Opinion for the Court filed by
Circuit Judge WILLIAMS. Opinion concurring in the judgment in part and dissenting in part filed by Circuit Judge MIKVA. WILLIAMS, Circuit Judge:TABLE OF CONTENTS
Page
I. Ixtrodiction...................................................... 13
A. Parties.......................................................... 14
B. Regulatory and Economic Context................. 15
II. Open-Access Requrements................................... 1“
A.Claimed Deficiencies in Statutory Authority to. Require Open Access.................................. 17
1. Natural Gas Act...................................... 17
2. Natural Gas Policy Act............................ 21
B. Alleged Failure to Comply with Mandate of Outer Continental Shelf Lands Act...................
C. Claims of Arbitrariness and Caprice...............
1. Failure to impose the nondiscriminatory access conditions on § 7(c) transportation certificates............................................
2. Capacity allocation on a “first-come, first served” basis........................................ 25
III. Rate Conditions.................................................. 27
A. Absence of Finding that Prior Rates Were Unlawful........................................................ 28
B. Allowance of Discounting Generally................ 29
C. Potential Discrimination Between Bundled and Unbundled Transportation........................... 29
D. Selective as Opposed to Uniform Discounts...... 30
E. Consistency of “Value-of-Service” Discounting with MPC II.............................................. 30
F. Impact of Discounting on Pipeline Solvency..... 32
IV. Contract Demand (“CD") Adjustment..................' 33
A. Legal Authority............................................ 34
1. Violation of Panhandle doctrine.............. 34
2. Alleged lack of compliance with § 7(b)..... 35
B. Adequacy of the Commission’s Reasoning in Support of CD Conversion.......................... 36
C. Adequacy of the Commission’s Reasoning in Support of CD Reduction............................• 38
D. Insufficiency Under MPC III.......................... 40
V.Producer-Pipeline Contracts................................. 41
A. The Commission’s Prior Activity and Its Inactivity in This Proceeding................................. 41
B. Analysis of FERC’s Decision.......................... 43
1. Lack of need for additional steps..............• 43
a. Likely effects of Order No. 436 on take- or-pay build-up............................... 43
b. Pipeline ability to shift costs downstream ........................................... 45
2. Policy considerations militating against any intervention..........................................
3. Reasons for rejection of specific proposals,
a. Section 5 action against jurisdictional contracts........................................ 47
b. Conditioning producer access.............. 48
Page
C.Conclusion..................................................... 50
VI. Optional Expedited Certification........................ 50
A. Ripeness....................................................... 51
B. Legality of the Presumption........................... 53
1. Alleged disregard of legally relevant factors 54
2. Unsupported assumptions.......................... 57
3. Alleged failures of reasoned decisionmaking 58
VII. Miscellaneous Arguments.................................... 58
A. Pipeline Sales Service Issues.......................... 58
B. Notice Provisions........................................... 59
C. Transportation Policies...................................• 59
D. Construction of Facilities for Use in § 311 Transportation.............................................* 60
E. Grandfathering Decisions................................ 60
F. System Supply Limitation...............................• 62
G. Transportation for Fuel-Switchable End Users . 63
VIII. Conclusion......................................................... 64
I. Introduction
On October 9, 1985 the Federal Energy Regulatory Commission (“FERC”1) issued Order No. 436, 50 Fed.Reg. 42,408 (1985) (codified at scattered sections of 18 C.F.R.). The Order envisages a complete restructuring of the natural gas industry. It may well come to rank with the three great regulatory milestones of the industry: the passage of the Natural Gas Act, 15 U.S.C. §§ 717 et seq. (1982) (“NGA”) in 1938, the imposition of price controls on independent producers’ wellhead sales under Phillips Petroleum Co. v. Wisconsin, 347 U.S. 672, 74 S.Ct. 794, 98 L.Ed. 1035 (1954), and adoption of the Natural Gas Policy Act (“NGPA”), 15 U.S.C. §§ 3301 et seq. (1982) in 1978.
At stake is the role of interstate natural gas pipelines. Although they are obviously transporters of gas, they have until recently operated primarily as gas merchants. They buy gas from producers at the wellhead and resell it, mainly to local distribution companies (“LDCs”) but also to relatively large end users. The Commission has concluded that a prevailing pipeline practice — particularly their general refusal to transport gas for third parties where to do so would displace their own sales (Joint Appendix (“J.A.”) 318-19) — -has caused serious market distortions. It has found this practice “unduly discriminatory” within the *14meaning of § 5 of the NGA. Order No. 436 is its response.
The essence of Order No. 436 is a tendency, in the industry metaphor, to “unbundle” the pipelines’ transportation and merchant roles. If it is effective, the pipelines will transport the gas with which their own sales compete; competition from other gas sellers (producers or traders) will give consumers the benefit of a competitive wellhead market.
Virtually every sector of the natural gas industry has challenged the Order, asserting a dazzling array of errors and omissions. They have filed some 85 briefs totaling about 2,000 pages. Oral argument spanned two days in a well-filled courtroom.2 We here uphold most elements of the Order, but remand the record to the Commission on certain issues.
A. Parties.
The bulk of the petitions for review are filed by representatives of the main actors in the industry: (1) producers, (2) pipelines, (3) LDCs and (4) end users. A word follows as to the varieties and regulatory status of each, and as to the developments underlying the Commission’s decision.
Producers are primarily independents, in the sense of being unaffiliated with any pipeline. Pipelines and pipeline affiliates produce only about 11% of the gas sold by pipelines in the interstate market. Interstate Natural Gas Ass’n of America, “The Gas Contracts Problem: Results of an IN-GAA Survey” 13 (Policy Analysis 83P-1, May 1983) (cited at J.A. 301 n. 34; data for 1982). The producers typically operate under oil-and-gas leases with owners of the land or mineral interests, subject to a duty to pay royalty and at some risk of losing their leases if production ceases.
Producers’ interstate wellhead sales have, through the operation of law or economics, achieved virtually complete release from binding price controls. The NGPA provided this almost immediately for certain categories of “high cost” gas. NGPA § 121(b), 15 U.S.C. § 3331(b) (1982). For most “new” gas, the NGPA established new ceilings, higher than those previously set by the Commission, and provided for gradual increases until scheduled deregulation on January 1, 1985 or July 1, 1987. NGPA §§ 102, 103, 121(a) & (c), 15 U.S.C. §§ 3312, 3313, 3331(a) & (c) (1982). In fact, current market prices at the wellhead are significantly below the ceilings remaining in these categories. Compare Selected National Average Natural Gas Prices, Natural Gas Monthly 10 (Jan. 1987) (average price at wellhead for March 1986 $2.16) with Natural Gas Ceiling Prices by Category of Gas, Type of Sale, or Contract, Natural Gas Monthly 30 (Jan. 1987) (ceiling price for § 102 and § 103 gas for same period ranged from $3,083 to $4,216). The NGPA made no provision for deregulation of “old” interstate gas, but, besides allowing escalation to reflect general price inflation, authorized the Commission to raise the former ceilings to higher “just and reasonable” levels. (The Commission actually exercised this authority on June 6, 1986, in Order No. 451, III FERC Statutes & Regulations ¶ 30,701, at 30,197 (1986).)
The pipelines are either intrastate or interstate. Since enactment of the NGA, the interstate pipelines have been subject to pervasive Commission regulation. Before performing any significant act — construction of new facilities or initiation of new transportation service or new sales for resale — they must secure a certificate of convenience and necessity from the Commission. NGA § 7(c), 15 U.S.C. § 717f(c) (1982). They also require Commission approval when they abandon any “certificated” facility, transportation or sale. NGA § 7(b), 15 U.S.C. § 717f(b) (1982). Finally, the Commission limits to “just and reasonable” levels the prices at which the interstates sell gas for resale or provide transportation service. NGA § 4(a), 15 U.S.C. § 717c(a) (1982). Under conventional public utility principles this allows the pipelines, at least in theory, to recover no more *15than cost of service, including a reasonable return on investment.
LDCs purchase gas for resale to end users, large and small. Their services and prices are subject to state regulation but not to that of FERC.
End users run the gamut both in size and in ability to use substitutes. At one end is the ordinary householder, who even in the intermediate run is limited to such expedients as installing better insulation, wearing more sweaters, or turning the thermostat down. At the other end of the spectrum are users that need only flick a switch to replace gas with oil.
B. Regulatory and Economic Context.
The Natural Gas Act has the fundamental purpose of protecting interstate gas consumers from pipelines’ monopoly power. See Sen. Doc. No. 92, part 84A, 70th Cong., 1st Sess. 588-91 (FTC Utility Corporations Rep. 1935). By the early 1980s, a number of developments suggested, for the first time since enactment, that assuring customer access to the wellhead market could be an important potential ingredient in Commission fulfillment of that goal. First, by then a nationwide pipeline network had matured. J.A. 279, 284-92. This made it physically possible, for the first time, for consumers to acquire gas supplies from virtually any region. It also ended, or at least sharply reduced, pipeline monopsony power over wellhead purchases, a power that had the tendency to keep wellhead prices below competitive levels. See S. Breyer, Regulation and its Reform 243 (1982); cf Note, Federal Price Control of Natural Gas Sold to Interstate Pipelines, 59 Yale L.J. 1468, 1478-79 (1950).
Second, the removal of wellhead price controls greatly increased an underlying risk of the regulatory system — that pipelines’ gas purchase costs would rise above competitive market levels. For while cost-based price regulation at least in principle prevented pipelines from enjoying monopoly profits, the combination of market power and regulation tended to blunt their incentives for careful gas purchasing. The pipelines’ market power dampened any fear of being unable to recover exorbitant costs; regulation dampened any hope of direct profit from thrifty purchases, as it prevented them from making any mark-ups not based on cost. Cf. Pierce, Reconsidering the Roles of Regulation and Competition in the Natural Gas Industry, 97 Harv.L. Rev. 345, 357-65 (1983). As it proved, this lulling effect of regulation — coupled with the gyrations of the energy market from 1973 to the present — brought on the phenomenon of embedded contract prices well above competitive levels at the wellhead. For example, the Commission has recently estimated average prices paid by pipelines at the wellhead at about $2.50 per thousand cubic feet (“Mcf”), compared with less than $2 per Mcf in the spot market. See Order No. 451, III FERC Statutes & Regulations 11 30,701, at 30,210 (1986). (If the volume of the interstate market is about seven trillion cubic feet (“Tcf”), this would amount to a $3.5 billion price discrepancy.) The Commission’s estimate of the price specified in long-term new contracts as of mid-1986 was about the same as the average price then being paid by the pipelines (about $2.50), see id.; but the pipelines attained their figure for average price actually paid in part by refusing to take high-priced gas and thus subjecting themselves to a build-up of very substantial take-or-pay liability (discussed immediately below).
Third, the conditions under which the NGPA began to relax wellhead price controls — namely acute gas shortage and sharply rising prices for alternative fuels— tended to divert pipeline attention from the hazards of incurring long-term obligations to buy high-priced gas. Under pressure from the Commission, the pipelines had typically purchased gas under contracts for very long terms. See, e.g., 18 C.F.R. § 2.61 (requiring pipelines to maintain supply reserves of up to 12-years’ projected demand); Columbia Gas Transmission Corp., 21 F.E.R.C. 1161,026, at 61,160-61 (1982) (example of long-term supply arrangements pipeline must make in order to extend new service); Department of Energy, The First Report Required by Section 123 of the Natural Gas Policy Act of 1978 3-2 (1984) (producer-pipeline contracts for *1610 or more years are common). Besides incorporating high prices (and provisions for escalation upward), the contracts commonly included “take-or-pay” provisions, requiring the pipeline to pay for some specified percentage, say 75%, of the deliverable gas even if it took less. While usually subject to recoupment later, and while a perfectly natural allocation of risk between producer and purchaser, the take-or-pay provisions effectively committed the pipelines to high gas costs in what by 1982 proved to be a time of falling prices, both for competing fuels and for substitute supplies of gas not covered by contract.
Fourth, various economic forces, including exhaustion of the cheaper supplies and the decline of pipeline monopsony at the wellhead, see S. Breyer, Regulation and Its Reform 243 (1982), raised the wellhead price — and thereby the potential loss to consumers if they should be saddled with the results of pipelines’ readiness to bid high prices. After hovering in the range of $.50 (in constant 1984 dollars) per million British thermal units (“Btu”) from 1930 to 1973, it rose to over $2.50 by 1982. See Energy Information Administration, An Analysis of the Department of Energy’s Notice of Proposed Rulemaking (NOPR), “Ceiling Prices: Old Gas Pricing Structure ” 3 (1986). While the wellhead price in 1972 represented only about 28% of the delivered price to consumers, by 1981 that fraction had risen to 57%. Id. at 5.
These developments lie in the background of the key Commission findings in support of Order No. 436. (1) Despite the growth of a competitive wellhead market, the interstate pipelines retain market power in gas transportation. J.A. 306. (2) Pipelines have generally declined to transport gas in competition with their own sales (except for transportation to customers that can switch to alternative fuels at little or no extra cost). J.A. 318-19. (3) Pipeline discrimination in transportation has denied consumers access to gas at the lowest reasonable rates. J.A. 320, 352.
Thus the early 1980s created the likelihood, for the first time, that the Commission could best fulfill the purposes of the NGA by adopting rules enabling customers to buy gas at the wellhead and to overcome the interstate pipelines’ general refusal to move gas that would compete with their own sales. Besides protecting consumers from the burden of the pipelines’ purchase contracts at supra-market prices, such rules would have the long-term effect of subjecting pipelines to the ordinary constraints of a middleman under competitive conditions. This the Commission set out to achieve in Order No. 436.
