Opinion for the Court filed by Circuit Judge BAZELON.
BAZELON, Circuit Judge:Petitioners seek review of Federal Energy Regulatory Commission orders1 issued in Opinion Nos. 7922 and 792-A,3 which set rates for the Texas Transmission Corporation (Texas Gas),4 an interstate natural gas *295transmission company.5 Petitioners Columbia Gas Transmission Corporation (Columbia), Consolidated Gas Supply Corporation (Consolidated)6 and intervenor, Public Service Commission of the State of New York (New York),7 challenge the Commission’s method of allocating the pipeline’s fixed storage and transmission costs among the pipeline’s rate zones as well as the Commission’s method of designing rates within each of the rate zones. Petitioners Texas Gas and Louisville Gas and Electric Company (Louisville)8 challenge only the Commission’s method of allocating fixed costs. Intervenor Memphis Light, Gas and Water Division (Memphis)9 supports the Commission in all respects.
The present dispute arose when the Commission decided to employ the methodology first set out in United Gas Pipeline Co.10 (the United formula) to classify costs for purposes of both cost allocation and rate design on the Texas Gas pipeline. The Commission thus departed from the Atlantic Seaboard Corporation11 methodology (the Seaboard formula), which it had used in allocating costs and designing rates on the Texas Gas pipeline system for nearly twenty-five years.12 At issue is whether *296the Commission’s decision to change from Seaboard to United meets the standards set forth in section 4 of the Natural Gas Act.13 Because we find that the Commission has failed to provide an adequate explanation for its changed approach, we remand the case to the Commission for further consideration.
I. BACKGROUND
On September 30, 1974, Texas Gas filed for a general rate increase,14 using the Seaboard formula for cost classification and allocation, and the United formula for rate design.15 The Commission accepted the filing but suspended the proposed increase until April 1,1975, and ordered a hearing on *297the justness and reasonableness of the proposed increase.16 Prior to the hearing, a settlement agreement resolved most of the issues raised by the filing, but several were reserved for hearing before an Administrative Law Judge (ALJ), including the proper methods for cost allocation and rate design.17 This hearing was held on December 2 and 3, 1975.
Texas Gas, Columbia, Consolidated, and the Commission’s own staff argued at the hearing that the Seaboard formula should continue to be used without modification for classification and allocation of costs among the pipeline’s four rate zones. In support of the Seaboard formula, the parties presented evidence that they claimed showed significant differences between the United pipeline in the Consolidated decision 18 and the Texas Gas pipeline. These differences included annual and seasonal but not peak-day curtailment on the pipeline, use of a demand charge credit provision, and the absence of significant non-jurisdictional sales by Texas Gas. The parties also presented evidence to show differences in the end use markets served by the United and Texas Gas pipelines and to show that the application of United discriminated against customers that had invested in storage facilities to improve their load factors.19 Memphis, the sole proponent of United formula for allocation of costs, simply contended that this method would more fairly distribute the costs of the pipeline’s unutilized capacity.
The Commission Staff20 supported adoption of the United formula for rate design. The staff claimed that the United formula would reduce the price discount that the Seaboard formula afforded large volume users, that the United formula more nearly reflects pipeline usage during curtailment, and that it would narrow the gap between the cost of natural gas and alternative fuels. Columbia and Consolidated argued for retention of the Seaboard formula for rate design. They contended that the United formula would unfairly penalize high load factor customers with storage while failing to limit consumption of natural gas. Further, they contended that, as with classification and allocation of costs *298among rate zones, the factual characteristics of the Texas Gas pipeline render the rationale of our Consolidated decision inapplicable for allocating costs among customers within a rate zone.
The ALJ found that the United decision was inapplicable to the Texas Gas pipeline and that the Seaboard formula was “just and reasonable” both for allocation of costs and for rate design.21 Use of the United formula for classification and allocation of costs among rate zones, he wrote, would be inconsistent “with general Commission policy” because
the United method in the instant case would not produce the result sought in United, but would simply shift substantial costs from one group of jurisdictional customers to another under circumstances where end use characteristics of both groups are substantially the same. . Such a result would not be just and reasonable or in the public interest.22
Similarly, with regard to rate design, the ALJ stated that
[a] change from the long established Seaboard formula of rate design to the United method is clearly unwarranted on the evidence in this proceeding. The rate design in United was adopted from the facts presented therein, to wit, low priority industrial sales, peak day curtailments, and the absence of demand charge credits for curtailed volumes, which do not obtain in the instant case.23
The Commission overruled the ALJ, and approved the United formula both for cost allocation and for rate design. “The most critical factor in favor of using the United method,” the Commission found, “[was] the degree of curtailment on the Texas Gas system.”24 Adoption of the United formula, the Commission asserted, would result in a distribution of cost responsibility that more closely parallels customer usage of the pipeline’s facilities.
In a brief opinion, the Commission subsequently denied the applications for rehearing and reconsideration filed by Texas Gas, Louisville, Columbia, and Consolidated, finding that “[n]o facts or issues [had] been raised . . . which warranted] any modification” of its opinion.25 The Commission added, in passing, that “possible minor anomalies in [use of the United formula] do not render [it] inappropriate for use on the Texas Gas System.”26
II. THE COMMISSION’S SHIFT FROM THE SEABOARD FORMULA TO THE UNITED FORMULA
The Natural Gas Act does not specify any specific formula for the Commission to use in arriving at just, reasonable and nondiscriminatory rates.27 We have recognized that, “[i]n setting rates, the Commission is ‘free, within the ambit of [its] statutory authority to make the pragmatic adjustments which may be called for by particular circumstances.’ ”28 But where, as here, the Seaboard formula has been used for many years, it is a “starting point” in determining “overall reasonableness” of pipeline rates.29 Thus, in this case, *299the Commission must not only show that the substitution of United for Seaboard results in rates that meet the standards of § 4 of the Act;30 the Commission must also provide a reasoned explanation for its decision to depart from the Seaboard formula, which it has applied to the Texas Gas pipeline for nearly twenty-five years.31
