the boundary separating those two zones. Mich Con presented evidence indicating that the delivery points on the Great Lakes system form three clusters spaced roughly 200 miles apart.3 All of the delivery points in the first cluster fall in the Western Zone; all of the delivery points in the third cluster fall in the Eastern Zone. But the then existing zone boundaries, Mich Con pointed out, split the second cluster of delivery points between the Eastern and Central Zones. As a result, two second-cluster delivery points only forty-five miles apart paid different zone rates (one paid Eastern, the other paid Central), while some second- and third-cluster delivery points over 200 miles apart paid the same zone rates (both ranged in the Eastern Zone). Mich Con argued that in view of the map of delivery points showing three distinct clusters, a zone boundary line segmenting a cluster could not be regarded as just and reasonable. Mich Con proposed expansion of the Central Zone about 100 miles eastward so that all the second-cluster delivery points would be in the Central Zone. Id. at 65,103.
Upon finding that “Mich Con has shown by substantial evidence that the present zone boundaries are unduly discriminatory and preferential,” id. at 65,105, the AU adopted the Mich Con boundary relocation proposal to “correct[ ] th[e] unlawfulness.” Id. The Commission summarily affirmed. Commission Decision, 24 F.E.R.C. at 61,-062-63.
For delivery points formerly in the Eastern Zone and now in the Central Zone, the boundary shift will lower transportation rates. However, enlargement of the Central Zone will occasion a greater allocation of total costs to that zone and, consequently, an increase in the transportation rate for delivery points that have always been in the Central Zone. Natural, a purchaser of gas from Great Lakes’ delivery points located in the Central Zone from the start, seeks our review of the zone boundary shift. Intervenor Mich Con defends the Commission’s decision.
T-6 Service — Under the T-6 Service, Great Lakes transports ANR gas during the summer months to certain ANR storage facilities. Unlike the other seasonal transportation services performed by Great Lakes for ANR, the T-6 service was initiated after Great Lakes’ main pipeline was completed and its original capacity fully utilized. To accommodate the T-6 service, Great Lakes added to its main pipeline two sections of pipeline “loops” and a compressor facility. These additions increased the carrying capacity of the entire pipeline the full amount necessary for the T-6 service.4
Great Lakes used two methods to allocate pipeline costs to its various services. Services other than T-6 were allocated pipeline costs according to the “rolled-in” method. Under this method the costs of pipeline facilities were rolled in together and then divided among all customers according to the amount of service each received. Great Lakes employed a different *319method — the “incremental” method — to allocate costs to the new T-6 service. Instead of rolling in the costs of the loops and compressor with all other pipeline costs and dividing the total among all customers, Great Lakes allocated all costs related to the new facilities — and only those costs— directly to ANR as part of the T-6 rate. All other services continued to be charged costs covering the original main pipeline facilities under the rolled-in method.
The cost of the facilities added to the completed pipeline to provide the T-6 service exceeded the outlay that would have been required to provide equivalent capacity in the original pipeline. As a result, ANR paid higher costs relative to services received than did Great Lakes’ other customers.
From the initiation of the T-6 service in 1975 until the orders now under review, the Commission approved use of the incremental costing method, as just described, for Great Lakes’ T-6 rate. ANR Brief at 8-9. Great Lakes did not seek any change in the methodology used to allocate costs to the T-6 service. However, at the hearings the Commission staff proposed that the costs of the newer facilities be rolled in with all the other pipeline costs and divided among all Great Lakes customers. A staff witness noted that the newer facilities were fully integrated into the main pipeline, and that all customers’ gas flows through both the original pipeline and the later-added facilities.
The staff proposal, by equalizing the costs relative to the service paid by each Great Lakes customer, would have yielded lower costs for ANR and higher costs for Great Lakes’ other customers. ANR therefore supported, and other Great Lakes customers argued against, the staff proposal. Initial Decision, 17 F.E.R.C. at 65,109.
The ALJ rejected the staff “rolled-in” proposal, holding that ANR should continue to be allocated the full costs of the newer facilities. But the ALJ did not stop there; he did not simply reconfirm the status quo. Instead, he ordered on his own initiative a change in cost allocation that had not heen proposed by the staff, by Great Lakes, or by any other party to the proceedings. The ALJ ordered that in addition to the full cost of the newer facilities, the T-6 rate also bear a portion of the costs of the original main pipeline. Id. at 65,112.