The Order’s regulatory package includes these elements: (1) If a pipeline seeks to take advantage of “blanket certification” of transportation [i.e., a certificate authorizing transportation services generically and thus obviating the need for unwieldy individual certification), it must commit itself to provide transportation on a nondiscriminatory basis (and thus become an “open-access” pipeline). (2) When demand outruns capacity for open-access transportation, the open-access pipeline shall allocate capacity on a “first-come, first-served” basis. (3) Rate regulation for open-access transportation will take the form of ceilings and floors, with the pipeline free to adjust rates within that band. (4) Any open-access pipeline, by applying for certification, agrees to allow its LDC customers to convert their “contract demand” (“CD”) (i.e., contract commitment to purchase gas) from an obligation to purchase gas to an obligation to use (or pay for) transportation services. The point of the option is to make open access a reality for the pipelines’ LDC customers despite long-term contractual service arrangements previously certificated by the Commission. The Order also requires an open-access pipeline to give its LDC customers the option to reduce contract demand. (5) The Commission will issue “Optional, Expedited Certificates” for new facilities, services and operations where the pipeline undertakes the entire economic risk of the project. The Commission declined to include in the package any special provision to relieve pipelines from the burden of take-or-pay contracts providing for prices well above current competitive levels.
*17Each component of the package contains many details not given above, some of them the source of challenges in this case. The details will be developed as appropriate in the opinion.
II. Open-Access Requirements
A. Claimed Deficiencies in Statutory Authority to Require Open Access.
Order No. 436 imposes an “open-access” commitment on any pipeline that (1) secures a “blanket certificate” to provide gas transportation, pursuant to § 7 of the NGA, 15 U.S.C. § 717f (1982), or (2) actually provides transportation under § 311 of the NGPA, 15 U.S.C. § 3371 (1982). See 18 C.F.R. §§ 284.8(b), 284.9(b).3 Several pipelines and others attack these conditions as beyond the scope of the Commission’s authority under the two statutes. The arguments under both statutes rely on the proposition that the “open-access” condition is equivalent to a “common carriage” requirement, as both the condition and common carriage have at their core a duty to accept shipments from all would-be shippers. The two statutes differ radically in their language, however, so we treat the claims separately, rejecting both.
1. Natural Gas Act.
The pipelines can point to no language in the NGA barring the Commission from imposing common carrier status on natural gas pipelines, and certainly none barring it from imposing upon the pipelines a specific duty that happens to be a typical or even core component of such status. They seek to overcome the statutory silence by means of legislative history. The task is uphill; “courts have no authority to enforce principles gleaned solely from legislative history that has no statutory reference point.” IBEW, Local No. 474 v. NLRB, 814 F.2d 697, 712 (D.C. Cir.1987) (emphasis deleted) (citing United States v. American College of Physicians, 475 U.S. 834, 106 S.Ct. 1591, 1598, 89 L.Ed.2d 841 (1986)).
The legislative history here consists entirely of congressional inaction. In 1906, when Congress brought oil pipelines under the wing of the Interstate Commerce Commission, it amended the Interstate Commerce Act to exclude natural gas pipelines from the category of common carrier, thus making clear that they were not covered by the extension of jurisdiction. Pub.L. No. 59-397, § 1, 34 Stat. 584, 584 (codified as amended at 49 U.S.C. §§ 10,102, 10,501 (1982)). In 1913 a bill was introduced in the Senate that would have reversed the 1906 decision, see S. 3345, 63d Cong., 2d Sess., 50 Cong. Rec. 5847-49 (1913), but it was never enacted. Finally, in 1935 a bill presaging the NGA — similar to the ultimate statute but explicitly imposing common carrier duties — died in committee. See H.R. 5423, 74th Cong., 1st Sess., 70 Cong. Rec. 1624 (1935).
This history provides strong support only for the point that Congress declined itself to impose common carrier status on the pipelines — a proposition that is unquestioned and is evident from the language of the statute itself. The chain of inference gets steadily weaker as we move toward more relevant issues. The history supplies modest support for the view that Congress did not intend the Commission to impose common carriage conditions at will. It affords weak — almost invisible — support for the idea that the Commission could under no circumstances whatsoever impose obligations encompassing the core of a common carriage duty.
The weakness of the legislative history is underscored when we examine the very component of “common carriage” on which the challenging pipelines rest their case: the duty to carry without discrimination. Insofar as they argue that a concern about discrimination has been a driving force behind legislative imposition of common carriage regulation, history is on their side. See, e.g., Louisville & Nashville R.R. v. United States, 282 U.S. 740, 749-50, 51 S.Ct. 297, 300-01, 75 L.Ed. 672 (1931); L. Gorton, The Concept of the Common *18Carrier in Anglo-American Law 42-43 (1971). But in the NGA Congress affirmatively addressed itself to that issue, giving the Commission power to stamp out undue discrimination; it is precisely that power that the Commission has here sought to exercise.
The Act fairly bristles with concern for undue discrimination. Section 4 prohibits any “undue preference” and any “unreasonable difference in rates, charges, service, facilities, or in any other respect,” and empowers the Commission to review tariffs filed by pipelines in order to reject ones violating the prohibition. 15 U.S.C. § 717c (1982). Section 5 — the primary authority invoked by the Commission here — directs the Commission to adopt corrective measures whenever it finds a “rate, charge or classification,” or any “rule, regulation, practice, or contract” affecting the same, to be “unjust, unreasonable, unduly discriminatory, or preferential.” Id. § 717d.
The issue seems to come down to this: Although Congress explicitly gave the Commission the power and the duty to achieve one of the prime goals of common carriage regulation (the eradication of undue discrimination), the Commission’s attempted exercise of that power is invalid because Congress, in 1906 and 1914 and 1935 and 1938 itself, refrained from affixing common carrier status directly onto the pipelines and from authorizing the Commission to do so. And this proposition is said to control no matter how sound the Order may be as a response to the facts before the Commission. We think this turns statutory construction upside down, letting the failure to grant a general power prevail over the affirmative grant of a specific one.
Thus we find little relevance in cases relied on by the pipelines for the proposition that a duty to provide service to all comers is the essence of common carriage. In each of those cases the court was construing the term as used in a statute, in one instance stating that a particular class of persons “shall not ... be deemed a common carrier,” FCC v. Midwest Video Corp., 440 U.S. 689, 699-702, 99 S.Ct. 1435, 1440-1442, 59 L.Ed.2d 692 (1979), in the others prohibiting agency exercise of jurisdiction over activities of certain persons classified as common carriers, see National Ass’n of Regulatory Utility Commissioners v. FCC, 525 F.2d 630 (D.C. Cir.1976); National Ass’n of Regulatory Utility Commissioners v. FCC, 533 F.2d 601 (D.C. Cir.1976). Such cases are not helpful on the issue of whether Congress’s omission of the term common carrier significantly undercuts its explicit provision of authority to prevent or correct undue discrimination.
Petitioners cite Florida Power & Light Co. v. FERC, 660 F.2d 668 (5th Cir.1981), and Richmond Power & Light v. FERC, 574 F.2d 610 (D.C. Cir.1978), for the following proposition: that any order by the Commission conditioning its approval of any “wheeling” (i.e., transmission of electricity owned by another) on the utility’s agreement to wheel for all constitutes an attempt by the Commission to impose indirectly duties that the Federal Power Act prevents it from imposing directly, namely common carriage. As the Federal Power Act establishes a regulatory scheme for electricity paralleling that which the NGA creates for gas, see, e.g., FPC v. Sierra Pacific Power Co., 350 U.S. 348, 353, 76 S.Ct. 368, 371, 100 L.Ed. 388 (1956), petitioners contend that the Richmond and Florida cases compel invalidation of the Commission’s open-access condition.
We think petitioners read Richmond and Florida far too broadly. First, we note that the legislative history of the two acts is, on this point, materially different. In its deliberations on the bill that ultimately emerged as the Federal Power Act, Congress considered and rejected a provision that would have “empowered the Federal Power Commission to order wheeling if it found such action to be ‘necessary or desirable in the public interest.’ ” Otter Tail Power Co. v. United States, 410 U.S. 366, 374, 93 S.Ct. 1022, 1028, 35 L.Ed.2d 359 (1973) (quoting S. 1725, 74th Cong., 1st Sess.). The evidence as to the NGA (surveyed above) is less direct: it consists exclusively of various occasions on which Congress did not adopt proposals actually *19making the natural gas pipelines into common carriers.
Second, neither Richmond nor Florida comes anywhere near stating that the Commission is barred from imposing an open-access condition in all circumstances. In Florida, the court expressly left open the question whether the Commission would be entitled to use open-access conditions as a remedy for anti-competitive conduct. 660 F.2d at 677-79. It stressed the failure of the Commission, in the orders under review, to “make any findings of anti-competitive activities or violations.” Id. at 678. Evidently because no party raised the issue, the court did not address the Commission’s power to exact such conditions as a remedy for undue discrimination. Cf. FPC v. Conway Corp., 426 U.S. 271, 276-77, 96 S.Ct. 1999, 2003-04, 48 L.Ed.2d 626 (1976) (accepting assumption that the Commission could not remedy a utility’s discrimination between wholesale (jurisdictional) rates and retail (nonjurisdictional) rates by ordering increases in the latter).
In Richmond the Commission had repulsed Richmond Power & Light’s demand that it condition approval of any transmission by either of two large interstate systems on their agreeing to transmit for all. In upholding the Commission we said little more than that unwillingness to transmit for all comers could not be automatically deemed an undue discrimination:
Thus Richmond spurned the opportunity to demonstrate that particular activities were unreasonably anticompetitive or discriminatory and claimed instead that the mere failure to wheel energy to and from Richmond while wheeling for any other utility was unlawful discrimination.
574 F.2d at 623. We went on to say that the Commission’s rejection of a per se rule (i.e., a rule that selective transmission was necessarily undue discrimination) followed logically from Congress’s refusal to impose common carrier duties on electric utilities. Id.
Petitioners’ reading of Richmond, moreover, is belied by this court’s later decision in Central Iowa Power Coop. v. FERC, 606 F.2d 1156 (D.C.Cir.1979). Pursuant to § 205 of the Federal Power Act, 16 U.S.C. § 824d (1982) (paralleling § 4 of the NGA), the Commission had reviewed the terms of a power-pooling agreement that established two classes of membership, one of them entitled to fewer privileges than the other. Finding the distinction discriminatory on its face, the Commission conditioned its approval on removal of the membership criteria that prevented the inferior class of participants from enjoying the privileges of the favored ones. We upheld the decision as a proper exercise of its power to prevent undue discrimination. 606 F.2d at 1170-72. See Reiter, Competition and Access to the Bottleneck: The Scope of Contract Carrier Regulation Under the Federal Power and Natural Gas Acts, 18 Land & Water L.Rev. 1, 47-50 (1983). The Commission’s open-access condition relies on precisely that principle.
It is true that in Central Iowa the court rejected South Dakota’s claim that the Commission should have conditioned approval of the power-pooling agreement on the parties agreeing to wheel for nongenerating electrical systems. See 606 F.2d at 1169. Such a condition would in effect have forced on the participants a broad extension of their agreement to wheel for each other. See id. at 1160 n. 7. Though the court brushed the request aside in fairly sweeping language, its approval of the Commission’s elevation of the inferior class of members clearly limits the negative impact of the discussion. The case upholds the power of the Commission to subject approval of a set of voluntary transactions to a condition that providers open up the class of permissible users.
Neither Florida nor Richmond presented the extreme factual circumstances that are now present in the gas industry. Here the Commission has found (a) that pipelines continue to possess substantial market power, J.A. 306; (b) that they have exercised that power to deny their own sales customers, and others without fuel-switching capability, access to competitively priced gas, J.A. 318; and (c) that this practice has denied consumers access to gas at the lowest reasonable rates, J.A. 318-21, *20352. Thus, despite the removal of regulation over the price and non-price aspects of wellhead transactions, and the evolution of an interconnected nationwide pipeline grid, discrimination in transportation has denied gas users, and the economy generally, the benefits of a competitive wellhead market.
Despite a sweeping suggestion that the Order is “unsupported,” Brief of Interstate Pipeline Group at 35, the pipelines do not in fact challenge these factual findings. Their objection is rather on matters of policy, grounded on beliefs that the kind of discrimination here prohibited is not “undue” within the meaning of the NGA.4 Indeed, the burden of their attack on the Order is precisely that it may disable them from passing on to customers gas purchase costs that they incur under contracts entered into years ago under premises now obsolete. Id. at 37-38. In other words, enforcement of Order No. 436 will expose them to competition that their discriminatory practices enable them to avoid. Their claim thus tends to substantiate the Commission’s views (1) that in the absence of Order No. 436 competition will be thwarted and (2) that the practices controlled by Order No. 436 are indeed anti-competitive and discriminatory.
The electricity cases cited thus provide only very weak support for the challenge. In contrast, our decision in Maryland People's Counsel v. FERC, 761 F.2d 780 (D.C. Cir.1985) (‘‘MPC II”), came about as close to endorsing the Commission’s approach as Article III permits. There we vacated orders of the Commission that had established “blanket certificate” transportation without any specific effort to prevent pipelines from offering such transportation on a discriminatory basis. We did not, of course, explicitly find “undue discrimination” such as would obligate the Commission to act under § 5. But we did find that the petitioners there had made a strong enough showing to require the Commission to address the issue. Specifically, we made it clear that blanket-certificate transportation, unconstrained by any nondiscriminatory access provision, might well require remedial action under § 5. We ended by saying:
We vacate the challenged orders to the extent that they allow transportation of direct-sale gas to fuel-switchable, non-“high-priority” end users without requiring pipelines to furnish the same service to LDCs and captive consumers on nondiscriminatory terms.