A. Use of the United formula for cost classification and allocation among the pipeline’s rate zones.
Initially, we note that the allocation of costs among rate zones and the design of rates to recover those costs within a zone are two separate and distinct matters,32 that sometimes turn on conceptually different considerations. Cost classification and allocation for the Texas Gas pipeline involve determining the pipeline’s total costs of rendering service and the portion of those costs incurred by each of the four rate zones.33 Rate design, on the other hand, involves recovering from individual customers within each rate zone their respective share of the zone’s cost responsibility.34
The undisputed effect of the Commission’s decision to use the United formula for cost allocation is to shift more than $3 million in cost responsibility from the jurisdictional customers35 in Rate Zones 1, 2, and 3 (in the southern part of the system) to the jurisdictional customers in Rate Zone 4 (at the northern terminus of the system).36 The Commission itself noted that this substantial inter-zone reallocation of cost responsibility is “the most significant *300impact resulting from the use of the United method.”37 Yet the Commission fails to provide an adequate explanation for or any evidence in support of this geographical shifting of costs. Instead the Commission offers only the conclusory finding that
[t]he United method of cost classification and cost allocation is appropriate for use on Texas Gas’ system because it appropriately allocates costs . . . among Texas Gas’ four rate zones.38
1. Curtailment
The Commission indicates that
the primary reason for the adoption of the United method in this case is the impact of the current natural gas shortage on the nation as a whole, and on Texas Gas and its customers in particular. Ratemaking concepts appropriate to a period when natural gas pipelines were struggling to find markets are not appropriate ... in the present era of curtailments.39
In support of its position, the Commission relies on our decision in Consolidated;40 in which we upheld, for the first time, the Commission’s use of the United formula.41 Petitioners, however, contend (as did the ALJ) that certain factual conditions critical to our rationale in Consolidated are absent from the record in the present case.
The record in Consolidated demonstrated that the United pipeline system was experiencing substantial peak-day curtailment,42 i. e., even on peak-days, curtailment resulted in unused pipeline capacity.43 The limiting factor on demand thus became gas supply instead of pipeline capacity. We therefore concluded that since the demand charge on the United pipeline “no longer [had] the same vitality as a premium for reserving priority use of a scarce resource,”44 the Commission had demonstrated a reasonable basis for reducing the pipeline’s demand charge.
The record now before us, however, does not provide the Commission with an analogous basis for reducing the demand charge on the Texas Gas pipeline. Peak-day use has not been curtailed — on peak-days full pipeline capacity is used.45 The Commission *301fails to confront the implications of this difference. Instead the Commission argues that Consolidated is applicable here because both the Texas Gas and United pipelines also have experienced annual curtailment. The Commission relies on its bare assertion that the result it reached “is at most an extension of [the United ] policy,”46 thereby avoiding the issue raised by lack of peak-day curtailment. The Commission’s failure to explain why the absence of peak-day curtailment is irrelevant undermines the adequacy of its justification for extending Consolidated to the facts of this case.47
The record contains substantial, unrebutted evidence showing that annual curtailment, alone, does not diminish the vitality of the demand charge (as did peak-day cur*302tailment in Consolidated ).48 On the Texas Gas pipeline, each customer remains free on any given day to take its full contract demand, even on peak-days, so long as its annual entitlement is not exceeded.49 Volumetric demands on peak-days have not decreased, notwithstanding annual curtailment.
Since the facilities constructed to meet peak-day demands continue to be used at pre-curtailment levels, petitioners contend that the Commission’s reallocation of fixed costs is unjustified. The Commission’s opinion fails to answer this contention. The Commission does not explain why the investments that created adequate capacity for peak-day service and the fixed charges associated with these investments now afford less benefit to customers that require peak period service than they did prior to annual curtailment.50 Before annual curtailment, vel non, can justify a substantial shift in system-wide pipeline cost responsibility to the commodity component, the Commission must provide an explanation for its assertion that annual curtailment has diminished the vitality of the demand charge.
Merely citing to Consolidated without a discussion of the significant factual differences between the United and Texas Gas pipelines is inadequate to discharge the Commission’s obligation to explain its change.51 Rather, what is required is an inquiry by the Commission into the function served by the $3 million in fixed transmission costs that adoption of United formula *303would reallocate to Zone 4.52 The Commission’s assertion that gas supply now places limits on the operation of the Texas Gas pipeline restates the obvious. It leaves unanswered the petitioners’ contention that the relative importance of the pipeline’s demand and commodity functions has not changed in a way which would justify departure from the Seaboard formula.53 On remand, the Commission must set out findings of fact sufficient to enable a reviewing court to discern whether a rational relationship exists between the Commission’s reallocation of costs among the pipeline’s rate zones and the respective cost responsibilities presently incurred by the zones.
2. Storage
The Commission’s decision to adopt United for cost allocation (thus shifting $3 million in cost responsibility from Zones 1, 2, and 3 to Zone 4) is particularly troubling here since the end use profiles of customers in all four zones are substantially similar.54 The principal difference between Zone 4 and the other zones is that customers in Zone 4 (both industrial and city-gate) have invested substantial sums in storage, thus increasing their load factors.55 It is precisely because Zone 4 customers have invested in storage (an action encouraged by the Commission, which benefits all of the pipeline’s customers) that they are required to assume additional costs with the switch from Seaboard to United.56
The Commission responds that users who built storage have benefited from lower than average unit costs of gas, a benefit they will continue to receive under the United formula, only to a lesser degree,57 *304and that the claim of Columbia and Consolidated, to have benefited the system by not requiring more peak-day capacity, was “too speculative” to justify any formula other than United.58
We agree with the Commission that the existence of the Zone 4 storage facilities, alone, is insufficient to preclude the Commission from altering a pipeline’s cost allocation formula. However, as in Consolidated, we are concerned that the storage issue does not seem to have been given adequate consideration or analysis.59 Moreover, where other record evidence in support of the Commission’s action is deficient, equitable considerations, such as storage, loom larger. Thus, we cannot say that the Commission’s analysis represents reasoned decisionmaking given the record now before us. Section 4 of the Natural Gas Act does not permit the Commission to reduce the cost responsibilities of one geographical area at the expense of another, without benefit of substantial evidence showing that a change in the relative costs of providing service to the area has occurred60 or that a new and permissible policy objective would thereby be facilitated.61