The ALJ’s ruling thus increased the costs allocated to ANR. But Great Lakes had not requested an increase in the T-6 rate. Therefore, as the AU noted, Great Lakes could not collect a higher T-6 rate until it made a new filing. However, the ALJ ordered Great Lakes to reduce immediately its rates to other customers by the amount of costs shifted to ANR. Id.
The Commission affirmed the AU’s holding on the T-6 rate with little discussion. Commission Decision, 24 F.E.R.C. at 61,-063. In response to requests for clarification, FERC “formally acknowledge^] the affirmance of the initial decision as to this issue.” Commission Clarification, 25 F.E. R.C. at 61,727 n. 4.
ANR does not challenge the Commission’s decision to continue to allocate all costs relating to the new facilities directly to ANR; it challenges only the additional imposition of costs relating to the original main pipeline.
The Commission asserts that, because the T-6 rate was not actually increased by the orders at issue, ANR is not “aggrieved” by the challenged Commission action and therefore may not petition for review. See 15 U.S.C. § 717r(b) (1982).
T-8 and T-10 Services — TransCanada and Natural request our review of FERC’s decision to alter the method of costing company use gas in the T-8 and T-10 rate schedules.
The T-8 and T-10 rate schedules apply to seasonal, storage-related “backhaul” services introduced only a few years ago and used exclusively by one Great Lakes customer — ANR. These services are described as “piggy-back” because they could not be offered absent a preexisting pipeline. Backhaul is accomplished not by actually transporting gas back west (which *320would require a separate pipeline) but rather by performing a swap. When ANR wishes to ‘move’ gas from east to west, gas is removed from the eastward flowing stream at the western destination point. ANR then ‘pays back’ the pipeline by restoring the appropriate quantity of gas to the gas stream at the eastern origination point.
Under the T-8 service, Great Lakes back-hauls gas from the Farwell interconnection (milepost 815.7) and delivers it near Crawford, Michigan (milepost 753), where ANR stores the gas during the summer. In winter, the gas is removed from storage and returned (by forward haul) to Farwell. Under the T-10 service, Great Lakes back-hauls gas from Farwell and delivers to ANR near the Chester, Michigan interconnection (milepost 748.3), where ANR also has facilities for summer storage. In winter, ANR removes the gas from storage and redelivers it to Great Lakes, which then backhauls the gas another 330 miles upstream and delivers to ANR at the Fortune Lake/Crystal Falls interconnection (milepost 487.8).
Company use gas is the gas used or consumed by the pipeline itself, including gas lost in meters, valves, leaks, and seepage. Since 1979, when the T-8 and T-10 schedules were certificated, the company use gas component of these rates has been costed incrementally. For all transportation services other than T-8 and T-10, the costs of company use gas have been allocated on a rolled-in basis. In the proceedings under review, staff challenged the incremental costing of T-8 and T-10 company use gas as unjust and discriminatory, proposing that company use gas be costed according to the rolled-in (system average cost per Mcf mile) method used for other transportation rates. Company use gas is only one component of the T-8 and T-10 rates.
The staff raised the company use gas issue for the first time by serving the Supplemental Testimony of the staff’s industry economist on the parties to the administrative proceeding. The relevant portion of the testimony consisted of less than one page and was not accompanied by exhibits or other explanatory material. The only response to the staff economist’s Supplemental Testimony was some brief commentary in the Supplemental Testimony of a Great Lakes witness. Neither of the petitioners here presented any evidence (an omission which they now attribute to the inadequacy of notice). ANR, which alone stood to benefit from the switch to the rolled-in method, took no position on the issue.
In his initial decision, the AU noted the “inexplicable” silence of ANR but, relying on the Supplemental Testimony of the staff’s economist, asserted that the staff had presented “a patent case” of preference and discrimination and ruled that the rolled-in method must henceforward apply. The AU applied Transco but concluded that because the Commission had made out a prima facie case, the burden shifted to the defenders of the status quo to justify the disparity between the T-8 and T-10 rates and other transportation rates.