761 F.2d at 789 (footnote omitted).
Our holding in MPC II obviously did not require the Commission to make the findings that it has. It surely carried the implication, however, that if it did make supportable findings of undue discrimination in pipeline use of the old blanket certificates, it would have the authority to employ suitable remedies. And it carried the further implication that among them might be a requirement that any pipeline offering blanket-certificate transportation agree to serve “LDCs and captive consumers on non-discriminatory terms.” Id.
The interstate pipelines appear to suggest that Order No. 436’s impact on their financial integrity is so grave as to be equivalent to a rule denying them the legal right to pass on costs, and invalid as such a rule would be. It is true that the Commission has only very limited power to deny the pipelines the legal right to pass gas purchase costs through to customers. See § 601(c)(2) of the NGPA, 15 U.S.C. § 3431(c)(2) (1982) (providing that the Commission must allow interstate pipelines to pass through gas costs not violating NGPA wellhead ceilings “except to the extent the Commission determines that the amount paid was excessive due to fraud, abuse, or similar grounds”); Office of Consumers’ *21Counsel v. FERC, 783 F.2d 206 (D.C.Cir.1986). But petitioners have called our attention to nothing that bars the Commission from devising rules that remedy a lack of competition by exposing pipelines to competition and its normal consequences. The Supreme Court has made clear, for example, that the due process clause affords no protection from losses inflicted by market conditions. In Market Street Ry. v. Railroad Comm’n, 324 U.S. 548, 65 S.Ct. 770, 89 L.Ed. 1171 (1945), it said of its decision in FPC v. Hope Natural Gas Co., 320 U.S. 591, 64 S.Ct. 281, 88 L.Ed. 333 (1944):
All that was held was that a company could not complain if the return which was allowed made it possible for the company to operate successfully. There was no suggestion that less might not be allowed when the amount allowed was all the company could earn____ The due process clause ... has not and cannot be applied to insure values or to restore values that have been lost by the operation of economic forces.
324 U.S. at 566-67, 65 S.Ct. at 779-80. Similarly, nothing in the NGA protects the pipelines from the market forces to which Order No. 436 subjects their gas marketing business, even though those forces are derived in part from a restriction on their discrimination in transportation.
It is finally argued that the Commission’s not having imposed any requirements like those of Order No. 436 in the period from enactment in 1938 until the present demonstrates the lack of any power to do so. Cf. FPC v. Panhandle Eastern Pipe Line Co., 337 U.S. 498, 513-14, 69 S.Ct. 1251, 1260-61, 93 L.Ed. 1499 (1949). But as our introductory review of the economic background sought to illustrate, the Commission here deals with conditions that are altogether new. Thus no inference may be drawn from prior non-use.
While the Supreme Court’s decision in Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), is not a wand by which courts can turn an unlawful frog into a legitimate prince, the case bolsters our conclusion. Congress has given the Commission in § 5 of the NGA a broad power to stamp out undue discrimination; in § 7 the power to approve certificates of service subject to “such reasonable terms and conditions as the public convenience and necessity may require”; and in § 16 the power to “perform any and all acts, and to prescribe ... such orders, rules, and regulations as it may find necessary or appropriate to carry out the [NGA’s] provisions.” The alleged negative restriction on this power is at best ambiguous, if indeed it exists at all. Under these circumstances, Chevron binds us to defer to Congress’s decision to grant the agency, not the courts, the primary authority and responsibility to administer the statute. The Commission’s view represents “a reasonable interpretation” of the Act, for which we may not substitute our view. 467 U.S. at 844, 104 S.Ct. at 2782.
2. Natural Gas Policy Act.
In enacting the NGPA Congress endeavored to break down the regulatory barriers between the interstate and intrastate markets. Sections 311 and 602(b)(2) were added to “facilitate[ ] development of a national natural gas transportation network without subjecting intrastate pipelines, already regulated by State agencies, to FPC regulation over the entirety of their operations.” H.R.Rep. No. 543, 95th Cong., 1st Sess. 45 (1977). See also Public Service Comm’n v. Mid-Louisiana Gas Co., 463 U.S. 319, 342-43, 103 S.Ct. 3024, 3037-38, 77 L.Ed.2d 668 (1983). Thus, § 311(a) permits the Commission to authorize transportation by an interstate pipeline on behalf of an intrastate or LDC and by an intrastate on behalf of an interstate or LDC. 15 U.S.C. § 3371(a) (1982). The Commission has exercised this authority, permitting such transportation as a general matter. See 18 C.F.R., part 284, subparts B and C. It has also exercised the authority provided by § 311(c), 15 U.S.C. § 3371(c) (1982), to prescribe terms and conditions.5 Order No. *22436 would add to these the requirement that any interstate or intrastate pipeline offering such transportation shall do so on a nondiscriminatory basis. 18 C.F.R. §§ 284.8(b) and 284.9(b). Petitioners challenge these open-access conditions, relying here on express statutory language. Section 602 of the NGPA, captioned “Effect on State Laws,” provides in subsection (b):
(b) Common carriers.
No person shall be subject to regulation as a common carrier under any provision of Federal or State law by reason of any transportation—
(1) pursuant to any order under section 3362(c) or section 3363(b), (c), (d), or (i) of this title; or
(2) authorized by the Commission under section 3371(a) of this title [section 311(a) of the NGPA].
15 U.S.C. § 3432(b) (1982).
Petitioners read this language as stultifying any effort by the Commission to control discrimination where that effort imposes on pipelines a duty—even though it be a conditional one—equivalent to the common carrier’s duty to provide nondiscriminatory service. We believe this interpretation is incorrect.
It seems to us that § 602(b)(2) was a congressional effort to protect § 311(a) from the consequences of pipeline concern that service thereunder would expose them generally to classification as common carriers, most likely by states, and thus to an unprecedented range of legal burdens. If pipelines shied away from use of § 311(a), its purpose would be defeated. Section 602(b)(2) could protect against that threat by assuaging the pipelines’ concern. A duty not to discriminate, imposed by the Commission on the basis of findings that the duty is necessary to assure consumers access to competitively priced gas, is utterly different. The imposition of the duty here facilitates the accomplishment of Congress’s purposes. At least it will do so if the gains in enhanced access offset whatever losses may result from the disincentive effect on pipelines. The judgment balancing those consequences is for the Commission to make, and it has made it in favor of imposing the duty.6
Congress had ample reason to fear that the risk of extraneous legal burdens would chill pipeline interest. Take the laws of Texas. Two years before enactment of the NGPA a Texas court ruled that “[w]hether the business conducted by a pipe line company is actually that of a common carrier is a question of fact,” which would depend on the court’s perception of whether “the line is available to all producers seeking its services.” China-Nome Gas Co. v. Riddle, 541 S.W.2d 905, 908 (Tex.Civ.App.1976). A pipeline’s transportation under § 311(a) would expose it to the risk of such “fact” findings, and thus enmesh it in state regulations. In Texas, for example, such classification would subject it to the jurisdiction of the Railroad Commission, see Op.Atty. Gen. No. M-175 (1967), to certain health requirements, see Tex.Civ.Stat.Ann. art. 4477-1 §§ 1 & 22 (Vernon 1976 & Supp. 1987), and to such duties and liabilities as a court might find the common law to prescribe, see id. arts. 882-884 (Vernon 1964 & Supp.1987). Most notably, a firm declared a common carrier is required under Texas law to carry, for anyone, any goods of the type for which it is suited. Id. § 884.
Pipeline fear of such extraneous burdens might well have rendered § 311 a dead letter. Though we have not identified similar federal hazards, we believe that Congress might well have included the reference to federal law out of anxiety that some overlooked federal provision would operate to thwart § 311’s purposes. Viewed in this light, § 602 has nothing to do with a Commission decision to impose a duty of nondiscrimination.
The structure of § 602 favors this reading over the broader one urged by the petitioners. Section 602 also provides that no one shall be subject to regulation as a *23common carrier by reason of providing transportation under 15 U.S.C. §§ 3362(c) or 3363(b), (c), (d), or (i), which involve presidential orders to transport in a natural gas supply emergency. The President’s emergency powers are extremely broad, and their exercise could well impose duties quite like those of common carriage, with the President dictating detailed priorities about whom the pipelines should serve. See id. § 3363. If the President determined that imposition of something like common carriage were necessary to meet such an emergency, it is hardly credible that § 602 would stand in the way. By the same token, that section must not flatly bar the Commission from imposing similar conditions on gas transportation under § 811.
Congress may conceivably have intended § 602 to bar FERC from conditioning § 311 transportation upon assurances of nondiscrimination. We doubt it. But apart from our independent conclusion that it has no such purpose, we regard FERC’s interpretation as reasonable. The reasonableness is underscored by FERC’s broad duties to assure consumers access to natural gas at prices such as would prevail in the absence of pipeline market power and its conclusion that under the present circumstances fulfillment of that duty requires such conditioning. Cf. American Trucking Ass’ns, Inc. v. Atchison, T. & S.F. Ry., 387 U.S. 397, 87 S.Ct. 1608, 18 L.Ed.2d 847 (1967). We are therefore under Chevron bound to uphold the Commission.
A parallel attack on Order No. 436 stresses § 601(a)(2)(A) of the NGPA, 15 U.S.C. § 3431(a)(2)(A) (1982), providing that transportation under § 311 (or under the Presidential emergency powers discussed above) shall not constitute “transportation in interstate commerce” within the meaning of § 1(b) of the NGA. But for this provision, a transporting intrastate pipeline would fall prey to the Commission’s NGA jurisdiction. Several petitioners argue that Order No. 436 imposes burdens substantially identical to those encompassed by NGA jurisdiction. Thus, they argue, application of the open-access condition to intrastate pipelines is an impermissible interference with state regulatory authority. See Brief of the American Gas Ass’n at 42-43; Brief of Intrastate Petitioners & Intervenors at 22-29.
Again consideration of the purpose of the provision refutes the attack. Section 601(a)(2)(A) is clearly intended to assure that pipelines' fear of the automatic imposition of the burdens of NGA jurisdiction does not make them so chary of § 311 that it languishes unused. This is altogether different from regulatory burdens imposed by the Commission in the exercise of its discretion under § 311(c) in order to make sure that § 311 transportation operates in harmony with the congressional purpose. Thus, even if it were true that the regulatory impact of Order No. 436 were identical to that of NGA jurisdiction, the decision to condition § 311 on acceptance of those burdens would not violate § 601(a)(2)(A).
In fact, of course, the nondiscrimination duties imposed by Order No. 436 by no means encompass all the burdens of NGA jurisdiction. (1) New service under § 311 does not require § 7 certification, and Order No. 436 carries that distinction forward where a pipeline brings itself under the Order: operations under a § 7 blanket transportation certificate entail notice-and-protest procedures more burdensome than the reporting requirements for § 311 transportation. See infra part VII.B. (2) Construction of facilities to be used exclusively for § 311 transportation requires no certification or FERC review at all. See infra part VII.D. (3) § 311 transportation does not subject an intrastate pipeline to the detailed accounting provisions applicable to a natural gas company under the NGA. Compare 18 C.F.R. part 201. (4) Intrastate pipelines may provide firm or interruptible service without providing both, while interstate pipelines providing one type must also provide the other. Compare 18 C.F.R. §§ 284.8(a)(1) & 284.9(a)(1) with id. §§ 284.8(a)(2) & 284.9(a)(2).
B. Alleged Failure to Comply with Mandate of Outer Continental Shelf Lands Act.
The Petitioner Industrial Groups (the Process Gas Consumers Group and *24American Iron and Steel Institute) claim that under §§ 5(e) and 5(f) of the Outer Continental Shelf Lands Act, 43 U.S.C. §§ 1334(e) & (f) (1982), the Commission must require every gas pipeline operating in the OCS to provide nondiscriminatory access for others’ OCS gas throughout the entire system of the “pipeline entity.”
Congress included § 5(e)7 in OCSLA at the time of original adoption, and then sought to “strengthen[]” it8 in 1978 by adding § 5(f):
(f) Competitive Principles Governing Pipeline Operation
(1) Except as provided in paragraph (2) [the gathering exemption], every permit, license, easement, right-of-way, or other grant of authority for the transportation by pipeline on or across the outer Continental Shelf of oil or gas shall require that the pipeline be operated in accordance with the following competitive principles:
(A) The pipeline must provide open and nondiscriminatory access to both owner and nonowner shippers____
43 U.S.C. § 1334(f) (1982).
The language will not bear the proposed load. The statute demands that any permit for transportation by pipeline on the OCS require that “the pipeline" be operated according to specified principles. The natural reading is that the subject pipeline is the physical facility in the OCS, not every facility owned or operated by the corporation operating that facility. The petitioners call our attention to remarks on the Senate floor reflecting concern about pipeline discrimination. See, for example, Senator Kennedy’s observation, “This amendment seeks to insure that OCS pipelines are true common carriers.” 123 Cong.Rec. 23,252 (July 15, 1977). But none of the remarks supports petitioners' proposed inference. All are completely consistent with a focus on what Congress had before it — the OCS.9
C. Claims of Arbitrariness and Caprice.
1. Failure to impose the nondiscriminatory access conditions on § 7(c) transportation certificates.
Maryland People’s Counsel finds Order No. 436 defective in that it (potentially) allows a pipeline to provide transportation under an individual certificate issued under § 7(c) without agreeing to provide the same on a nondiscriminatory basis.