B. Use of the United formula to design rates within each rate zone.
The Commission concluded that
[t]he United rate design method is appropriate for use on the Texas Gas system because it reflects a more equitable distribution of costs among its jurisdictional customers based upon the levels of curtailment now existing on the system.62
The Commission contends that our decision in Consolidated supports its use of United for rate design as well as for allocation.63 Our previous discussion of the factual differences between the Texas Gas and United pipelines, which rendered the application of Consolidated questionable for cost allocation in this case, applies with equal force to rate design and need not be repeated.64
Moreover, to the extent that the Texas Gas pipeline curtails volumes on an annual basis, it gives customers demand charge credits.65 Revenues equivalent to the amount of the demand charge credits then are recouped through a commodity surcharge. As a consequence, fixed costs are automatically “reclassified” from demand to commodity in direct relationship to the degree of annual curtailment on the system. The Commission eliminated a similar provision from the United pipeline tariff and its United decision was, in part, based on this fact.66 Here the Commission dismisses this difference by stating that
this further shifting of costs is consistent with the Commission’s goals of shifting costs to reflect curtailments on Texas *305Gas’ system and is not a reason [for] retaining Seaboard.67
The Commission’s statement simply acknowledges that use of the United formula for rate design, when coupled with a system of demand charge credits, results in recovery of more than 75% of the fixed storage and transmission costs through the commodity charge. It does not explain why the Commission found this result was just and reasonable under section 4 of the Natural Gas Act.68
In addition, the Commission offers two policy arguments in support of its decision to use the United formula for rate design. The Commission claims that use of the United formula will discourage industrial use of natural gas by reducing the “price discounts” high load factor customers receive under the Seaboard formula,69 and, secondarily, by narrowing the gap between the price of natural gas used for industrial purposes and that of alternative fuels.70 We recognize that the Commission “may . reconsider its approach even in the absence of any new evidence” if “the change in policy [is] avowed and reasoned.” 71 However, the record now before us fails to demonstrate that these objectives, which the Commission finds consonant with an era of natural gas shortages, will be achieved by use of the United formula for rate design on the Texas Gas pipeline.
Petitioners and respondent both agree that
[t]he objective of rate design is to determine rates that enable the pipeline to sell its gas at charges which recover from the various types of loads their reasonably associated costs of service, keeping in mind the factors pertinent in this regard.72
The parties also agree that the United formula will reduce the price differential between high load and low load factor customers in a given rate zone by shifting costs to the former. Columbia and Consolidated, however, object that the Commission has not demonstrated that this admitted shift in costs to high load factor customers is supported by a change in the relative cost of providing service to the higher load factor customers as compared to the lower load factor customers within each rate zone. Secondly, they reiterate that, contrary to the Commission’s belief, not all high load factor users are industrial users. City-gate customers, such as Louisville, that have built storage, also take gas at a high load factor.73 Thus, the record fails to demon*306strate that reducing the differential between high load factor and low load factor customers will discourage industrial sales.74 At best, application of United to Texas Gas will discourage high load factor sales, both industrial and city-gate.
We are also unable to find evidence in the record supporting the Commission’s contention that use of United for rate design will accomplish the Commission’s secondary objective of narrowing the gap between the price of natural gas and alternative fuels.75 Elsewhere the Commission has recognized that rate design is, at most, a “limited means” of “reducing] industrial gas use under present day circumstances,”76 because the prices of the alternative fuels greatly exceed the price of natural gas. Based on the record before us, it appears that use of the United formula for rate design, as for cost allocation, will simply shift costs in each zone from wholesale customers without storage to those with storage rather than reduce the price gap.
Judicial scrutiny under the National Gas Act is limited to assuring that the Commission’s decisionmaking is reasoned, principled, and based upon the record. For a reviewing court to perform this task, it is imperative that the Commission articulate the critical facts upon which it relies when it decides to reallocate fixed cost responsibility. Similarly, when the Commission finds it necessary to make predictions or extrapolations from the record, it must fully explain the assumptions it relied on to resolve unknowns and the public policies behind those assumptions.77 Where the Commission balances competing interests in arriving at its decision, it must explain on the record the policies which guide it. Only if the Commission observes these minimum standards can we be confident that missing facts, gross flaws in agency reasoning, and statutorily irrelevant or prohibited policy judgments will come to a reviewing court’s attention. Moreover, by requiring that the Commission fully articulate the basis for its decision, we assure the Commission, itself, the first opportunity to correct any defects which may emerge from such disclosure.
Since the Commission has failed to comply with these standards, we vacate the Commission’s order insofar as the order adopts the United formula for cost classification, allocation, and rate design, and remand the case to the Commission for further proceedings on these matters consistent with this opinion.
It is so ordered.
. The orders on review here were issued by the Federal Power Commission. Pursuant to the provisions of the Department of Energy Organization Act, Public Law 95-91, 91 Stat. 565 (1977) and Executive Order No. 12009, 42 Fed. Reg. 46267 (1977), the Federal Power Commission (FPC) ceased to exist on September 30, 1977. The Federal Energy Regulatory Commission (FERC) succeeded to the FPC’s ratemaking authority under the Federal Power Act and the Natural Gas Act on October 1, 1977, Id, § 402(a)(1)(B) & (C), 91 Stat. 583 (1977) (to be codified at 42 U.S.C. § 7172(a)(1)(B) & (Q).
Section 705(a) of the Act, 91 Stat. 606 (1977) (to be codified as 42 U.S.C. § 7295(a)) provides that all FPC orders in effect on October 1, 1977 “shall continue in effect . . . .” Section 705(e) of the Act, 91 Stat. 607 (1977) (to be codified as 42 U.S.C. § 7295(e)) provides that FERC shall be substituted for the FPC in suits such as this, which were pending when FERC was created.
References to “the Commission,” when used in the context of an action taken or statement made before October 1, 1977, refer to the Federal Power Commission; when used otherwise, such references refer to the Federal Energy Regulatory Commission.
. Opinion and Order Determining Proper Methods for Cost Classification, Cost Allocation, and Rate Design and Deciding Zone Boundaries Issue, 58 F.P.C. — (April 11, 1977), Joint Appendix (J.A.) 187-218 [hereinafter “Commission Opinion”].
. Opinion and Order Denying Rehearing, 58 F.P.C. — (June 7, 1977), J.A. 260 66.