The Commission Decision (Opinion No. 179) rejected exceptions taken by petitioners and by two other Great Lakes customers (Northern and Central Gas Corporation and Union Gas Limited). The Commission Clarification (Opinion No. 179-A) reiterated FERC’s approval of the initial decision’s company use gas holding; the fact that “the T-8 and T-10 services are in some ways functionally unlike other transportation services ...,” FERC stated, “does not mean ... that similar items, e.g., company use gas, should not be treated similarly.” The Commission added that “[cjompanyuse gas, like other gas for other services, is commingled in the pipeline. As such, it is not possible to trace the flow of particular molecules of gas and not always easy to determine on a customer-by-customer basis the benefits derived thereby.” FERC also accepted the representation made in Staff’s Brief Opposing Exceptions that T-8 and T-10 actually save other customers some gas compression costs because “some volumes of gas which are delivered to them *321did not traverse the entire length of the system en route.”
In this court, TransCanada argues that it lacked adequate notice and opportunity for hearing on the company use gas issue. TransCanada and Natural contend that the Commission conclusion is unsupported by the record. They also contend that the Commission improperly ignored the existence of negotiated agreements between them and ANR mandating the use of incremental costing for the T-8 and T-10 schedules.
II. Burden of Proof
In Transco, 642 F.2d 1335 (D.C.Cir.1980), this court focused on the different burdens of proof that operate when the Commission acts under section 4(e) of the Natural Gas Act, 15 U.S.C. § 717c(e) (1982), and under section 5(a) of the Natural Gas Act, 15 U.S.C. § 717d(a) (1982).
Section 4(e) provides in part that “[a]t any hearing involving a rate or charge sought to be increased, the burden of proof to show that the increased rate or charge is just and reasonable shall be upon the natural-gas company.” Section 5(a) authorizes the Commission to “determine the just and reasonable rate, ... classification, [or] practice, ... and [to] fix the same by order” whenever, after a hearing, it “find[s] that any rate, charge, or classification ... collected by any natural-gas company ... or that any ... practice ... affecting such rate ... is unjust, unreasonable, unduly discriminatory, or preferential.”
As we explained in Transco, sections 4(e) and 5(a) are not alternatives within FERC’s discretion to select; each is a specific and limited grant of authority to the Commission. The Commission’s authority under section 4(e) is limited to review of increases proposed by the natural gas company. When the Commission seeks to impose its own rate determinations — rather than to accept or reject a change proposed by the company — the Commission must act under section 5(a). These different sources of authority are attended by different burdens of proof: When a natural gas company seeks a rate increase, it bears the burden of showing that the increase is just and reasonable; but when the Commission seeks to impose a rate change not proposed by the company, the Commission must first find that the existing rates are unjust or unreasonable, and its finding must be supported by substantial evidence. Transco, 642 F.2d at 1345.
The ALJ in this case applied Transco correctly; he placed the proof burden as to any particular change in rates on the party proposing the change — be it Great Lakes, the intervening Great Lakes customers, or the Commission staff. Initial Decision, 17 F.E.R.C. at 65,087-88. On administrative appeal, the Commission disregarded the message of Transco. Instead of following the instruction that decision supplied, FERC declared that its mission was “to assure a just and reasonable end result” and that, in arriving at such a result, it need only “provide a reasoned explanation to support its conclusions.” Commission Decision, 24 F.E.R.C. at 61,-062.
Before this court, FERC does not attempt to defend its previous position that, in imposing an unsolicited change, it bears no evidence-adducing onus in any case. The Commission appears to acknowledge that it was obliged to adduce substantial evidence tending to show that the then existing T-6 rate, which Great Lakes had not proposed to change in any way, was unjust or unreasonable. See Brief for Respondent Federal Energy Regulatory Commission at 20. But FERC argues that the T-8 and T-10 company use gas charges and the location of zone boundaries were unchanged parts of rate increases proposed by Great Lakes, and therefore Great Lakes had the burden of showing that these unchanged parts continued to be just and reasonable. To support its argument, FERC points to cases holding that when an unchanged component of a proposed rate increase is an integral part of the rate, FERC may exercise its authority under sec*322tion 4 (or its Federal Power Act counterpart section 205) to suspend the rate increase in its entirety and order refunds after investigation. See Cities of Batavia v. FERC, 672 F.2d 64, 77 (D.C.Cir.1982) (Federal Power Act); Laclede Gas Co. v. FERC, 670 F.2d 38, 41-42 (5th Cir.1982) (Natural Gas Act).