We lack jurisdiction over the claim. Section 19(a) of the NGA, 15 U.S.C. § 717r(a) (1982), prohibits any “proceeding to review” an order of the Commission in the absence of an application for rehearing filed within 30 days after issuance of the order, setting forth specifically the ground on which the application is based. The Commission addressed the problem of individual § 7 certificates in Order No. 436, stating in its analysis of comments that it did not intend to apply the nondiscriminatory access provision “on a generic basis to all section 7 certificates at this time.” J.A. *25388. MPC did not file an application raising the point until March 1986, long after expiration of the 30 days from issuance of Order No. 436 on October 9, 1985.
MPC acknowledges the jurisdictional difficulty, but states that it could not have realized, until February 1986, that the Commission’s refusal encompassed certificates for transportation to fuel-switchable customers. At that time, in Texas Gas Transmission Corp., 34 F.E.R.C. ¶ 61,203 (1986), the Commission actually did issue such a certificate without open-access conditions. MPC argues that, in view of the obligations imposed by MPC II, and language of the Commission elsewhere in Order No. 436, it was entitled to believe that the Commission’s statement referred only “to transport in support of [the pipelines’] merchant function (e.g., where one pipeline transports gas owned by another pipeline) or in ways that would not create the discrimination that the Commission found unlawful in Order No. 436 (e.g., transportation for high priority customers ...).” Reply Brief of MPC at 4.
We must reject MPC’s reading of Order No. 436’s disclaimer. The statement is broadly phrased to encompass “section 7 transportation certificates,” and is justified in terms of the Commission’s opportunity to scrutinize “individual section 7 transportation arrangements ... on a ease-by-case basis when [the certificates] are applied for.” J.A. 388. Both the language and the explanation are fully as applicable to transportation to fuel-switchable users as to any other transportation. MPC of course remains free to challenge the policy by seeking review of individual §• 7 orders in which the Commission applies it.
The Commission briefly sought to refute MPC’s contention in Order No. 436-D, issued on March 28, 1986. Such discussion was merely dictum, as MPC’s March 1986 application was time-barred under § 19 and the Commission so recognized. In any event, the Commission cannot waive the jurisdictional bar of § 19 by selective discussion of belated rehearing applications. See Boston Gas Co. v. FERC, 575 F.2d 975, 979-80 (1st Cir.1978).
2. Capacity allocation on a “first-come, first-served” basis.
Several parties attack as arbitrary and capricious the Commission’s “first-come, first-served” formula for determining priorities among those who seek transportation service under the Order.
The Commission did not announce this formula in any regulation but merely in material supporting the regulations. See J.A. 342, 401-04. It also said that certain claims on pipeline capacity might enjoy favored treatment, “outside the general first-come, first-served rule.” J.A. 400. These preferred claims include those of “a firm sales customer,” on the grounds that such a customer “has already booked the transportation capacity currently ‘bundled’ with ... the sale.” Id. (emphasis in original). A similar special priority applies to LDCs that exercise the CD conversion option discussed in part IV of this opinion. Id. at 401. Further, the “first-come, first-served” concept is evidently not to apply in cases of sudden capacity interruption, but only to the process of “contracting for available capacity.” Id. at 1095-96 (emphasis in original). But see El Paso Natural Gas Co., 35 F.E.R.C. ¶ 61,440, at 62,061 (June 27, 1986).
Apart from introducing the complexity of these exceptions and superior claims, Order No. 436 and its supporting statements contain no guidance about how pipelines are to implement this formula. “First come, first served” is an easy principle to apply in a bakery where each customer pulls a numbered ticket on entering and is served in that order. But in an industry such as natural gas transportation it may often be difficult to say who “comes first.”
Here are a few sample questions that the rule fails to resolve: (1) Suppose that A, an end user, contracts with a pipeline for the right to transmit up to five billion Btu per day for five years. At the end of four years A seeks to renew the contract on the same terms. But others have earlier filed requests that in the aggregate exceed the pipeline’s capacity. Does A go to the end of the line? Such a result would probably *26disrupt most notions of ordinary business arrangements in this market. But if A and persons similarly situated enjoy a sort of super-priority, open access will be an empty promise for new would-be users. (The Commission appears to lay great stress on the date on which a request is filed, J.A. 1191, but the full implications are nowhere spelled out.) (2) What of efforts by A to secure not merely continued but additional transportation at the end of a fixed-term contract? Would A go to the end of the line for the increment? (See El Paso Natural Gas Co., supra, at 62,060.) (3) What is the impact of a minor change in point of receipt? If such a change forces the user to go to the end of the line, then an LDC or end user may find it hard to shift from one supplier to another. See Brief of Baltimore Gas & Elec. Co. at 17. (4) May a pipeline charge a fee for accepting requests for service? If not, how can it prevent all potential users from filing immediately for virtually unlimited claims on capacity for an indefinite period of time?
The most potent objection to the Commission’s treatment of the problem — and one that is unquestionably ripe — is the contention that it leaves an intolerable gap in the regulatory structure. That gap creates some risk that pipelines may use the resulting leeway to persist in the discrimination that Order No. 436 nominally forbids. It leaves even the most willing pipeline uncertain as to what full compliance requires. On the other side, shippers cannot know what steps they must take to secure adequate priority status or what should guide them in choosing between bundled and unbundled service.
The Commission’s brief treats the problem dismissively, noting that aggregate annual gas consumption has fallen from a peak of 22.6 trillion cubic feet in 1973 to only about 17-18 Tcf currently. FERC Brief at 101. This comment seems utterly irrelevant: the problem is surely capacity at peak periods. The Commission expressly concedes that generally pipelines have operated at capacity during the winter peak. See FERC Brief at 103 n. 3; J.A. 280.
Nonetheless, as each pipeline elects to become an open-access transporter, it must file tariffs with the Commission to govern the service, which tariffs must include any “operational conditions” the pipeline proposes to apply. 18 C.F.R. §§ 284.7(a), 284.-8(c), 284.9(c). See, e.g., El Paso Natural Gas Co., 35 F.E.R.C. ¶ 61,440 (June 27, 1986). These filings afford the Commission an opportunity to develop standards of permissible capacity allocation. Accordingly, the essential legal issue is the validity of the Commission’s choice to address the problem in that format rather than in the format of generic rulemaking.
It is clear that this choice “lies primarily in the informed discretion of the administrative agency.” SEC v. Chenery Corp., 332 U.S. 194, 203, 67 S.Ct. 1575, 1580, 91 L.Ed. 1995 (1947). The Supreme Court has noted that such resolution is appropriate where problems arise that the agency could not reasonably foresee, or where its experience makes adoption of a “hard-and-fast” rule unsuitable, or where the problem is so specialized and variable as to be “impossible of capture within the boundaries of a general rule.” Id. at 202-03, 67 S.Ct. at 1580-81. See also NLRB v. Bell Aerospace, 416 U.S. 267, 294, 94 S.Ct. 1757, 1771, 40 L.Ed.2d 134 (1974). While the Commission has said little or nothing to explain why it is sensible to postpone these decisions to the stage of review of individual proposals, the necessity of its conducting that review seems to us* at this point, an adequate ground for deferring to its judgment.
This is not to say, however, that the Commission may endlessly postpone the necessary decisions. Failure to make the rule reasonably determinate by the time a pipeline starts Order No. 436 operations would severely constrain the Commission’s authority to enforce the Order against the pipeline. Such failure would at least complicate actions seeking injunctive relief against pipelines under § 20 of the NGA, 15 U.S.C. § 717s (1982), and would probably make impossible any effort to secure penalties under § 21, 15 U.S.C. § 717t (1982). See NLRB v. Majestic Weaving Co., 355 F.2d 854, 860 (2d Cir.1966) (Friend*27ly, J.) (“the [judicial] hackles bristle still more when a financial penalty is assessed for action that might well have been avoided if the agency’s changed disposition had been earlier made known, or might even have been taken in express reliance on the standard previously established”). Cf. Boyce Motor Lines, Inc. v. United States, 342 U.S. 337, 340, 72 S.Ct. 329, 330, 96 L.Ed. 367 (1952) (requirement of adequate notice for criminal enforcement of administrative regulations). Moreover, if the Commission approves plans of compliance so vague that enforcement is impaired, its posture will be essentially that found fatally defective in MPC II: it will have authorized blanket certificate transportation under rules not adequately grappling with the potentially discriminatory effects.
The Commission’s oracular procrastination (a blend of Delphi and Fabius) makes challenges to the specifics of “first come, first served” unripe. These challenges include assertions that the policy (1) gives inadequate attention to contractual commitments or to “dependency” as bases of distinction; (2) unduly threatens the security of supply of LDCs; and (3) disregards equities based on prior payments for pipeline capacity. See, e.g., Brief of Interstate Pipeline Group at 35-36; Brief of Associated Gas Distributors at 39-41. One intervenor also poses a carefully reasoned attack on the Commission for its failure to consider alternatives such as an auction system. See Brief of Baltimore Gas & Elec. Co. at 19-20. Though the point is much closer, the Commission’s vagueness and lack of commitment are such that even this attack appears unripe. The ripeness doctrine seeks to
prevent the courts, through avoidance of premature adjudication, from entangling themselves in abstract disagreements over administrative policies, and also to protect the agencies from judicial interference until an administrative decision has been formalized and its effects felt in a concrete way by the challenging parties.
Abbott Laboratories v. Gardner, 387 U.S. 136, 148-49, 87 S.Ct. 1507, 1515-16, 18 L.Ed.2d 681 (1967). As the Commission confined its disposition of the issue to some general remarks in its supporting statement, our involvement in the merits at this stage would defy these principles.
III. Rate Conditions
With the stated intention of imposing on pipelines more of the risk and responsibility for their own business decisions, the Commission has established a system of flexible rates. See 18 C.F.R. §§ 284.7, 284.8(d), 284.9(d).10 Tariffs are to provide for ceilings and floors, with the pipeline free to charge anywhere within that band. Each maximum rate is to be based on what is typically known as “fully allocated cost,” i.e., a rate such that, if the pipeline carries projected volume at the specified unit price, it should exactly recover all costs allocable to the relevant service for the period. See 18 C.F.R. § 284.7(c)(3). Minimum rates are to be based on average variable cost. See 18 C.F.R. § 284.7(d)(4)(h). The maximum rates are to vary depending on whether the service is in a peak or off-peak period, and on whether it is firm or interruptible service. A pipeline discounting any service from the maximum rate must, within 15 days of the close of the billing period, report the maximum rate for the transaction, the rate actually charged, the shipper’s identity, and any corporate affiliation between pipeline and shipper. 18 C.F.R. § 284.7(d)(5)(iv).
Pipelines may charge a “reservation fee” for firm service. Otherwise shippers could request whatever volume they liked, without cost and regardless of intent to use. As requests would vastly exceed capacity, the pipeline could not rationally plan capacity allocation. See J.A. 457-60. Apart *28from the reservation fee, pipelines are required to charge on a “volumetric” basis, i.e., a simple charge per unit actually transported, without a “demand charge” or “minimum bill.”
A. Absence of Finding that Prior Rates Were Unlawful.
The Interstate Pipeline Group objects that the Commission did not make specific findings that any rates charged by individual pipelines were unlawful before imposing the new rate conditions. The Commission is not required to make individual findings, however, if it exercises its § 5 authority by means of a generic rule. See, e.g., Wisconsin Gas Co. v. FERC, 770 F.2d 1144, 1165-68 (D.C.Cir.1985), cert. denied, — U.S. -, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986). The pipelines seek to distinguish Wisconsin Gas on the ground that it “involved specific findings as to a single billing term,” to wit, minimum bill provisions that included variable costs. (Minimum bills charge for specific portions of contract demand even as to gas that is not taken; the Commission believed that inclusion of variable costs in such bills imposed an unjustifiable restriction on customer choice of gas supply and improperly sheltered pipelines from competition.) The distinction is irrelevant. What justified the generic approach in Wisconsin Gas was the Commission’s conclusion that any tariff violating the rule would have such adverse effects on the interstate gas market as to render it “unjust and unreasonable” within the meaning of § 5. That is precisely what the Commission has concluded here.
The pipelines may be claiming that the Commission’s failure to adduce evidence meeting the standards of adjudication breaches the substantial evidence requirement of § 19 of the NGA, 15 U.S.C. §• 717r (1982). Again Wisconsin Gas is dispositive. There the court reaffirmed the court’s conclusion in American Public Gas Ass’n v. FPC, 567 F.2d 1016 (D.C.Cir.1977), that § 19’s reference to “substantial evidence,” located as it is in the provision guiding judicial review, does not dictate the procedure to be employed in FERC’s notice- and-comment rulemakings. Wisconsin Gas, 770 F.2d at 1167-68.
Finally, the pipelines’ complaint may be that the Commission adopted its new rate criteria without “factual” submissions tracing a relationship between rate practices formerly permitted and. the evils that it sought to correct. There may be circumstances in which such a claim would prevail. In Electricity Consumers Resource Council v. FERC, 747 F.2d 1511, 1514 (D.C.Cir.1984), for example, this court declared that “mere reliance on an economic theory cannot substitute for substantial record evidence and the articulation of a rational basis for an agency’s decision.” In fact, however, the court in Electricity Consumers was persuaded that the Commission had “inexplicably distorted” the theory that it claimed to apply. Id. Here the pipelines point to no such inexplicable distortion.