. Texas Gas is an interstate natural gas transmission company subject to the Commission’s jurisdiction under the Natural Gas Act, 15 U.S.C. §§ 717 to 717w (1976). It operates an interstate pipeline system through which it transports and sells natural gas pursuant to certificates of public convenience and necessity issued by the Commission. Its pipeline facilities originate from gas fields in Texas and Louisiana, traverse Arkansas, Mississippi, Tennessee, Kentucky and Indiana, and terminate near Cincinnati, Ohio. It sells gas to 81 wholesale (or “jurisdictional”) customers in four separate rate zones. The rates for these sales are regulated by the Commission. In addition, Texas Gas makes a de minimus volume of direct sales to industrial customers at rates that are not subject to the Commission’s jurisdiction. Sales to these customers are referred to as “nonjurisdictional” sales. Among the pipeline’s customers located in its northernmost rate zone, Zone 4, are petitioners Columbia, Consolidated, and Louisville. Intervenor Memphis Light, Gas and Water Division is a customer in the southernmost rate zone, Zone 1. Texas Gas Brief at 4 -5; Commission Brief at 4; Columbia-Consolidated Joint Brief at 3.
. Section 19(b) of the Natural Gas Act, 15 U.S.C. § 717r(b) (1976), provides, inter alia, 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 United States Court of Appeals for the District of Columbia . . [S]uch court shall have jurisdiction, which . . shall be exclusive, to affirm, modify, or set aside such order in whole or in part.
. Columbia and Consolidated are wholly-owned subsidiaries of the Columbia Gas System, Inc. and Consolidated Natural Gas Company, respectively. They purchase approximately one-third of the gas sold by Texas Gas. Their purchases are made at a relatively high load factor and resold to their customers at relatively low load factors. Columbia-Consolidated Brief at 3-4. “Load factor” is defined infra at note 19.
. New York is a state regulatory body with jurisdiction over the sale of natural gas at retail in New York. Although the Texas Gas pipeline serves neither New York gas distribution companies or end consumers directly, it is a major supplier of gas to Columbia, Consolidated, and Texas Eastern Transmission Corporation, which, in turn, are major suppliers of gas to New York.
. Louisville is a major city gate distribution company in Zone 4 which purchases all of its natural gas supply from Texas Gas except for small emergency purchases. Louisville Brief at 4.
. Memphis, a division of the City of Memphis, is a city gate customer of Texas Gas, located in Zone 1. It purchases gas from the pipeline at a low load factor. During the test year, Memphis accounted for more than 70% of Zone l’s sales. Columbia-Consolidated Joint Brief at 16, n.13.
. 50 F.P.C. 1348 (1973), rehearing denied, 51 F.P.C. 1014 (1974), aff’d sub nom. Consolidated Gas Supply Corporation v. Federal Power Comm’n, 172 U.S.App.D.C. 162, 520 F.2d 1176 (1975) [hereinafter Consolidated].
. 11 F.P.C. 43 (1952).
. Three separate steps are involved in rate-making for a pipeline, such as Texas Gas, that is divided into geographical rate zones. The first step, cost classification for cost allocation, involves determining the pipeline’s total costs of rendering service and the proportion of these costs incurred by each of the pipeline’s rate zones. Within each zone, costs are then allocated between jurisdictional and non-jurisdictional customers. The latter step is relatively inconsequential in this case because non-jurisdictional sales represent less than 1% of the pipeline’s annual sales volume. (Commission Opinion at J.A. 199, 205.) Rate design, the third step, involves the formulation of rates to be charged customers in each rate zone. These rates enable the pipeline to recover from its customers in each rate zone the portion of total system costs allocated to that zone. See Texas Gas Brief at 5-6; Commission Brief at 7-14. A more detailed description of this process is presented in P. Garfield & W. Lovejoy, Public Utility Economics (1964) at 182-88 [hereinafter cited as Garfield and Lovejoy]. See also Consolidated, supra note 10, 172 U.S.App.D.C. at 165-66, 520 F.2d at 1179-80.
The Commission set a “demand-commodity” rate for gas sales to distributors by Texas Gas. This is a commonly used, two-part rate with which a pipeline recovers its service costs and a return on its investment through two separate changes, — a demand charge and a commodity charge. The demand component of the two-part rate is related primarily to the utility’s fixed costs. These are costs associated with a customer’s basic entitlement to receive gas and with the system’s maintenance of capacity sufficient to serve maximum (or “peak”) needs. *296Fixed costs include investment in pipeline facilities, taxes, depreciation. The commodity component relates more directly to the utility’s variable costs (e. g., the cost of the gas itself and the cost of compressor station fuel). These costs vary in relation to the volume of gas delivered to any given customer. See Commission Brief at 7-11; Aman & Howard, Natural Gas and Electric Utility Rate Reform: Taxation Through Ratemaking?, 28 Hastings L.J. 1084, 1121 (1977); Pierce, Natural Gas Rate Design: A Neglected Issue, 31 Vanderbilt L.Rev. 1089, 1094-97 (1978).
The principal issue in this case is whether the Commission acted properly in switching from the Seaboard to the United formula for apportioning fixed costs between the demand and commodity charges. The Seaboard formula takes it name from the Commission’s opinion in Atlantic Seaboard Corporation, supra note 11, wherein the Commission determined that 50% of the pipeline’s fixed transmission costs are to be recovered through the demand component of the pipeline’s two-part “demand-commodity” rate. The remaining 50% of the fixed costs and all of the variable costs are to be recovered through the commodity charge. The United formula takes name from the Commission’s opinion in United Gas Pipeline Company, supra note 10. Simply stated, the United formula provides that 25% of the fixed transmission costs are to be recovered through the demand charge, and the remaining 75% of fixed costs, together with all variable costs, are to be recovered through the commodity charge.
.Section 4 of the Natural Gas Act, 15 U.S.C. § 717c (1976), provides that:
(a) All rates and charges made, demanded, or received by a natural-gas company for or in connection with the transportation or sale of natural gas subject to the jurisdiction of the Commission, and all rules and regulations affecting or pertaining to such rates or charges, shall be just and reasonable, and any such rate or charge that is not just and reasonable is declared to be unlawful.