The reasoning of the court decisions cited by the Commission does not inevitably lead to the position FERC espouses here. Neither decision discusses burden of proof. Batavia and Laclede hold only that when a proposed increase interacts with unchanged aspects of a rate so as to create a new unreasonable rate, the purposes of section 4 (or section 205 of the Federal Power Act) would be thwarted if the Commission did not have suspension and refund authority over the changed and unchanged components of the proposed higher rate. See Batavia, 672 F.2d at 76-77; Laclede, 670 F.2d at 41-42. The proof burden allocation as set forth in Transco, on the other hand, does not foil FERC’s remedial mission. The Commission is in no case prevented from substituting its own just and reasonable rate for any provision found unreasonable; all that is required, for those changes not proposed by the natural gas company, is that the Commission actually make a finding — supported by substantial evidence — that the replaced provision is unjust or unreasonable.
Adhering to Transco, we hold that when the Commission imposes a change not proposed by the natural gas company— including an alteration in an unchanged part of a proposed higher rate — it must first find that the existing provision is unjust or unreasonable. As we observed in Transco, this rule comports with the policy of the Natural Gas Act to have rates set by pipelines, to be set aside and replaced by the Commission only when the privately-ordered rates are unreasonable. Transco, 642 F.2d at 1343-45. Moreover, by requiring each proposed change to be justified by the proponent of the change, the focus of evidence and hearings will be trained on the provisions in fact in controversy. If, on the other hand, the pipeline had to justify every aspect — old and new — of an increased rate, it would be forced to gather for presentation before the Commission evidence on unchanged provisions that might never be challenged. “Forcing the petitioning company to justify not only the novel portions of its petitions but the unchanged parts as well would seriously increase the burden upon these regulated companies without any corresponding improvement in reasoned decisionmaking.” Id. at 1345.
None of the three changes at issue in this case was proposed by Great Lakes; therefore the burden, as delineated in Transco, rests with the Commission to justify each.
III. Change in Zone Boundaries
We uphold the Commission’s decision to shift Great Lakes’ Central Zone/Eastern Zone boundary. As the ALJ determined, and the Commission confirmed, substantial evidence presented by Mich Con demonstrated that the previous zone arrangement was not reasonable.5
Mich Con, we have already recounted, see supra pp. 509-510, showed that the delivery points on the Great Lakes system formed three clusters, and that the Central Zone/Eastern Zone boundary line cut through the middle cluster of delivery points. Mich Con urged, persuasively, that the previous boundary placement violated the ratemaking principle that zone rate differentials should reflect transportation cost differentials. First, Mich Con observed, by dividing the zones in the midst of closely clustered delivery points, the Central/Eastern boundary created a significant rate differential where no significant transportation cost differential existed. Second, Mich *323Con pointed out, by placing some second-cluster delivery points and the third-cluster delivery points — some 200 miles away — in the same (Eastern) zone, the old boundary “fail[ed] to reflect a significant cost differential that does exist between deliveries to points in the westernmost portion of the Eastern Zone and deliveries to points in the easternmost portion of that zone.” Initial Decision, 17 F.E.R.C. at 65,103.
None of the participants before the agency disputed the existence of the three delivery point clusters, and none could convincingly attack the reasonableness of the conclusions drawn from the cluster analysis. The record we review, in short, provides ample justification for the AU’s finding of “substantial evidence that the present zone boundaries are unduly discriminatory and preferential.” Id. at 65,105.
Petitioner Natural, opposing the change, argues that the previous zone boundaries deserve “deference” because they had been unchanged for ten years. Brief of Petitioner, Natural Gas Pipeline Company of America at 32. But all the respect due an existing rate was afforded in this instance when the ALJ tested Mich Con’s position under the burden of proof analysis clarified in Transeo. Natural also objects that the ALJ did not adequately consider Natural’s suggestion that a fourth zone be created. The ALJ observed: “Natural has not presented this proposal in any detail. Consequently, its effect is unknown and it surely is not supported by substantial evidence.” Initial Decision, 17 F.E.R.C. at 65,105. We find this observation accurate and adequate. Further objections raised by Natural, while accorded our full attention, do not warrant discussion.