Promulgation of generic rate criteria clearly involves the determination of policy goals or objectives, and the selection of means to achieve them. Courts reviewing an agency’s selection of means are not entitled to insist on empirical data for every proposition on which the selection depends. Wisconsin Gas made that clear. For example, in the rulemaking proceeding parties had objected that curtailment of the minimum bill would result in the pipelines shifting costs to its most captive customers. The Commission responded in part with a prediction that “the increased incentive to compete vigorously in the market would eventually lead to lower prices for all consumers.” 770 F.2d at 1161. The court accepted this without record evidence, presumably because it viewed the prediction as at least likely enough to be within the Commission’s authority. Clearly nothing in Electricity Consumer’s reference to “economic theory” was intended to invalidate agency reliance on generic factual predictions merely because they are typically studied in the field called economics. Agencies do not need to conduct experiments in order to rely on the prediction that an unsupported stone will fall; nor *29need they do so for predictions that competition will normally lead to lower prices.
In support of this objection the pipelines do not identify any factual proposition, relied on by the Commission, that they regard as requiring additional support. Accordingly, the objection cannot succeed.
B. Allowance of Discounting Generally.
Several petitioners object that the Commission’s allowing pipelines to discount from the maximum rates is in effect an approval of “undue preferencefs]” and “undue discrimination” in violation of §§ 4 and 5 of the NGA. But “the mere fact of a rate disparity” is not enough to constitute unlawful discrimination. Cities of Bethany v. FERC, 727 F.2d 1131, 1139 (D.C.Cir.), cert. denied, 469 U.S. 917, 105 S.Ct. 293, 83 L.Ed.2d 229 (1984). The reporting system will enable the Commission to monitor behavior and to act promptly when it or another party detects behavior arguably falling under the bans of §§ 4 and 5. This provision for flexibility conforms to Congress’s intention in the NGA to allow a vital role for private contracting between the parties. See United Gas Pipe Line Co. v. Mobile Gas Service Corp., 350 U.S. 332, 338-39, 76 S.Ct. 373, 377-78, 100 L.Ed. 373 (1956); see also SeaLand Service, Inc. v. ICC, 738 F.2d 1311, 1316-19 (D.C.Cir.1984) (rejecting proposition that “contract rates,” based on individual contract but available to similarly situated shippers of like commodities, are automatically violative of nondiscrimination principle). Accordingly, given the Commission’s broad latitude to choose between rulemaking and adjudication, see SEC v. Chenery Corp., 332 U.S. 194, 67 S.Ct. 1575, 91 L.Ed. 1995 (1947),11 we could find the provisions illegal only if they carried such a risk of allowing undue discrimination or preferences as to be arbitrary and capricious. We do not find the risk so high.
The Associated Gas Distributors call our attention to the problem of discounts in favor of a pipeline’s gas trading affiliate. We recognize that such transactions may carry more than the usual risk of undue discrimination. Cf. NGPA § 601(b)(1)(E), 15 U.S.C. § 3431(b)(1)(E) (1982) (imposing a special limit on pipeline recovery of cost of gas purchased from affiliate). But we see no reason to think that such a discount should be per se unduly discriminatory. If a pipeline gives its gas trading affiliate discounts identical to those given to unaffiliated parties in identical circumstances, the discount would not be unlawful merely on account of the affiliation. Accordingly, the risk of such discounts proving invalid is insufficient to justify invalidation of the rule.
C. Potential Discrimination Between Bundled and Unbundled Transportation.
Other petitioners suggest that the Commission’s rate regulations are invalid because they sanction undue discrimination between unbundled transportation and the transportation component of a bundled sales transaction. That the criteria governing permissible rates in the two categories are different, however, does not establish discrimination between them. Most notably, the petitioners point to no reason to suppose that, as a whole, unbundled transportation service will recover a lower proportion of its costs than will the transportation component of unbundled sales. Indeed, the rate provisions specify that the maximum rates for each subcategory of unbundled transportation are to be designed to recover “solely those costs which are properly allocated to the service to which the rate applies.” 18 C.F.R. § 284.-7(d)(4)(i). That the pipelines may offer discounts does not alter the case. They do so at their own risk, see especially id. at § 284.7(a)(5)(iii) (disallowing any rate seeking to recover losses from a prior period); pipeline managements will presumably aim at a pricing strategy that will, in fact, fully recover costs allocable to unbundled transportation. We cannot evaluate the rule on the basis of an assumption that they will not succeed. (We address below a claim *30that the rate provisions disable pipelines from full recovery of unbundled transportation costs.)
The claim of discrimination in favor of unbundled transportation contains a more subtle argument (or at least the seeds of such an argument): even though such rates may recover exactly the cost of service (just as for the transportation component of sales service), perhaps the flexibility afforded pipelines will in effect give unbundled transportation an advantage over sales service. The possibility is hardly one that we may rule out a priori. But we think it a problem that the Commission should be free to solve if and when it develops. As the Commission points out, the historical problem has been that unbundled transportation rate provisions put it at a disadvantage as against sales service. J.A. 318. No one appears to dispute that finding. It seems wholly suitable for the Commission to experiment with one rate structure in this specialized area; if it proves a triumphant success, the Commission will doubtless have opportunities to extend it to sales.
D. Selective as Opposed to Uniform Discounts.
Some parties accept the concept of price discounting but argue that the Commission should allow only “uniform” discounting (in effect requiring a pipeline to promulgate in advance the criteria under which it would provide discounts). The Commission, however, made the judgment that such a rule would unduly stifle discounting. J.A. 478-83. It saw substantial gains from such discounts: cheaper fuel supplies for the price-elastic customers receiving the discounts; reduced revenue short-falls for pipelines that would otherwise lose the business altogether; and protection for non-favored customers from rate increases that would ultimately occur if pipelines lost volume through inability to respond to competition. J.A. 483.
For much the same reasons that courts allow administrative agencies the leeway to choose between rulemaking and adjudication (variability of circumstances, difficulties of foresight), we think that the Commission was within its power to allow pipelines a parallel choice. But, just as courts insist on a degree of agency consistency, see, e.g., Local 32, American Federation of Gov’t Employees v. FLRA, 774 F.2d 498, 502 (D.C.Cir.1985), we expect that the Commission will exact from the pipelines as much consistency of application as is necessary for both to be in conformity with §§ 4 and 5.
E. Consistency of “Value-of-Service” Discounting with MPC II.
The American Public Gas Association and others contend that the general consent to selective discounting violates this court’s decision in Maryland People’s Counsel v. FERC ("MPC II”), 761 F.2d 780 (D.C.Cir.1985). The attack is directed especially to Commission suggestions — in supporting statements, not the rule itself— that discounting intended to meet competition from alternative fuels or indeed from other pipelines is not per se unduly discriminatory. J.A. 476.
Petitioners misconceive the scope of MPC II. Pipelines were using their market power in the transportation market to discriminate (indirectly) in the sale of gas, a commodity that Congress had concluded was produced under roughly competitive conditions. In the sale of such a commodity there is no economic justification for charging different prices based on the purchasers’ differing access to substitutes {i.e., their price elasticity of demand). Indeed, if a product is produced under competitive conditions, such price discrimination cannot occur unless a bottleneck with market power stands between it and the customers. By contrast, pipeline transportation service is marked by a degree of natural monopoly, J.A. 305-06, 352, 481 (i.e., longrun average costs decline in the relevant range of production). See 2 A. Kahn, The Economics of Regulation: Principles and Institutions 119-23 (1971). In such an industry, “value-of-service” ratemaking (i.e., rates varying on the basis of differing demand characteristics) has an *31established place,12 though not an uncontested one.13 The equitable argument in favor of such differentials is that they may benefit captive customers by making a contribution to fixed costs that otherwise would not be made at all. (The efficiency argument is that such differentials will raise total volume closer to the level it would attain if all sales were priced at marginal cost.)
These justifications were missing in MPC 11. There the court found that the then-existing blanket certificate regulations allowed pipelines to deny captive consumers access to the spot market for gas, while providing it for the non-captives. 761 F.2d at 788. This allowed pipelines to preserve the revenues attributable to transportation of gas to fuel-switchable customers, while continuing to sell their inventory of overpriced gas to captive customers. The Commission advanced an argument that the pipelines’ receipt of transportation revenues would redound to the benefit of captive customers — an argument that sounds like the one advanced above. The court said no. First, we said that the Commission had offered no reason to think that the captives could not enjoy the fuel switchables’ contribution to fixed costs even if the Commission conditioned the program on equal access for captive consumers — precisely what the Commission has done here. Id. Second, we pointed out that the Commission had nowhere answered the petitioners’ argument that the captives’ loss through lack of access to the wellhead market would greatly exceed their gain through the fuel-switchables’ contribution to fixed costs. Id. Here, of course, the Commission is providing access to the spot market. Thus the facts here obviate our two reasons for rejecting the Commission’s argument on contributions to fixed costs.
To read MPC II as a rule that price differentials based on demand conditions are always unduly discriminatory would render the decision a defiant and unreasoned exception to the general pattern. The judicial acceptance of such price differentials is longstanding. For nearly 100 years, for example, the courts have interpreted the antidiscrimination provisions of the Interstate Commerce Act to allow the ICC to approve differentials justified exclusively by competition. See, e.g., Texas & Pacific Ry. v. ICC, 162 U.S. 197, 218-19, 16 S.Ct. 666, 674-75, 40 L.Ed. 940 (1896); Dresser Industries, Inc. v. ICC, 714 F.2d 588 (5th Cir.1983) (review under three different anti-discrimination provisions); National Gypsum Co. v. United States, 353 F.Supp. 941, 946-49 (W.D.N.Y.1973) (enumerating cases following this view). Indeed, the Supreme Court has even struck down an ICC finding of unlawful discrimination where it appeared to be based on an absolute rule that competitive conditions could never justify a rate differential. Eastern-Central Motor Carriers Ass’n v. United States, 321 U.S. 194, 64 S.Ct. 499, 88 L.Ed. 668 (1944).
We have answered the claims that the rate provisions of Order No. 436 put it in violation of our mandate in MPC II. This is not to say, of course, that the Commission is free to uphold every price distinction based on different demand elasticities. It has long been contended, for example, that rate differentials based exclusively on competition between transporters with similar cost functions may end up forcing captive customers to bear disproportionate shares *32of fixed costs without any offsetting gain in efficiency. See, e.g., 1 A. Kahn, The Economics of Regulation: Principles and Institutions 159-81, esp. 170 (1970). The contention is not self-evidently true: if the demand of buyers with access to competing carriers is at all price elastic, the price reductions they enjoy will raise their demand close to competitive levels. In any event, the Commission may properly defer its ultimate resolution of these issues to another day and another proceeding. Cf. American Commercial Lines, Inc. v. Louisville & Nashville R.R., 392 U.S. 571, 88 S.Ct. 2105, 20 L.Ed.2d 1289 (1968) (finding broad discretion in ICC to choose format in which to resolve issues of price discounting in competition between railroads and barge-truck combinations).
F. Impact of Discounting on Pipeline Solvency.
Petitioners ANR Pipeline Company and Colorado Interstate Gas Company fault the regulations for allowing the pipeline to discount below the ceilings but never to charge more. To the Commission’s defense that the discounting mirrors the world of unregulated firms, they respond that in such a world the circumstances where market conditions force a firm to discount are likely to be matched by ones allowing the charge of a premium. (In equilibrium firms will earn a normal profit.) Here, they argue, the rules parallel only the downside of the unregulated market. As a result, they say, return will necessarily be less than in other industries with corresponding risks, in violation of FPC v. Hope Natural Gas Co., 320 U.S. 591, 603, 64 S.Ct. 281, 288, 88 L.Ed. 333 (1944).
We can imagine a rate methodology under which this contention would be sound. Suppose that a pipeline has a capacity for transporting 120,000 units a year, that each year’s share of fixed costs amounts to $90,-000, and that variable costs are $.10 per unit. In an initial rate case, the Commission projects-volume at 100,000 units, and thus sets a maximum price of $1.00 per unit ($.90 as a share of fixed costs and $.10 for variable costs).
While those rates are in effect, suppose the firm in fact carries 100,000 units at $1.00, but, spotting market opportunities, carries another 10,000 units at $.20 per unit for customers who would switch to alternative fuels if the transportation charge rose above $.25 per unit. (If the pipeline knew that $.25 was the switchover point, it would charge that, but it may not know exactly.)
In the next rate case, suppose the Commission projects use at 110,000 units, and accordingly sets the maximum price at $.92 per unit ($.10 for variable costs and $.82 ($90,000/110,000) for fixed costs). Such a rate would be sufficient to recover costs only if the pipeline carried 110,000 at the maximum rate; but the evidence overwhelmingly suggests that it will not be able to do so — the extra 10,000 units of business were due to the discount. Unless some change in circumstance saves the pipeline, revenue will be $94,000 ($92,000 for 100,000 units transported at the maximum rate plus $2,000 for 10,000 units at $.20), against costs of $101,000 ($90,000 fixed and $11,000 variable).
We see no reason, however, to suppose that the Commission intends such calculations. Its only statement relating to projections, 18 C.'F.R. § 284.7(c)(3), indicates the contrary:
The pipeline’s revenue requirement allocated to firm and interruptible services should be attained by providing the projected units of service in peak and off-peak periods at the maximum rate for each service.
In its commentary, the Commission pointed to this passage as proof of its agreement with MPC’s suggestion that “revenue projections in rate filings [should] assume that all sales and transportation volumes will be charged at the maximum rate.” J.A. 484. Thus, it appears that “rate” in § 284.7(c)(3) refers to the maximum unit price, not to projected throughput. This would appear to undermine any fear that the Commission might employ the dubious procedure hypothesized above.