(b) No natural-gas company shall, with respect to any transportation or sale of natural gas subject to the jurisdiction of the Commission, (1) make or grant any undue preference or advantage to any person or subject any person to any undue prejudice or disadvantage, or (2) maintain any unreasonable difference in rates, charges, service, facilities, or in any other respect, either as between localities or as between classes of service.
Our review is pursuant to § 19(b) of the Act. See note 5 supra. In 1968, the Supreme Court described the nature of a reviewing court’s function under § 19(b):
First, [the reviewing court] must determine whether the Commission’s order, viewed in light of the relevant facts and of the Commission’s broad regulatory duties, abused or exceeded its authority. Second, the court must examine the manner in which the Commission has employed the methods of regulation which it has itself selected, and must decide whether each of the order’s essential elements is supported by substantial evidence. Third, the court must determine whether the order may reasonably be expected to maintain financial integrity, attract necessary capital, and fairly compensate investors for the risks they have assumed, and yet provide appropriate protection to the relevant public interests, both existing and foreseeable. The court’s responsibility is not to supplant the Commission’s balance of these interests with one more nearly to its liking, but instead to assure itself that the Commission has given reasoned consideration to each of the pertinent factors. Judicial review of the Commission’s orders will therefore function accurately and efficaciously only if the Commission indicates fully and carefully the methods by which, and the purposes for which, it has chosen to act, as well as its assessment of the consequences of its orders for the character and future development of the industry.
Permian Basin Area Rate Cases, 390 U.S. 747, 791-92, 88 S.Ct. 1344, 1373, 20 L.Ed.2d 312 (1968).
. F.P.C. Docket No. RP75-19.
. The Commission Staff’s cost classification is shown in Exhibit 17 and Schedule 3 of Exhibit 18. J.A. 83. The Staffs cost allocation, using the Seaboard formula without modification, is shown in Exhibit 18. J.A. 83-86; 118-20. Exhibit 19 shows the rate designs proposed by Staff and Texas Gas. J.A. 83-86.
. Under section 4(e) of the Natural Gas Act, 15 U.S.C. § 717c(e) (1976), the Commission may suspend a filing for up to five months pending a hearing on its lawfulness.
. Although the five month suspension ended before the hearings were held, the parties stipulated in the settlement agreement filed on June 12, 1975, and approved by the Commission on October 6, 1975, that the Commission’s decision on the reserved issues would become effective only after the decision became final and nonappealable. Record (R.) 759. The Commission took note of this stipulation in its opinion. It indicated that Texas Gas was expected “in any new general rate increase filing . to reflect fully the United methodology” or subject itself to additional payment for undercollections not to exceed the amount which would have been collected under the United formula. Commission Opinion at J.A. 216-17.
The ALJ also heard the issue of whether the rate zone boundaries along the Texas Gas pipeline should be changed. He subsequently ruled that no grounds existed to support any change in rate zones. Presiding Administrative Law Judge’s Initial Decision on Reserved Issues Upon Settlement of Rate Proceeding, Docket No. RP75-19 (May 11, 1976) [hereinafter “ALJ Decision”], J.A. 115, 150-51. No party filed an exception to this ruling, nor did the Commission, on its own motion, disturb it. Therefore, it is not now before this court.
. Consolidated Gas Supply Corporation v. Federal Power Comm'n, 172 U.S.App.D.C. 162, 520 F.2d 1176 (1975).
. A customer’s annual “load factor” is the percentage relationship of its average daily demand to its maximum daily demand. Thus, high load factor customers tend to take a relatively steady supply of gas throughout the year, either because they have constructed storage facilities or because they serve industrial customers not subject to weather-related fluctuations characteristic of residential consumers. Low load factor customers tend to have needs that fluctuate considerably. These are often “city gate” customers, typically local distribution companies that do not have storage facilities, that take gas at the “city gate” for sale to commercial and residential users. See Garfield & Lovejoy, supra note 12, at 171-72, 175.
. Although Texas Gas used the United formula in its rate design filing, the ALJ states that “[Texas Gas] refrains from arguing in support thereof and notes that there is much to be said in favor of using the unmodified Seaboard formula [for rate design] on the Texas Gas System.” ALJ Decision, J.A. at 136-37.
. ALJ Decision at J.A. 130-31; 151-52.
. Id at J.A. 130-31.
. Id. at J.A. 140.
. Commission Opinion at J.A. 200; see also Commission Brief at 13.
. Opinion and Order Denying Rehearing, Opinion No. 792-A (June 7, 1977), J.A. 260, 265.
. Id. at J.A. 264.
. Alabama-Tennessee Natural Gas Co. v. FPC, 359 F.2d 318, 331 (5th Cir.) cert. denied, 385 U.S. 847, 87 S.Ct. 69, 17 L.Ed.2d 78 (1966).
. Consolidated, supra note 10, 172 U.S.App.D.C. at 171, 520 F.2d at 1185 quoting F.P.C. v. Natural Gas Pipeline Co., 315 U.S. 575, 586, 62 S.Ct. 736, 86 L.Ed. 1037 (1942).
. American Louisiana Pipe Line Company, 29 F.P.C. 932, 940 (1963), rev’d on other grounds, 120 U.S.App.D.C. 140, 344 F.2d 525 (1965). Although “Seaboard is not an inexorably fixed rule,” the Commission has acknowledged previously that any “departure from [Seaboard] should be permitted only upon a substantial showing ” because
the Seaboard classification of costs protects the domestic consumer from excessive fixed costs, affords the commission a fair and defi*299nite point of reference in regulating pipeline rates, and provides the pipelines a known and consistent standard for developing their rates and markets.
Id. (emphasis added).
. 15 U.S.C. § 717c(a) (rates must be just and reasonable); 15 U.S.C. § 717c(b) (no natural gas company shall grant any undue preference or maintain any unreasonable difference in rates between localities or classes of service).