IV. The T-6 Rate Schedule
A. The Standing Question
Before turning to the merits of the T-6 issue, we must first address a threshold question whether ANR has standing to petition for review of FERC’s orders concerning the T-6 rate schedule. Under section 19(b) of the Natural Gas Act, 15 U.S.C. § 717r(b) (1982), a party must be “aggrieved” by an order of the Commission to have standing to obtain judicial review.6
The Commission argues that ANR is not “aggrieved” by its order because the projected increase in rates to ANR resulting from the new rate methodology approved for the T-6 schedule cannot be effected in this proceeding. Because Great Lakes did not file for an increase in the T-6 rates as part of this proceeding, the projected increase must await a separate rate filing. Thus, the Commission contends that ANR has yet to suffer any injury requisite to attain standing in this court. We reject this contention and hold that, under the Natural Gas Act, ANR plainly has standing to contest the T-6 rate determination in this proceeding.
Aggrievement is essentially fact-specific. As we have previously observed, “[wjhether a party is aggrieved within the meaning of section 19(b) and similar statutes is not controlled by any general rule but must be determined on the basis of the specific facts in each case.” Northwestern Public Service Co. v. FPC, 520 F.2d 454, 457 (D.C.Cir.1975) (citing United States ex rel. Chapman v. FPC, 345 U.S. 153, 156, 73 S.Ct. 609, 612, 97 L.Ed. 918 (1953)). The test is whether a party has sustained injury-in-fact to an interest arguably within the zone of interests to be protected or regulated by the Natural Gas Act. Id. at 458. Cf. Sierra Club v. Morton, 405 U.S. 727, 733, 92 S.Ct. 1361, 1365, 31 L.Ed.2d 636 (1972). We have interpreted the injury-in-fact requirement to include not only present and immediate harm, but also a “looming unavoidable threat” of harm. Cincinnati *324Gas & Electric Co. v. FPC, 246 F.2d 688, 694 (D.C.Cir.1957); see also Northwestern Public Service Co., 520 F.2d at 458 n. 6. In other words, it is sufficient for standing purposes if a party has an “immediate prospect of future injury.” Transwestern Pipeline Co. v. FERC, 747 F.2d 781, 785 (D.C.Cir.1984).
In the instant case, there is no doubt that the Commission has made a determination that is adverse to ANR’s interest. Absent judicial review of this determination, ANR likely will be bound by the Commission’s order in any subsequent filing by Great Lakes to increase the T-6 rates. In addition, it appears unavoidable that Great Lakes will file for an increase in the rates it charges to ANR on the T-6 schedule, as a result both of the Commission’s order changing the methodology, and of the Commission’s order lowering the T-6 rate for other customers. The record and arguments in this case compel us to conclude that Great Lakes does intend to file for higher T-6 rates for ANR; indeed, neither Great Lakes nor the Commission disputes this point.
Accordingly, we find that the injury-in-fact requirement is met in this case. Lynchburg Gas Co. v. FPC, 336 F.2d 942, 945 (D.C.Cir.1964) (standing based on “economic injury [that was] likely to flow from the action sought to be reviewed”). It is uncontested that the interest involved is arguably within the zone of interests to be protected or regulated by the Natural Gas Act. Thus, ANR has standing to contest the Commission’s T-6 determination.
B. The T-6 Rate Methodology
The Commission ordered Great Lakes to utilize simultaneously two rate methodologies to assess ANR for the T-6 service — both an incremental rate to cover the cost of the loop’s facilities and services, and a pro rata share of the rolled-in costs of the mainline system. This new methodology resulted in an immediate lowering of T-6 rates for Great Lakes’ other customers, with, as explained above, the increase to ANR to occur later.
ANR protests that this methodology, which was not suggested by either Great Lakes (which did not propose any change in the existing T-6 rate methodology) or by the Commission’s staff (which proposed changing the existing methodology from an incremental to a rolled-in method), “requires ANR, its customers, and the ultimate gas consumer, to pay twice for the same service.” Brief of ANR at 10. Because we are unable to find substantial evidence in the record to support the Commission’s conclusions concerning the T-6 rate, we reverse this portion of the Commission’s order.
Although agency expertise is entitled to respect in a case of this sort, we nonetheless must ascertain that the Commission’s order is supported by substantial evidence and reached by reasoned decision-making — that is, a process demonstrating the connection between the facts found and the choice made. Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 167-68, 83 S.Ct. 239, 245, 9 L.Ed.2d 207 (1962); Electricity Consumers Resources Council v. FERC, 747 F.2d 1511, 1513-14 (D.C.Cir.1984); see also City of Charlottesville v. FERC, 661 F.2d 945, 950 (D.C.Cir. 1981) (“What is basic is the requirement that there be support in the public record for what was done.”) (citation omitted).