*33Thus we find no legal defect m the rate provisions of Order No. 436.
IV. Contract Demand (“CD”) Adjustment
Local distribution companies require a firm supply of gas. Typically they have looked to pipeline sales service to fill this need. Firm sales contracts give the customer the right to demand, and obligate the pipeline at all times to stand ready to deliver, a certain quantity of gas per day, generally known in the industry as “Contract Demand” or “CD.” Once the arrangement receives the necessary certificate under § 7 of the NGA, the LDC’s entitlement and the pipeline’s obligation acquire a legal existence independent of the contract and persist until the Commission issues formal approval of “abandonment.” See California v. Southland Royalty Co., 436 U.S. 519, 98 S.Ct. 1955, 56 L.Ed.2d 505 (1978); Panhandle Eastern Pipe Line Co. v. Michigan Consolidated Gas Co., 177 F.2d 942, 945 (6th Cir.1949). For a “full requirements” customer, relying on a single pipeline, the CD will amount to its entire anticipated gas needs; “partial requirements” customers, as the name suggests, rely on more than one pipeline.
“Demand charges” are based on CD and are payable regardless of the customer’s actual use; having thus committed itself to partial payment for the gas covered by its CD, a customer pays only a “commodity charge” when it actually takes gas.14 Thus, if the demand charge is $1 and the commodity charge $3, the customer will switch to an alternative supply only when the alternative’s total cost (transportation and gas) is under $3, even though (in a sense) gas at $3.50 would be a better bargain. (It is better only “in a sense” because the customer gets a security of supply from its pipeline supplier that it does not get in the spot market.) The relation with a regular pipeline supplier thereby constrains the customer’s practical freedom to take advantage of open access to the wellhead market. The higher the CD in relation to its total usage, and the higher the demand charge as a proportion of total price, the more severe is the constraint.
To make the customers’ access meaningful, Order No. 436 provides customers a limited right to unilaterally modify their contracts with pipelines who elect to operate under the Order. It entitles any party with a firm sales contract on the date the pipeline becomes subject to Order No. 43615 to (a) convert specified percentages of its CD from gas purchase (i.e., fully bundled service) to unbundled gas transportation or (b) reduce its CD by the same percentages.
The entitlement to convert or reduce16 is 15% for each of the first and second years, 20% for the third year, and 25% for each of the fourth and fifth years. The entitlements are cumulative. For example, a firm customer could reduce contract demand not at all in the first two years and then reduce by 50% in the third year. Or it could refrain from exercising the option at all in the first four years, and then reduce by 100% in the fifth year or any year thereafter. See Order 436-C, J.A. 1355-60.
Petitioners attack the Commission’s CD adjustment program on several fronts. A number of pipelines assert that the conditions violate the rule of Panhandle Eastern Pipe Line Co. v. FERC, 613 F.2d 1120 (D.C.Cir.1979), cert. denied, 449 U.S. 889, 101 S.Ct. 247, 66 L.Ed.2d 115 (1980), restricting the use of the Commission’s power to impose conditions on certificates is*34sued under § 7. In addition, the pipelines attack the sufficiency of the Commission’s reasoning for adopting CD conversion and CD reduction. With respect to CD reduction a number of LDCs lend their voices to the attack. Finally, Maryland People’s Counsel contends that this court’s opinion in Maryland People’s Counsel v. FERC, 768 F.2d 450 (D.C.Cir.1985) (“MPC III”), compels the Commission to permit customers to convert 100% of their CD to transportation immediately.
A. Legal Authority.
1. Violation of Panhandle doctrine.
Several pipelines contend that the Commission’s creation of the CD conversion/reduction options violates the principle established by this court in Panhandle, supra. Panhandle had applied for certification under § 7(c) of new transportation service to an industrial gas user. Exercising its § 7(e) authority to attach “such reasonable terms and conditions as the public convenience and necessity may require,” the Commission approved the certification subject to a condition requiring Panhandle to “flow” the resulting revenues through to its wholesale gas customers. The condition effectively gave the latter a rate reduction equal to the new revenues. The Commission’s theory was that the wholesale gas customers had already paid for the capacity used to provide the service to the industrial user (or had obligated themselves to do so under existing rates). 613 F.2d at 1123. This court held, however, that such a use of the Commission’s § 7 conditioning authority was an illegal circumvention of § 5. The latter authorizes the Commission to modify pipeline rates and charges when it finds them “unjust, unreasonable, unduly discriminatory, or preferential,” but only after notice and hearing in which, it is well established, the Commission bears the burden of proof. See, e.g., Sea Robin Pipeline Co. v. FERC, 795 F.2d 182, 184 (D.C. Cir.1986).
In the present case, the Commission's creation of the CD conversion/reduction option similarly modifies previously approved arrangements that are separate from the transportation authority that a pipeline would seek under Order No. 436. On its face, it appears to challenge Panhandle’s strictures against extensions of the § 7(e) conditioning power that would erode substantive or procedural limits on the Commission's power.
The Commission has taken high ground, which we think quite untenable. It argues that since application for a blanket certificate under § 7 is entirely voluntary, there is no need for it to point to any express congressional grant of power. It asserts that in the CD modification conditions it was “not requiring the adjustment of previously-certificated service____” FERC Brief at 107 n. 1 (emphasis added). Obviously the suggestion that Panhandle presents no problem because the application for certification is voluntary is unacceptable. As all § 7 applications are voluntary in a legal sense, the Commission’s theory would extirpate the Panhandle doctrine.
Apart from the voluntariness theory, the Commission notes that it has invoked § 7(b), authorizing it to permit natural gas companies to abandon certificated service. In 18 C.F.R. § 284.10(f)(3) it expressly finds that pipeline abandonments of service, pursuant to customer elections under the Order, are “permitted by the present or future public convenience and necessity.” But this finding looks only to the pipeline’s obligation, as is fitting under § 7(b). Neither that section nor the finding thereunder supports the Commission’s relieving customers of their contract obligations.
For that, it appears one must turn to § 5 of the NGA, which allows the Commission to set aside any unjust, unreasonable or unduly discriminatory “contract affecting” rates and charges. Much of the Commission’s reasoning suggests a belief that under present circumstances inflexible CDs in fact qualify as “unjust, unreasonable, [or] unduly discriminatory” terms. But the Commission has expressly declined to rely on § 5, and has not explained why not. J.A. 1059.
*35The “voluntariness” theory and the invocation of § 7(b) appearing inadequate, and the Commission having disclaimed reliance on any other provision (notably § 5), the CD modifications are without basis in law insofar as they condition blanket certificate transportation under the NGA on release of customers from contract obligations. (Immediately below we address the issue of transportation under § 311.) On remand, the Commission can proceed under such grants of power as it believes are relevant, undistracted by the notion that its power under § 7(e) entitles it. to sweep aside the substantive and procedural constraints of the NGA. In so doing, of course it may employ rulemaking. Cf. Wisconsin Gas Co. v. FERC, 770 F.2d 1144 (D.C.Cir.1985), cert. denied, — U.S.-, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986).
CP National and other LDCs contend that the Commission is without authority to impose the CD conversion/reduction option as a condition of authority to transport under § 311 of the NGPA. No analysis is offered in support of the claim. The premises of the Panhandle doctrine are absent here. Perhaps because of its expectation that § 311 would operate simply to forge interstitial links between the hitherto separate interstate and intrastate markets, see Public Service Comm’n of the State of New York v. Mid-Louisiana Gas Co., 463 U.S. 319, 342, 103 S.Ct. 3024, 3037, 77 L.Ed.2d 668 (1983); Process Gas Consumers Group v. United States Dep’t of Agric., 694 F.2d 728, 764 (D.C.Cir.1981) (other portions vacated and reconsidered en banc, 694 F.2d 778 (D.C.Cir.1982)), cert. denied, 461 U.S. 905, 103 S.Ct. 1874, 76 L.Ed.2d 807 (1983), Congress never created for § 311 transportation any elaborate structure paralleling that of the NGA. This claim fails.
In summary, the Commission has failed to ground the CD adjustment provisions in any adequate section of the NGA. Because the conditions can legally apply to § 311 transportation, however, and because we have no reason to expect that the Commission’s interest in attaching the options to § 7 transportation has waned, we proceed to address a number of other attacks.
2. Alleged lack of compliance with § 7(b).
As noted above, the Commission has, in the CD modification provisions, identified circumstances under which pipelines are automatically entitled to abandonment of service — namely, when the customer exercises the election provided. In support of this it has made thé necessary finding under § 7(b) that such abandonment serves the “public convenience or necessity.” 17 CP National and others fault the Commission for failing to make various specific findings said to be subsidiary parts of that conclusion. They cite, for example, Transcontinental Gas Pipe Line Corp. v. FPC, 488 F.2d 1325, 1329-30 (D.C.Cir.1973), cert. denied, 417 U.S. 921, 94 S.Ct. 2629, 41 L.Ed.2d 226 (1974), overturning the Commission’s decision to permit certain producers to abandon sales to an interstate pipeline. In that context, the court held that the Commission must study various factors and make a “broadly conceived comparison of the needs of the two natural gas systems [the current purchaser and the producers' intended substitute] and the public markets they serve.” Id. at 1330 (footnote omitted).
The petitioners implicitly assert that the substantive ingredients of the “public convenience or necessity” are the same regardless of the type of abandonment. This makes no sense. Clearly the substantive concerns relevant to terminations at the option of LDCs are altogether different from those relating to producers’ abandonment of their sales to pipelines in a period of acute shortage, as was the case in Transcontinental.
*36Elizabethtown Gas Company derives from Transcontinental the proposition that the Commission cannot wholly defer to private parties’ choice. Id. at 1328-29. We see no conflict between that precept and the Commission’s action here: nothing in Transcontinental prevents the Commission from identifying circumstances which, when coupled with the purchaser’s election, satisfy the public convenience and necessity.
B. Adequacy of the Commission’s Reasoning in Support of CD Conversion.
Pipelines attack CD conversion as arbitrary and capricious, focusing mainly on its impact on preexisting supply arrangements. They entered into these arrangements primarily in response to their Commission-imposed obligations to maintain sources of supply adequate to meet their sales commitments,18 see 18 C.F.R. § 2.61, and in reliance on their ability to recoup the costs through firm sales contracts. The arrangements consist largely of long-term supply contracts with producers, but also include investments in expensive facilities for the importation of liquefied natural gas (“LNG”), see Trunkline LNG Co. (Opinion No. 796), 58 F.P.C. 726 (1977); Trunk-line LNG Co. (Opinion No. 796-A), 58 F.P.C. 2935 (1977).19 In addition to arguing that the rule defeats their justifiable reliance on their sales contracts, the pipelines argue that the Order is shortsighted: seeing their treatment in this situation, pipelines will hardly jump to meet any future shortage or come to the assistance of a converting LDC that later finds itself in trouble when the market tightens.
An assertion that agency conduct was arbitrary and capricious requires the court to explore the links between that conduct and the agency’s statutory authority. We must examine the agency’s reasoning to determine whether it considered the relevant factors and drew “a ‘rational connection between the facts found and the choice made.’” Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962)). Here we are hampered because, insofar as the Commission has attached the CD conditions to § 7 blanket certificate transportation, its asserted statutory basis, § 7(e), is legally insufficient. See supra part IV.A.l. However, in attaching the condition to § 311 transportation, the Commission plainly invokes the authority of § 311(c), which allows it to prescribe “terms and conditions.” Section 311 itself states no explicit standards for the exercise of the power, but the overall purposes of the NGPA provide .a standard — somewhat amorphous, to be sure— against which we can and must measure the Commission’s decision. See, e.g., Permian Basin Area Rate Cases, 390 U.S. 747, 776, 88 S.Ct. 1344, 1364, 20 L.Ed.2d 312 (1968). Given the overlap in the purposes of the NGA and the NGPA this process will have implications for possible future exercises of the Commission’s NGA authority, presumably § 5, but in view of the Commission’s disclaimer of reliance on § 5, our analysis does not directly apply to such an exercise.
Unilateral abrogation of a contract is an extreme measure. This is true even where the abrogation is partial, as it is under the conversion option, and even though common law doctrines of impracticability and impossibility shift the risk allocation nom*37inally arrived at by the parties. (They may well do so only to arrive at the allocation the parties would have made had they, considered in advance the risk that eventuated. Posner & Rosenfield, Impossibility and Related Doctrines in Contract Law: An Economic Analysis, 6 J. Legal Stud. 83 (1977).) Nonetheless, we find the reasoning underlying CD conversion persuasive.
Unlike the typical contract, those at issue here necessarily reflect the pipelines’ monopoly power. The Commission found the transportation network “highly monopolistic in some markets, fairly competitive in others.” J.A. 281. Historically, in fact, many customers have been served by only one pipeline. J.A. 279. This is not disputed. Until the recent partial unbundling of pipeline sales and transportation service, the pipelines were the only parties from whom LDCs might practicably buy gas. Once the unbundling of services began, the LDCs’ position changed little, as the pipelines wielded their monopoly power over transportation to deny them the ability to purchase from would-be competing suppliers. J.A. 318, 352 (finding practice “unduly discriminatory and preferential”). Absent these market restraints, there is no reason to believe that the LDCs would have agreed to the long-term sales contracts binding them to pay rates based on the pipelines’ costs, whatever those might be.20 Yet, by virtue of these arrangements, the LDCs found themselves denied access to the spot market, which offers prices at least 20% below the pipelines’ average gas costs.21 See supra part I.