. We disagree with the Commission that its action can be sustained simply because Texas Gas, Columbia-Consolidated, and New York did not present adequate cost studies. Commission Opinion at J.A. 214; Commission Brief at 40. See also J.A. at 253-54. Decisions of both the courts of appeals and the Commission have held that the Commission bears the burden of explaining the reasonableness of any departure from a long standing practice, and any facts underlying its explanation must be supported by substantial evidence. See Consumers Union v. Federal Power Commission, 166 U.S.App.D.C. 276, 279, 510 F.2d 656, 659 (1974); Columbia LNG Corporation v. Federal Power Commission, 491 F.2d 651, 654 (5th Cir. 1974); United Gas Pipe Line Company, 25 F.P.C. 26, 30-32 (1961). If, as it appears (see Commission Opinion at J.A. 202, 211; Commission Brief at 37 -40), the Commission has decided to abandon Seaboard because the application of that formula to Texas Gas no longer generates rates that are just and reasonable within the meaning of § 4 of the Natural Gas Act, the Commission must demonstrate that its finding is supported by substantial evidence. 15 U.S.C. § 717r(b).
. The Commission’s Opinion is in accord. See Commission Opinion at J.A. 199. Louisville also stresses this point in its petition for rehearing before the Commission. J.A. at 221, 228-29. See also Consolidated, supra note 10, 172 U.S.App.D.C. at 165-66, 520 F.2d at 1179-80; Garfield and Lovejoy at 187-89.
. See note 12 supra, and Commission Opinion at J.A. 198- 99.
. See note 12 supra, and Commission Opinion at J.A. 199, 209 -17.
. We are not concerned with the effect of the Commission’s decision on Texas Gas’ direct, non-jurisdictional sales, since these sales account for less than one percent of the test year deliveries. See note 12 supra. This was not true, however, in Consolidated, where almost one third of the test year sales were made to non-jurisdictional customers. As a consequence of adopting the United formula in Consolidated for cost allocation and rate design, a substantial amount of cost responsibility was shifted to non-jurisdictional (largely industrial) customers and away from United’s wholesale customers. United Gas Pipeline Company, supra note 10, 50 F.P.C. at 1355-56; 1365-67; Consolidated, supra note 10, 172 U.S.App.D.C. at 168, 520 F.2d at 1182. See also J.A. 167.
. Commission Opinion at J.A. 199. Louisville, for example, states that the change to United allocates to it, on an annual basis, $409,000 in additional cost responsibility. J.A. 220. The cost shift to Zone 4, of course, increases the costs not only of high load factor customers, such as Louisville, but also of the zone’s thirteen relatively small, low load factor customers. See also note 73 infra and accompanying text.
. Commission Opinion at J.A. 199.
. Commission Opinion at J.A. 198. The Commission's discussion of cost classification and allocation is set out at J.A. 198-206.
. Commission Brief at 20. See also the Commission Opinion at J.A. 200-01. The era of pipeline expansion referred to by the Commission is described in Garfield and Lovejoy at 167- 170.
. Consolidated Gas Supply Corporation v. Federal Power Comm'n, 172 U.S.App.D.C. 162, 520 F.2d 1176 (1975).
. Consolidated involved the use of the United formula for cost classification, allocation and rate design. See Consolidated, supra note 10, 172 U.S.App.D.C. at 165, 520 F.2d at 1179.
. The term “curtailment" as used by the Commission means the difference between the amount of gas that an interstate pipeline is required by contract or certificate to deliver (either on a daily or an annual basis) and the amount it is actually able to deliver.
Curtailments first occurred in 1970-71 and have increased steadily each year since, reaching 3.4 trillion cubic feet (or about 23% of the combined contractual obligations of all pipelines) in 1976-77. See S. Herman, R. Pierce, M. Tropin and B. Tyree, Natural Gas Users’ Handbook 3 (1976) and Federal Power Commission, Requirements, Curtailments, and Deliveries of Interstate Pipeline Companies Based on Form 16 Reports Required to be Filed on April 30, 1977 (1977).
. See, for example, the ALJ’s presentation of data in his Initial Decision in United, supra note 10, 50 F.P.C. at 1380-82 (1973). At the time of our decision, United had been affected by the natural gas shortage more than any other pipeline. Id. at 1375.
. Consolidated, supra note 10, 172 U.S.App.D.C. at 172, 520 F.2d at 1186.
. According to the Commission, “the record indicated that there is not severe underutilization ... on the peak-day on the Texas Gas system,” although “there is quite substantial underutilization of Texas Gas’ system on an annual basis.” Commission Opinion at J.A. 201. See also Commission Brief at 29.
Texas Gas, Columbia, and Consolidated point out that the underutilization referred to by the Commission results from reduced demand during years in which weather conditions are warmer than they were during the 1974 test period or because the minimum billing provisions of the Texas Gas tariff during the test *301period induced customers to avoid takes in excess of 95 percent of contract demand. Texas Gas Brief at 15, n.2; Columbia-Consolidated Joint Brief at 9. The Commission accepted the ALJ’s determination that Texas Gas had not curtailed peak period service in any respect:
The [Administrative Law] Judge cites with approval the arguments of Texas Gas and others that the annual peak day usage is still significant on the Texas Gas system because Texas Gas is able to meet all of its customers’ demands for gas (including those of Memphis) on that day, in spite of curtailments on an annual basis on the system.
Commission Decision at J.A. 195-96. See also J.A. 42-44.
The Texas Gas FPC Tariff, Third Revised Volume No. 1 indicates that the design capacity of the system is 2,656,682 Mcf. Commission witness Chinn presented the average three-day peak use to be 2,427,080 Mcf. Texas Gas indicated a comparable figure of 2,461,104 Mcf. for the average of January 12-14, 1975 and 2,471,-106 Mcf. for the average of January 7-9, 1976. Commission Opinion at J.A. 202, n.ll.
. Commission Brief at 32. Annual curtailment on the Texas Gas system, furthermore, is substantially less than the annual curtailment which United experienced.
In its brief, the Commission cites various cases, none of which have undergone judicial review, in which it claims it “has determined that the United method for cost classification and rate design is proper for pipelines experiencing curtailment.” Commission Brief at 27-28. In particular, the Commission refers us to its opinions in Natural Gas, Pipeline of America, Opinion No. 782, 56 F.P.C.—(1976), rehearing denied, Opinion 782-A, 57 F.P.C.—(1977), in which annual but not peak-day curtailment was found. Commission Brief at 28-29.