In this case, we can fathom no support in the record for the Commission’s decision regarding the T-6 methodology. The record presents evidence in favor of either the incremental or the rolled-in method, but not both. The staff originally proposed changing from the incremental to the rolled-in method for two reasons: first, that it was unlawful and unjust for ANR not to pay a portion of the rolled-in costs of the mainline system reflecting its use; second, that there were actual or potential benefits flowing from the loop to all customers. On this latter point, the staff’s witness, Mr. Zebot, testified that
Great Lakes’ system is a single pipeline, approximately 1,100 miles long ... [along which] [s]ales and transportation services are rendered at various *325points____ The pipeline is thus an integrated system. In my opinion, the facilities utilized to provide the T-6 basic service are a part of the integrated system, and the associated costs should be allocated on a rolled-in basis as is currently done for the other rate schedules.
J.A. 112-13. Zebot viewed the looped portion of the system as a potential benefit to all customers because “there is nothing in the certificate or the contract underlying the T-6 service that restricts the use of the looping facilities solely to the T-6 schedule or solely to any specific customer or customers.” Id. Therefore,
should it be necessary to shut down Great Lakes’ main line along this looped portion of the system, Great Lakes would be able to use this looping to continue service to its customers. The net effect is that the looping provides additional pipeline system security, a potential benefit to all customers. [Also,] ... the looping is actually available for serving other customers [and] ... actually has been used to provide service for other customers.
J.A. 113.
It is uncontroverted that Zebot was proposing to substitute the rolled-in methodology for the existing incremental methodology, and not to combine the methods. He testified that Great Lakes had not justified its use of the incremental method, id., and that that method was improper in an integrated system. J.A. 113, 114.' Therefore, according to the staff’s witness, “[t]he more appropriate rate is a rolled-in cost approach.” J.A. 114.
The staff counsel, Ms. Boyle, testified that the staff viewed the incremental rate design on the T-6 schedule to be a carryover that had never been determined to be just and reasonable. J.A. 191. She explained that,
incremental costing is unlawful because ... there is no other service, even though there are similar services such as T-8 and T-10 services, that are costed incrementally, so this is preferential treatment and it is unjust and unreasonable because the incremental costs of the T-6 service has [sic] decreased due to reduction in volumes, but actual costs have increased due to increased fuel costs____
Furthermore, the incremental costing fails to recognize that other customers benefit from the facilities, such as by added system security and fuel savings—
____ [R]olled-in costing is just and reasonable because rolled-in costing is appropriate for an integrated system which [Zebot’s] testimony shows Great Lakes to be.
J.A. 194. In short, staff presented no evidence or argument to support an incremental methodology, either by itself or in combination with rolled-in costing. In staff’s view, incremental costing is unlawful.
On the other hand, TransCanada’s witness argued that the incremental, not the rolled-in, methodology should remain as the basis for the allocation of costs under the T-6 service. J.A. 389-92, 430-31. Natural Gas Pipeline’s witness also supported incremental costing, J.A. 481-82, as did Great Lakes’ witnesses. J.A. 503D-E; 503H-J; 514C-E. Great Lakes’ consultant, Mr. Chinn, agreed that the incremental method should be maintained. He explained:
I disagree with Mr. Zebot on his use of a rolled-in method to design the rate for the transportation service under the T-6 Rate Schedule. Mr. Zebot’s argument boils down to the proposition that if at any time a transmission company provides a new service for the benefit of single customer ... the rates of the existing customers of such transmission company should be increased and they should be required to bear additional costs related to the new facilities because there may be a potential ancillary benefit, however small, to some of such existing customers. Mr. Zebot ... [reasons] ... that Great Lakes operates an integrated system and, therefore, that the incremental facilities also benefit other system customers____ The plain fact is that the other customers do not benefit *326from this service____ Mr. Zebot alone proposes without factual basis that the T-6 facilities benefit other customers on the system, an assumption which simply is not supportable.
J.A. 547C-D.
Thus, the only evidence in the record concerning the T-6 schedule supports the adoption of either the rolled-in method of cost allocation (the staff proposal) or the incremental method (the proposal of Great Lakes and its customers), not both. The AU discussed the evidence for and against the adoption of either method, J.A. 2545-48, and concluded that the staff had “not met its evidentiary burden” of showing that the incremental method was preferential and unduly discriminatory. J.A. 2549. Applying a functional test, the AU found that, in fact, the system was not integrated because the looped facilities did not benefit all customers. He concluded that, “since there must be ‘equal treatment for customers receiving equal service,’ the T-6 rate should be computed on an incremental basis.” J.A. 2550.