FERC thus found that to remedy these effects, it was essential to permit limited LDC abrogation of pipeline sales contracts:
The transitional contract demand reduction and conversion options are essential if the goal of nondiscriminatory access to transportation is to be achieved____ With such an option, full-requirements customers — especially small, sole-supplied local distribution companies ...— will have access to competitively-priced supplies of the gas commodity. Thus, they will no longer be dependent on a single merchant for their gas supplies ____
J.A. 407-08. See also J.A. 424 (“only through conversions can sole-supplied customers have access ... to the national market”). Failing to provide such an option, FERC concluded, would be “to condone the fundamental form of undue discrimination by monopoly power which the NGA intended to prohibit.” J.A. 426.
Thus while the CD conversion option partially denies pipelines the benefits of their contracts, it does so only because those contracts are vestiges of their monopoly power, and only in order to correct the consequences of that power. This action therefore conforms to the purposes of the NGPA. In Transcontinental Gas Pipe Line Corp. v. State Oil & Gas Board, 474 U.S. 409, 106 S.Ct. 709, 88 L.Ed.2d 732 (1986), the Supreme Court declared that enactment of the NGPA left as the “aim of federal regulation ... to assure adequate supplies of natural gas at fair prices,” id., 106 S.Ct. at 716, and referred to Congress’s “determination that the supply, the demand, and the price of high cost gas [the type at issue there] be determined by market forces,” id. at 716-17. Congress’s continued concern for the market power of pipelines is expressly reflected in NGPA § 601(b)(1)(E), providing that the price of *38gas purchased by a pipeline at the wellhead from its affiliate is deemed just and reasonable only to the extent that it does not exceed independents’ prices in comparable sales.
Thus it would appear that the Commission has been correct in its belief that under § 311 it should assert “the traditional regulatory approach in areas where it is needed to protect the public from market dominance by natural gas companies.” J.A. 271. Provision of the CD conversion option for customers of pipelines offering § 311 transportation properly implements that view. Any principle quashing the Commission’s chosen remedy would seem to block pro-competitive regulatory reform and run counter to a long judicial tradition favoring agency development of whatever pro-competitive policies are consistent with the agency’s enabling act. See, e.g., Gulf States Utils. Co. v. FPC, 411 U.S. 747, 760, 93 S.Ct. 1870, 1878, 36 L.Ed.2d 635 (1973); Denver & Rio Grande W.R.R. Co. v. United States, 387 U.S. 485, 492-93, 87 S.Ct. 1754, 1759-60, 18 L.Ed.2d 905 (1967).
CD conversion may to a degree shift the costs of overpriced gas and take-or-pay liability to those pipeline customers least able to switch to reliance on the wellhead market. But circumstances limit the pipelines’ power to shift the costs. Any customer of a pipeline electing to provide transportation under Order No. 436 has, by definition, the power to go out into the market to secure gas and unbundled transportation. That strategy has its costs — including, of course, the management costs of negotiating and coordinating long-term supplies, or fees to brokers for doing so. But these costs form the ceiling on what pipelines may charge. Accordingly we see no basis for rejecting FERC’s conclusion that the cost-shifting risk was tolerable.
We accept FERC’s conclusion that the relevant factors under § 311 tilt in favor of the CD conversion option.
C. Adequacy of the Commission’s Reasoning in Support of CD Reduction.
Several LDCs attack the Commission’s authorization of CD reduction. They contend primarily that it failed to meet the “substantial evidence" requirements of § 19(b) of the NGA and § 506(a)(4) of the NGPA, which are understood by the parties to be equivalent, in the rulemaking context, to the “arbitrary and capricious” standard. See Wisconsin Gas Co. v. FERC, 770 F.2d 1144, 1156 (D.C.Cir.1985), cert. denied, — U.S.-, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986); Mid-Tex Elec. Coop., Inc. v. FERC, 773 F.2d 327, 338 (D.C.Cir.1985). This standard requires the agency to “articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.’ ” Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962)).
We agree with the challengers that the Commission has failed to develop an adequate rationale in support of CD reduction. Review of the Order on this point is considerably hampered by the Commission’s tendency to wrap its justification for CD reduction together with the case for CD conversion. The two almost always appear together, like Rosenkranz and Guildenstern, making it hard to extract the portion of the argument that has any real connection with reduction. This is understandable in terms of the proposals’ history. In the NOPR, the Commission offered CD reduction as the sole device by which customers could escape their long-term contracts with pipelines. When the Commission later recognized and embraced CD conversion, its analysis of CD reduction became largely obsolete. Yet it has continued in places to justify it as necessary to effect the goal of providing consumers access to competitively priced gas. E.g., J.A. 407-08. See also J.A. 1055-58. If CD conversion is available, however, this justification fails. Below we review various other purposes asserted by FERC, as well as its treatment of the cost-shifting consequences of CD reduction.
*39(a) FERC argues that “[t]he transitional contract demand reduction and conversion options are essential if the goal of non-discriminatory access to transportation is to be achieved when pipelines operate under the new transportation rules.” J.A. 407. In a later passage, rejecting a proposal that it afford only CD conversion, the Commission reasoned that a CD option limited in such a way would restrict customers’ access to “transportation services on the same pipeline” and deny them the ability to shop around for unused transportation capacity booked on other pipelines. J.A. 448.
As justifications for CD reduction, these arguments seem peripheral to the problem the Commission set out to solve in this rulemaking: access to gas competitively priced at the wellhead. CD reduction would of course help each LDC secure access to all the different producing areas of the country, in addition to the ones from which its existing pipeline supplier(s) draw their gas. But the record contains no suggestion that competitive wellhead prices are subject to important regional variations. While easy LDC access to all producing regions would help correct or prevent regional price variations, the Commission makes no argument that any such variations pose so great a problem as to require such drastic action as 100% CD reduction.
Of course, competition among pipelines in transportation services may well be expected to engender lower prices (at any given level of service quality). If so, it would fulfill the general consumer-benefit purposes of the Order. But the Commission does not spell out any such arguments. Moreover, any analysis along these lines would trigger attempted rebuttals, essentially based on the view that in a monopolistic or oligopolistic industry unrestricted consumer choice may lead to duplication of capacity and higher costs for consumers. Without assessing the validity of those arguments or the Commission’s authority to adopt rules moving the industry towards more competition in transportation, we simply cannot find any Commission effort to justify CD reduction in such terms.
(b) The Commission argues that the levels of sales service that customers have contracted for on a firm basis “may no longer correspond to what they desire to purchase.” J.A. 406 (footnote omitted). The Commission found that this imbalance between actual needs and contracted-for capacity has caused at least two problems. First, the discrepancy results in an undesirable inequity as “on some systems, interruptible transportation is as a practical matter virtually the same quality of service as firm,” though available at lower rates. Id. (emphasis in original). Second, the discrepancy also has the effect of unnecessarily denying firm service to some potential users, making the reduction option necessary “to allow a freeing up of firm capacity.” J.A. 411. See also J.A. 417.
While the argument seems highly relevant to CD reduction, it hardly supports the broad remedy adopted. Even in its terms, it refers to a limited portion of the industry. The finding’s lack of general application is underscored by the Commission’s observations that most firm sales customers need their full contract demand on peak days. J.A. 420. If so, then the obsolescence referred to is clearly far from universal. If such obsolete certificates exist only “on some systems,” it is unclear why the Commission believes that an industry-wide solution is needed, especially one that permits all LDCs — or rather, all LDCs in their relations with pipelines opting to offer service under Order No. 436 — a right to reduce contract demand 100%.
The Commission argues that Wisconsin Gas Co. v. FERC, 770 F.2d 1144 (D.C.Cir.1985), cert. denied, — U.S.-, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986), allows it to make § 5 determinations generically. J.A. 412. True but irrelevant. Neither Wisconsin Gas nor any other case of which we are aware supports an industry-wide solution for a problem that exists only in isolated pockets. In such a case, the disproportion of remedy to ailment would, at least at some point, become arbitrary and capricious. This is not to say, of course, that the Commission could not use generic rules to identify a limited class of LDCs to be *40entitled to reduce CD when special conditions are present. But here the Commission has said nothing to link the asserted obsolescence of CD levels to the broad class of purchasers made eligible.
(c) Responding to the complaints of some LDCs that the CD reduction option will force them to bear an additional share of pipelines’ capital costs (i.e., the costs now borne by the customers that will exercise the option), the Commission asserts that customers are likely to seek little net CD reduction. J.A. 420. Accordingly there will be little net change, nationwide, in aggregate recovery of costs. Id.
The improbability of major aggregate cost-shifting, however, provides little answer to the concerns of captive customers of pipelines that are likely to lose out in the competitive race. While the Commission might justify the losses of such a captive by reference to potential aggregate gains, it has neither confronted the problem nor developed in any detail its reasons to expect the net gains.
D. Insufficiency Under MPC III.
Maryland People’s Counsel argues that under the principle established in Maryland People’s Counsel v. FERC, 768 F.2d 450 (D.C.Cir.1985) (“MPC III’’), the Commission was required to afford firm customer an immediate option to convert any purchase obligation completely.22 We find the claim unpersuasive.
In MPC III this court addressed the validity of certain “special marketing programs” (“SMPs”), successors to the SMPs held invalid in Maryland People’s Counsel v. FERC, 761 F.2d 768 (D.C.Cir.1985) (“MPC I”). The SMPs reviewed in MPC I authorized the following approach to the mounting problem of high-cost gas subject to high take-or-pay burdens: A producer would resell, at market rates, high-cost gas previously committed to a pipeline. A limited class of persons was eligible to purchase, and that class excluded captive consumers. The producer would credit the pipeline’s take-or-pay liability for the sale, and the pipeline would transport the gas to the new purchaser. This court held the program invalid because it excluded captive consumers from the class of eligible new purchasers, without adequate consideration of possible anti-competitive and anti-consumer consequences. FERC then amended the program to permit anyone with a firm contractual entitlement to purchase a pipeline’s gas to nominate up to 10% of its contract demand to be purchased under the SMP. Thus the second-generation SMPs gave “captive” customers access to gas at competitive wellhead prices, but only up to the 10% figure. This court overturned the second-generation SMPs, finding them to be “of a piece with” the first. MPC III, 768 F.2d at 455.
We think that MPC finds more in MPC III than it contains. The Commission there made no effort to justify the 10% option in the second-generation SMPs by reference to classical “grandfathering” values. Here, by contrast, it invokes those values emphatically and we think plausibly, arguing that the pipelines require time to adjust to the new dispensation, particularly to resolve the take-or-pay problems presented by the producer-pipeline contracts. J.A. 1152. The specific phase-in period and percentages and election procedures resulted from detailed consideration of various options.23 See J.A. 1145-55. In the event that the Commission modifies its disposition of the producer-pipeline contracts in light of our treatment of the take-or-pay issue or for any other reason, however, the *41Commission may wish to reconsider the phase-in of the conversion option.
V. Producer-Pipeline Contracts
As noted above, certain contracts entered into by producers and pipelines between 1977 and 1982 have been at the root of the Commission’s endeavor in Order No. 436. These problem contracts provide for prices far in excess of current market levels. They contain take-or-pay clauses requiring the pipelines either to purchase a specified percentage of the producer’s deliverable gas or to make “prepayments” for that percentage anyway. The contracts typically permit the buyer to recoup prepayments by applying those amounts to subsequent “takes” of gas occurring within a limited period after prepayment. See 18 C.F.R. § 154.103 (mandating that purchase contracts for gas to be transported in interstate commerce carry a minimum five-year make-up period).
Although all parties refer to the problem posed by these contracts as the issue of take-or-pay, it is the combination of high prices with take-or-pay clauses that causes the difficulty. A 100% take-or-pay contract for gas priced at $1.00 per Mcf would usually cause the purchaser no trouble; gas purchased at the wellhead for $5.00 per Mcf, however, cannot be resold at that price (plus normal transportation mark-ups).
Between 1977 and 1982 the pipelines “rolled in” high-priced gas with low-priced gas (the latter largely due to wellhead ceilings imposed under Phillips Petroleum Co. v. Wisconsin, 347 U.S. 672, 74 S.Ct. 794, 98 L.Ed. 1035 (1954)), selling at an average price that was competitive with alternative fuels. But drops in the price of the latter exposed the pipelines to intense market pressure. This was soon reflected in reductions in the wellhead price for newly available gas {i.e., gas not hitherto subject to contract). Pipelines reduced their takes of high-priced gas, and in 1983 prepayment liabilities for the period between 1982 and 1985 were predicted to reach $7 billion. J.A. 301 & n. 34.
At the heart of the industry’s immediate problem is the discrepancy between the average cost of gas that pipelines have under contract and the much lower price of gas now available at the wellhead. The essence of that discrepancy is the same whether the pipelines buy over-priced gas and sell it at a loss, or decline to buy such gas and thereby incur take-or-pay liabilities. The price discrepancy represents a sunk loss of billions of dollars (doubtless reflected in actual drilling expenses). At issue among the parties is who should bear it. All actors in the natural gas industry— producers, pipelines, LDCs and consumers — are candidates for this dismal position. There is one exception: fuel-switchable users, who can employ the cheapest fuel competing with gas and thus cannot be induced to pay more than the current competitive price.
By enabling pipeline customers (and some end users) to obtain gas at current wellhead prices and thus escape the supraeompetitive contract prices, Order No. 436 appears to relieve consumers from the threat. Conversely, it heightens the likelihood that pipelines will play the fall guys. Considered in this section of the opinion is whether FERC’s lack of direct action as to the uneconomic contracts is permissible, particularly in light of Order No. 436’s shift in the balance of forces.