We have searched these cases for some indication of the basis on which the Commission apparently has concluded that pipelines experiencing only annual curtailment should be treated in the same manner as those experiencing annual and peak-day curtailment. We find none. In Natural, as in this case, the Commission simply relies on our opinion in Consolidated to conclude “as a matter of judgment” that the United formula should be adopted. Natural, supra at 56 F.P.C.-, slip. op. at 16. Nor is any additional discussion provided in the Commission’s opinion denying rehearing. Natural, supra at 57 F.P.C.—, slip. op. at 5.
. Our recent comments in Richmond Power & Light Co. v. FERC, 187 U.S.App.D.C. 399, 574 F.2d 610 (1978), about the desirability of a thorough discussion of the economic principles on which the Commission relies, bear repeating here. In Richmond we said:
We are not entirely pleased with the Commission’s discussion of the economic principles operable in this situation. . . . Irrespective of the direction in which the Commission ultimately heads, it could travel a smoother road if it would lay out the competing principles of ratemaking, explain why it accepts particular theories and rejects others, and then elucidate how the principles adopted support the specific allocation of costs. Such an explication would not only aid our review but, we believe, would lead to better informed and better reasoned decisions by the agency itself.
187 U.S.App.D.C. at 410, 574 F.2d at 621 n.43.
Neither the Commission’s opinion nor its brief makes any effort to take into account the substantial body of scholarly analysis of natural gas ratemaking. See, for example, Garfield and Lovejoy, supra note 12; Aman and Howard, Natural Gas and Electric Utility Rate Reform: Taxation Through Ratemaking? 28 Hastings L.Rev. 1084 (1977); Lome, Natural Gas Pipelines, Peak Load Pricing and the Federal Power Commission, 1972 Duke L.J. 85 (1972) and the citations therein at nn.l & 2; Pierce, Natural Gas Rate Design: A Neglected Issue, 31 Vanderbilt Law Review 1089 (1978); and, Note, 57 North Carolina L.Rev. 287 (1979). As a consequence, one commentator has observed that
the manner in which the economic literature has been ignored by those who deal closely with the FPC is surprising. As a result, there are now two entirely separate bodies of literature, one dealing with economic theory and the other with the FPC regulation of gas pipeline rates. The interaction between the two, although necessary, has thus far been nonexistent.
Lorne, supra at 90.
. In United, the Commission concluded “that greater emphasis must be placed upon the annual use of United’s pipeline system,” but offered no explanation of how annual curtailment affected its conclusion. United, supra note 10, 50 F.P.C. at 1361 (emphasis added).
. See, e. g„ J.A. 41-44; 140.
. The Commission has not satisfied its responsibility under the Natural Gas Act by conclusorily asserting that
[it] has not overemphasized the annual use of the [Texas Gas] system but rather has made a proper determination based on current operations of the Texas Gas system under curtailment conditions.
J.A. 213-14. Where, as here, the Commission has neither specified the evidence on which it relied nor explained how that evidence supports the conclusion it reached, a reviewing court cannot uphold its orders.
. In questioning the applicability here of the Consolidated rationale, We in no way intimate that the Commission must or should adopt the Seaboard formula nor that United is necessarily inappropriate for the Texas Gas pipeline. All parties acknowledge that the relative importance of the demand and commodity charges under Seaboard was not measured with scientific accuracy. Moreover, we recognize that the Seaboard formula has been widely criticized. One commentator, for example, notes that
[t]he most disturbing aspect of Atlantic Seaboard is the arbitrary fifty-fifty allocation of system-wide fixed costs between the demand charge and the commodity charge. That allocation has no justification other than the vague notion that ‘all gas transported by the pipeline will share in all of the various kinds of expenses incurred to transport the gas.’ There is no reason to suspect a connection between the amount of the expense that ‘should’ be borne by a class of consumers and the amount that is allocated to them by the fifty percent approach.
Lorne, supra note 47 at 99 (citation omitted). See also Garfield & Lovejoy at 182-83 (suggesting that even the fundamental determination of which costs are fixed and which are variable was inappropriately made in Seaboard). The Commission has departed from strict application of the Seaboard formula on numerous occasions. During the 1960’s, the Commission frequently “tilted” the Seaboard costs away from the commodity component to make the price of natural gas low enough to be competitive with alternative fuels. See Fuels Research Council, Inc. v. FPC, 374 F.2d 842, 852 (7th Cir. 1967).
On remand, the Commission is free to consider further the economic effects of annual curtailment and, if it chooses, any alternative grounds which might exist for shifting fixed costs to the commodity category. Here the Commission’s failure to address adequately the economic implications of the facts presented by the record has resulted in an order that lacks the reasoned explanation and factual support required for a reviewing court to uphold the Commission’s action. We have previously noted that “[w]hat is basic [on review] is the requirement that there be support in the public record for what is done.” The Second National Natural Gas Rate Cases, 567 F.2d 1016, 1029 (1977), citing City of Chicago v. FPC, 458 F.2d 731 (D.C.Cir. 1971), cert. denied, 405 U.S. 1074, 92 S.Ct. 1495, 31 L.Ed.2d 808 (1972).
. See, Consolidated, supra note 10, 172 U.S.App.D.C. at 172 73, 520 F.2d at 1186-87.
. See, for example, Texas Gas Petition for Rehearing, J.A. 226.
.Texas Gas witness Joseph L. Benson, for example, testified that
[t]he end-use profiles of [distributors who invested in storage] are not materially different from the end-use profiles of distributors without storage facilities because of the reduction in industrial sales through the operation of curtailment plans.
J.A. 68- 69.
.See Commission Opinion at J.A. 204-06; 240 -42. Customers with storage facilities are able to take gas at a relatively constant rate, thereby developing high load factors, even though many of their customers have needs that fluctuate considerably. See note 19 supra.
The ALJ Decision discloses the following levels of investment by the Zone 4 petitioners as of December 31, 1974:
Storage Plant Investment Gas Stored Investment
Columbia $206,333,154 $106,205,520
Consolidated 146,801,149 90,416,425
Louisville 20,671,000 10,933,000
J.A. 131.
.The Commission appears to ignore this important fact. Instead the Commission incorrectly equates high load factor customers with industrial customers. See, e. g., Commission Opinion at J.A. 206: “[B]oth United and Texas Gas serve their pipeline customers in their northernmost zones at high load factors whereas their respective lower load factor city-gate customers are served in their southern zones.” See also Commission Opinion at J.A. 206-08; 211.