Having determined that the incremental method should apply, however, the AU added sua sponte that the
[s]taff is correct, however, in stating that the T-6 rate is unlawful because it does not include any portion of the cost for use of the mainline system. Although the Great Lakes system is not sufficiently “integrated” to require the rolling-in of the cost of the T-6 facilities, it is uncontroverted that the mainline system is used to provide the T-6 service to Mich Wis. Therefore, Mich Wis should bear its share of the costs of the mainline system.
* * * * * '
Upon a review of the record ... it is held that substantial evidence of record supports the conclusion that the T-6 rate should continue to be calculated on an incremental basis, and further, that the mainline fixed costs should be allocated among all customers.
J.A. 2550.
Although we agree that substantial record evidence supports the continued use of the incremental basis, we find absolutely no record evidence, not even a “mere scintilla,” to support the use of the rolled-in method as well. Since our review is based on the substantial evidence standard, and since we find that standard not to be satisfied here, we deem it unnecessary to discuss the precedents cited by the parties. Our review is based on the record in this ease. If there were evidence in the record supporting the concurrent adoption of the two methods of cost allocation, we would examine past judicial and Commission decisions to determine whether the statute permits such an approach. We need not reach that question here because we lack the factual predicate. Accordingly, we remand the case to the Commission. On remand, FERC should either supplement the record before concluding that the combined methods are appropriate, or should elect one of the methods advocated by the parties on the basis of the present record.
V. The T-8 and T-10 Rate Schedules
TransCanada and Natural contend that the Commission’s decision with respect to Rate Schedules T-8 and T-10 is unsupported by the record and further contend that the Commission erroneously neglected to consider the existence of negotiated agreements between the parties in reaching its determination. We agree and consequently remand this portion of the case to the Commission. Because we find remand in order in any event, we do not address the question whether the minimal notice afforded by the Commission was adequate under the circumstances.
The sole record evidence underlying the Commission’s decision is the Supplemental Testimony of Cyril J. Zebot, the staff’s industry economist. Zebot was first asked whether he agreed with the rate design for Schedules T-8 and T-10. He testified that
[although the concept is similar to my proposal for all transportation rates, I feel it more appropriate to use the same Mcf mile approach used in my direct testimony for these rate schedules.
*327Zebot was then asked to explain the rate design for Schedules T-8 and T-10. Zebot stated:
Rate Schedules T-8 and T-10 are based on a rolled-in or a systemwide basis of allocation. ;This is the same basic type of allocation that I have proposed for all of Great Lakes transportation service, including the T-6 rate design. The difference between my proposal and Great Lakes’ T-8 and T-10 rate design is quite subtle; Great Lakes’ proposal credits the revenues to the Cost of Service based on a rate per Mcf mile; whereas my proposal allocates cost in the same manner as the zone allocation.
Finally Zebot was asked to explain the difference between this rate design and the rate design for the T-6 service. He merely asserted:
Great Lakes uses an incremental rate design for the T-6 service presumably because the incremental cost is higher than the rolled-in allocation of costs. The T-8 and T-10, as explained above, are determined on a rate per Mcf mile as presented by Great Lakes in Docket No. RP80-134.
J.A. 123-24. This was the entirety of the testimony relied upon to give adequate notice to all interested parties that company use gas was in question and to provide substantial evidence for the Commission’s deviation from the status quo. The ALT relied on these sentences both for the assertion that “staff witness proposes that the cost of company use gas included in the T-8 and T-10 rates be calculated on the basis of the total system Mcf-miles,” 17 F.E.R.C. at 65,100, and for the conclusion that staff has presented “a patent case of preference and undue discrimination within a class of customers.” Id. at 65,101.
We find the testimony too slim to support the conclusions based upon it. All concede that incremental costing was the certificated method and constituted the status quo. Consequently, the burden was on the Commission to establish that the resultant rates are unjust and unduly discriminatory within the meaning of the NGA. Zebot’s testimony does not establish that ANR is “within the same class of customers.” It does not establish that the seasonal, storage-related “backhaul” services to which schedules T-8 and T-10 pertain are the same as or substantially similar to the basic year-round forward-haul transportation services. In light of the Commission’s own observation that the “T-8 and T-10 services are in some ways functionally unlike other transportation services,” 25 F.E. R.C. at 61,726, the Commission has failed to present substantial evidence that the differing treatment of these rates was unjust or unreasonable.
TransCanada and Natural have also contended that the Commission’s decision is deficient in that neither the ALJ nor the Commission gave appropriate consideration to the fact that the decision to employ incremental costing for company use gas derived from a negotiated agreement among Great Lakes customers. FERC seems to acknowledge that this agreement existed and that neither the AU nor the Commission took it into account, but contends that “the fact that these agreements were negotiated by all the parties” is “irrelevant.” FERC’s Brief at 24.
The existence of contractual arrangements among the parties is not binding on FERC or dispositive of the question of the legality of rates. The Commission is not justified, however, in cavalierly disregarding private contracts. This court has said that
Because the preservation of private contract within the context of a rate-setting statutory scheme promotes economic stability, the Supreme Court held in the Mobile and Sierra cases, that statutory provisions governing public utilities’ rates should be construed, when possible, as compatible with private rate agreements.
Cities of Bethany v. FERC, 727 F.2d 1131, 1137, 1139. (D.C.Cir.1984). This court has further suggested that settlement agreements can constitute “a factual difference that may justify a rate disparity.” Id. *328Other cases have pointed out that the Commission is obligated to consider equitable considerations that may derive from contracts in making its determinations. Thus the Commission’s assertion that contracts are irrelevant is erroneous, and the Commission’s failure to take the existence of the negotiated agreements into account is a material deficiency in its reasoning.
In sum, we are not convinced that the “Commission has given reasoned consideration to each of the pertinent factors.” Cf. In re Permian Basin Area Rate Cases, 390 U.S. 747, 792, 88 S.Ct. 1344, 1373, 20 L.Ed.2d 312 (1968); Transco, 589 F.2d 542, 549 (D.C.Cir.1978). Consequently, we remand this portion of the case for more thorough consideration and a more carefully reasoned determination. In so ruling, we do not suggest that a change to rolled-in costing is necessarily unacceptable. We hold only that the present record is madequate to support such a departure from the negotiated status quo.
Conclusion
This case involves three separate challenges to FERC’s decision with respect to rateS 0n the Great Lakes system. It also raises an important question as to the prop-er allocation of the burden of the proof, We find that our previous decision in Tran-sco controls this case and that the Commission erred in ruling otherwise. Our review the. record indicates the Commission s decision to alter the zone boundaries was reasonable and supported by substantial evidence. We uphold the Commission on that issue- We find> however, that in light of Transco the Commission’s determinations with resPect to the T“6 addltlonal facilities issue and the T-8 and T-10 corn-Pany use £as *ssue 'a<dc adequate record suPPort and do not reflect reasoned consideration of a11 the Pertinent factors. Consewe reverse those asPects of the Commission’s decision and remand to the a£ency for further Proceedings,
It is so ordered.
. TransCanada owns 50% of the equity of Great Lakes.
. ANR was formerly Michigan-Wisconsin Pipeline.
. A map showing the delivery points on the Great Lakes pipeline is appended to this opinion.
. The "loops" are additional sections of pipe, laid parallel to portions of the existing pipe, which empty into the existing pipe at both ends of the "loop pipeline." This has the effect, because of the characteristics of gas physics, of increasing the carrying capacity of the whole pipeline. Similarly, compressor facilities increase the capacity of the whole pipeline, by compressing the gas so that the same pipe can carry more.
Initial Brief of Petitioner ANR Pipeline Company at 5-6.
. Although the Commission erroneously believed it was unnecessary to find the prior arrangement unreasonable, see supra p. 513, it stated that "once the showing [of an unjust and unreasonable practice] was made, as here, the [initial] decision cannot be faulted for changing the practice to one which reason shows to be just and reasonable." Commission Decision, 24 F.E.R.C. at 61,062-63.
. Section 19(b) of the Natural Gas Act provides in relevant part that
[a]ny party to a proceeding under this chapter aggrieved by an order issued by the Commission in such proceeding may obtain a review of such order in the court of appeals of the United States for any circuit wherein the natural-gas company to which the order relates is located or has its principal place of business, or in the United States Court of Appeals for the District of Columbia____
15 U.S.C. § 717r(b) (1982).