Columbia and Consolidated assert that the distribution companies which purchase gas in Zones 1, 2, and 3 actually sell relatively more gas for (low-priority) industrial use than do the major distributors in Zone 4 served by Columbia and Consolidated. Columbia-Consolidated Joint Reply Brief at 12. Similarly, with United cost allocation, many low-load factor residential users in Zone 4 have costs shifted to them from the three other zones simply because their city-gate distributor (Louisville) built storage to achieve a higher load factor (and thereby provide better service), whereas city-gate distributors in Zones 1, 2, and 3 (such as Memphis) did not.
The Commission’s staff witness, Mr. Chinn, acknowledged this result at the administrative hearing in testifying that the United formula was appropriate for rate design but not for cost allocation because the United formula
would allocate less costs to Zones 1, 2 and 3, and over $2 million more cost to Zone 4, than the Seaboard method. I consider these results unreasonable because four customers in Zone 4 purchasing 93% of the gas have extensive storage operations and that method would penalize them. In view of the continuous use of the Seaboard method on the system, I recommend that the method be maintained. However, in the design of rates I have departed from allocated costs to reflect a greater commodity value of the gas.
J.A. 251 (emphasis added).
.J.A. 204-05.
. J.A. 205.
. See Consolidated, supra note 10, 172 U.S.App.D.C. at 174, 520 F.2d at 1188. We observed there that “ ‘the agency [must] give due consideration to the equities, if any, arising out of commitments based on previous rulings’," id., 172 U.S.App.D.C. at 173, 520 F.2d at 1187 (citing City of Chicago v. FPC, 128 U.S.App.D.C. 107, 115, 385 F.2d 629, 637 (1967).
. 15 U.S.C. 717c(b) (1976); see also Northern Natural Gas Company, 14 F.P.C. 11 (1955), aff’d sub nom. Interstate Power Company v. F. P. C., 236 F.2d 372 (8th Cir. 1956).
. In justifying its choice of United for cost allocation, the Commission disclaims reliance on its policy goals of discouraging industrial usage of natural gas and of narrowing the gap between the price of natural gas used for industrial purposes and alternative fuels. See J.A. 207. It discusses these goals in the context of rate design. See J.A. at 211 16 and notes 69-76 infra and accompanying text.
. Commission Opinion at J.A. 209.
In its brief, the Commission states that its Texas Gas decision simply implements “its recent policy of adopting the United method for pipelines experiencing curtailment.” Commission Brief at 15 (emphasis added).
. Commission Opinion at J.A. 209 10.
. See notes 40-53 supra and accompanying text.
. Commission Opinion at J.A. 196. The record also shows that low priority users are curtailed on a seasonal basis through a priority system. J.A. 139.
. Supra note 10, 50 F.P.C. at 1367 68.
. Commission Opinion at J.A. 215.
. In American Louisiana Pipeline, supra note 29, 29 F.P.C. at 932, the Commission noted that the Seaboard formula is “but one factor in reaching a determination on [the question of rate design].” On remand, the Commission must thoroughly delineate its consideration of all of the factors it relies on to justify the rate design formula it adopts. See Public Service Commission v. F. P. C., 167 U.S.App.D.C. 100, 107, 511 F.2d 338, 345 (1975). Anything less falls short of “its duty to indicate ‘fully and carefully the methods by which, and the purposes for which, it has chosen to act, as well as the consequences of its orders for the character and future development of the industry.’ ” Id. 167 U.S.App.D.C. at 108, 511 F.2d at 346, quoting Permian Basin Area Rate Cases, supra note 13, 390 U.S. at 792, 88 S.Ct. 1344.
. J.A. 210-11, J.A. 264.
. J.A. 207, 215-16.
The Commission does not rely on these objectives as an alternative justification on which to rest its decision. J.A. 207, 215. Thus, we need not decide in this case whether the pricing of natural gas to discourage industrial use is unlawful. See Federal Power Commission v. Hope Natural Gas Co., 320 U.S. 591, 615-16, 64 S.Ct. 281, 88 L.Ed. 333 (1944).
. The Second National Natural Gas Rate Cases, supra note 51, 186 U.S.App.D.C. at 38, 567 F.2d at 1031.
. American Louisiana Pipe Line Company, supra note 29, 29 F.P.C. at 943.
. We also agree with petitioners that the record does not demonstrate that the correlation between high load factor users and high volume users or between high load factor and low priority industrial users is significantly different intra-zone than it is inter-zone. See our discussion supra at notes 54-61 and accompanying text. In the absence of such a showing, we fail to see how use of the United formula for rate design will further the Commission’s avowed policies.
It appears from the record that United rate design would tend to shift costs in Zone 4 from *306that zone’s thirteen low load factor customers to all high load factor customers in the zone, including Louisville as well as Columbia, and Consolidated. In contrast, the Commission decision actually results in a decrease in the commodity rates for Zones 1, 2, and 3. Thus the effect of applying United to cost allocation and rate design in Zones 1-3 is actually antagonistic to the Commission’s stated goals. See Exhibits 22 and 24, J.A. 91-102.
. Since the volume of non-jurisdictional gas sales on the Texas Gas pipeline is de minimus, see note 35 supra, the United rate design formula also cannot be justified on the ground that it shifts cost responsibility to non-jurisdictional, industrial customers.
. ALJ Decision at J.A. 140.
. Proposed Rulemaking, Docket No. RM75-19 (February 20, 1975), p. 3, J.A. 170.
In its decision in Natural Gas Pipeline, supra note 46, in which it used the United formula for cost allocation, the Commission produced some evidence showing that the United formula tended to reduce the gap between the price of natural gas used for industrial purposes and alternative fuels. But, unlike this case, the shift to the United formula on the Natural Gas pipeline did not result in a differential impact on various customer classes across rate zones.
. Industrial Union Department, AFL-CIO v. Hodgson, 162 U.S.App.D.C. 331, 339-40, 499 F.2d 467, 475-76 (1974).
This task is impeded, where, as here, the Commission’s explanations do not clearly set out the linkage between the record evidence and the policy determinations drawn from that evidence.