A. The Commission’s Prior Activity and Its Inactivity in This Proceeding.
In April 1985, shortly before issuing the Notice of Proposed Rulemaking (the “NOPR”) that culminated in Order No. 436, the Commission issued a policy statement on the regulatory treatment that it would give payments made by pipelines to extinguish take-or-pay liabilities (referred to as “buy-out” payments). Regulatory Treatment of Payments Made in Lieu ofTake- or-Pay Obligations, 50 Fed.Reg. 16,076 (1985) (codified at 18 C.F.R. § 2.76). The policy statement provides: (a) take-or-pay buy-out payments are not counted toward NGPA price ceilings (i.e., contracts to buy gas at NGPA ceilings do not breach the ceilings when the pipeline pays the producer in exchange for relief from take-or-pay *42obligations); (b) pipelines may file to include buy-out payments in their rate bases; (c) the method of cost recovery and the allocation among customers will be determined on a case-specific basis; (d) customers will maintain their NGA § 4 right to question the prudence and apportionment of buy-out payments; and (e) where the take-or-pay buy-out covers “jurisdictional” gas,24 FERC will handle requests for the certificate amendments or abandonments terminating the producer’s legal obligation with respect to such gas on an expedited basis.
Subsection (a) removes one potential difficulty to take-or-pay settlement (the price ceiling issue) and subsection (b) holds out to the pipeline the possibility of recovering the cost. Subsection (e) promises a reduction in red tape. Otherwise the policy is pretty noncommittal, and, more important, does nothing whatever to prevent the cost from flowing downstream to consumers.
The NOPR proposed more drastic action. It would have created “a rebuttable ‘safe harbor’ presumption of prudence for certain one-time payments made to extinguish all minimum payment or purchase obligations in certain qualifying contracts.” J.A. 529. Qualification would require that the buy-out payment not exceed some percentage of the take-or-pay liability discharged. (The Commission did not specify the percentage in the NOPR. See J.A. 536-37.) The responses to this proposal were overwhelmingly negative and reflected a fear that the safe harbor would merely establish a floor for take-or-pay settlements. J.A. 529-40. FERC regarded these complaints as valid and withdrew the safe harbor proposal.
FERC also considered a number of alternate proposals for dealing with the take-or-pay problem. Most prominent of these were proposals that FERC (a) directly invoke its power under § 5 to set aside the take-or-pay clauses of contracts for jurisdictional gas,25 see J.A. 330-33, 1119-21, or (b) condition producer access to Order No. 436 transportation on the granting of take- or-pay relief, J.Á. 381-83, 1065-74.26 *43FERC rejected these proposals. Concluding that further regulatory action on problem contracts was neither necessary nor desirable, it made only a trivial adjustment in the April 1985 policy statement.27
Virtually all parties other than producers attack FERC’s refusal to directly address the producer-pipeline contracts. The essential thesis is this: (1) Order No. 436 denies pipelines much of their leverage over producers — the threat to refuse a producer transportation of new gas when the producer refuses to compromise liabilities under old contracts — at excessive prices. (2) Many pipeline customers will take advantage of open access and CD conversion to escape from dependence on their pipeline suppliers. (3) The escapes will have a spiralling effect — as each additional LDC drops bundled service the gas cost burden will grow, driving still others off. (4) The only LDCs who will remain as pipeline sales customers will be the least nimble— those for whom it is most costly to develop secure supplies from non-pipeline sources. (5) As a result, the consumers who purchase from these LDCs will be stuck with the burden of the overpriced gas, thus defeating the purpose of the Order and violating the consumer-protective purposes of the NGA. See, e.g., Atlantic Ref. Co. v. Public Serv. Comm’n, 360 U.S. 378, 388, 79 S.Ct. 1246, 1253, 3 L.Ed.2d 1312 (1959); FPC v. Hope Natural Gas Co., 320 U.S. 591, 610, 64 S.Ct. 281, 291, 88 L.Ed. 333 (1944). Further, FERC’s inaction will permanently distort the structure of the natural gas market: by creating an artificial advantage for unbundled transportation service, it will cause the pipelines’ merchant role to atrophy, despite the greater efficiency of bundled service. (The greater efficiency derives from pipelines’ incurring lower transaction costs in providing a full range of service, including load-balancing service, than would any non-pipeline entity.)
We conclude that FERC’s decision to do nothing more than reaffirm the April 1985 policy statement reflects questionable legal premises and fails to meet the requirement of “reasoned decisionmaking.” Accordingly, we reverse and remand for further proceedings.
B. Analysis of FERC’s Decision.
FERC made several contentions in support of its decision against further direct action on the uneconomic contracts. First, it determined that further action was not necessary because Order No. 436 would not cause take-or-pay liabilities to increase, or if it did, the increased liabilities could be shuffled from the pipelines to consumers. Second, FERC invoked a number of policy reasons generally militating against regulatory relief for pipelines. Third, FERC found flaws with the leading alternative proposals advanced by participants in the rulemaking, namely that it use its § 5 power to modify contracts involving jurisdictional gas, and that it condition producer access on the granting of some measure of take-or-pay relief (or allow the pipelines to impose such conditions). We address each of these findings in turn.
1. Lack of need for additional steps.
a. Likely effects of Order No. 436 on take-or-pay build-up. FERC accepted an industry estimate of about $7 billion in take-or-pay obligations, J.A. 301-02, though noting the much smaller sum included in pipeline rate bases, J.A. 831, 1069-70. But it made no detailed evaluation of what proportion of these payments — or ones to be made in the future — the pipelines could recoup by later “takes” of gas. It found that pipelines had been able to buy-out substantial portions of their liabilities for about 20<p on the dollar. J.A. 540. It concluded that these prices were reasonable and that buyouts were sufficiently widespread to represent a reasonable solution to the problem.
*44FERC also offered several arguments challenging petitioners’ death-spiral scenario for pipeline sales service.
(a) As to the pipelines’ loss of a bargaining chip, FERC observes that the nondiscriminatory access and CD reduction/conversion conditions “are not intended to affect such renegotiations or litigation.” J.A. 1070. But FERC’s intent is not at issue. What is in dispute is the likely consequence of its acts. On that, FERC offers nothing to undermine the challengers’ inherently plausible suggestion that these conditions will have an adverse impact on the pipelines’ take-or-pay problems.
FERC also alludes to the “voluntary” character of pipeline provision of Order No. 436 transportation. E.g., J.A. 1072. There are two flaws in this. First, refusal of the option may spell bankruptcy: inability to provide blanket-certificate transportation for fuel-switchable users may in current market circumstances cause critical load loss. Of course acceptance of the option may also be fatal. But when a condemned man is given the choice between the noose and the firing squad, we do not ordinarily say that he has “voluntarily” chosen to be hanged.
Second, the argument obscures distinctions between pipelines in the aggregate and alone. To be sure, Order No. 436 gives pipelines an option, blanket-certificate transportation, which as a result of this court’s decision in Maryland People's Counsel v. FERC, 761 F.2d 780 (1985) (“MPC II”), is not available outside of Order No. 436. But as soon as a single pipeline finds it attractive enough to accept, each competing pipeline will come under competitive pressure to match the first's flexibility. Thus even if only one pipeline actually preferred to use Order No. 436 (hanging rather than being shot), competition might force others — ultimately perhaps all the others — to switch their preference. Thus the Order effectively reduces pipeline ability to face down recalcitrant producers.
(b) FERC argues that the CD conversion/reduction provisions may not injure the pipelines at all.
[T]he Commission considers it more likely that [the LDCs] will either convert to firm transportation on the same pipeline, or else free up underutilized capacity under uneconomic CDs for use by other customers on the same pipeline. In either case, the pipeline may actually increase throughput and, therefore, gain the net benefits of spreading its fixed costs over greater units of gas service.
J.A. 1068-69.
But customers’ conversion to transportation will clearly aggravate a pipeline’s ability to resolve the problem of its overpriced gas inventory. Nor can the Order’s CD reduction provision be painless: a pipeline would voluntarily agree to release an LDC whenever it had equally profitable business opportunities for the pipeline capacity thus made available; accordingly, the only cases where Order No. 436 causes CD reduction will be ones where the reduction does injure the pipeline. FERC’s analysis provides no reason to suppose that those instances will be negligible.
(c) FERC contends that under competitive pressures a pipeline “will seek to adjust its gas purchasing practices so as to lower its weighted average cost of gas to all customers.” J.A. 1069 (emphasis in original). This reasoning assumes away the problem of the uneconomical contracts to which pipelines are presently bound. All FERC does here is to admit that Order No. 436 dramatically increases the consequences of not getting out from under an uneconomical contract. It seems to confuse the pipelines’ incentives to renegotiate contracts with their ability to do so.
(d) FERC suggests that “[a] pipeline may determine that certain of its existing business arrangements make participation in the new rules unwise or undesirable,” and suggests possible alteration of the pipeline’s “leveraged capital structure” and “the extent of its non-pipeline business interests.” J.A. 1067. To the extent that this implicitly recommends some sort of spin-off of assets or borrowing increase, preparatory to bankruptcy, it seems unrealistic in light of market conditions, the rules *45against transfers in fraud of creditors, and “pierce-the-corporate-veil” principles.
(e) FERC finds two affirmative advantages for pipelines in Order No. 436. First, the Order affords producers a procedure for expedited “abandonment” (i.e., extinction of the legal commitment of the gas to the seller named in the producer’s certificate under the NGA) where a take-or-pay settlement has been reached or where a producer experiences substantially reduced takes without payment. 18 C.F.R. §§ 2.76(e), 2.77. In the latter case, it argues, abandonment will enable the pipeline to invoke contract defenses against a producer that resells the gas, specifically mitigation of damages. J.A. 1071-72. It is unclear how useful this may be, for the pipelines’ whole problem is the excess of the contract price over the current market price. That differential is the source of pipeline vulnerability, and expedited abandonment does little or nothing to cure it.28
Second, the Commission suggests that Order No. 436 might aid the resolution of take-or-pay problems by not precluding
any gas shipper, whether end-user, LDC, or independent broker, from negotiating on its own behalf & condition in a “self-help” gas sales contract which would require the producer-seller to grant take- or-pay relief to the shipper’s pipeline supplier before the shipper takes delivery of the gas.
J.A. 1073-74 (emphasis in original). Assuming FERC will stand by the underscored phrase, the scenario seems most unlikely. No reason appears why any of those persons would, in its bargaining with a supplier, sacrifice anything of value in order to secure a benefit for the pipeline. (Nor would the supplier agree to it without securing something in exchange.)
In sum, FERC's seeming blindness to the possible impact of Order No. 436 on take- or-pay liability, and its tendency to elevate into affirmative benefits what are at best palliatives, seem impossible to square with the requirement of reasoned decisionmaking.
b. Pipeline ability to shift costs downstream. The Commission suggests that pipeline complaints about Order No. 436 are undercut by its having reserved the ability to shift costs downstream to customers.
To the extent a pipeline in fact incurs additional take-or-pay liability due to transportation services under Order No. 436, nothing in Order No. 436 is intended to preclude the Commission from determining the appropriate allocation of revenue responsibility among the pipeline and its customers in an individual rate case filed to reflect the adjustments in sales and transportation service.
J.A. 1070-71.
While it is unclear what weight the Commission attaches to this possibility, it seems important to call to mind two legal limits on the Commission. On one hand, simply moving costs downstream to customers must at some point conflict with the Commission’s duty to “adequately attend [ ] to the agency's prime constituency — the consumers whom the [NGA] was designed ‘to protect ... against exploitation at the hands of natural gas companies.’ ” MPC II, 761 F.2d at 781 (quoting FPC v. Hope Natural Gas Co., 320 U.S. 591, 610, 64 S.Ct. 281, 291, 88 L.Ed. 333 (1944)). On the other hand, the Commission’s power to directly restrict pipelines from passing costs through is limited: it can limit passthrough of gas purchase costs made in compliance with the NGPA only if the payments arose out of “fraud or abuse,” NGPA § 601(c)(2), 15 U.S.C. § 3431(c)(2) (1982), and it can limit passthrough of other costs (including, under the Commission’s view, take-or-pay settlement costs), only to the extent not “prudently” incurred, see 18 C.F.R. § 2.76(d).
Despite these constraints on the Commission's power to limit passthrough by «te*46cree, it has considerable ability to protect consumers by bringing about market conditions that prevent a pipeline from passing costs forward. The NGPA’s legal limits on restricting passthrough clearly do not bar rules tending to generate such market conditions. Cf. supra part IV.B. Indeed, that is the principle underlying Order No. 436. At least two other factors, however, restrict the Commission’s room for maneuver here. (1) To the extent that the Commission allows gas costs in a pipeline’s rate base for transportation purposes, a possibility that it seems to leave open, J.A. 423, the costs will flow through the customers despite open access. (2) To the extent that it allocates large take-or-pay liabilities exclusively to gas costs, it seems likely to eradicate the pipelines as merchants, for under such a rule all customers would convert their entire contract demand, except for a limited class of customers: those who find it unusually costly to arrange their own nonpipeline sources of supply or who are pressured by state regulatory commissions to adhere to reliance on pipeline supplies.
In sum, the Commission seems to exaggerate passthrough to customers as a solution to the problem of uneconomic contracts. If Order No. 436 works as intended, economic constraints will make pass-through impossible, except for customers— likely not accounting for a large proportion of the interstates’ total gas sales — that are exceptionally sclerotic about purchasing gas from brokers or producers.
2. Policy considerations militating against any intervention.
In refusing to modify producer/pipeline contracts, FERC noted that such action would “raise extremely serious questions regarding the ability of private parties in the gas production industry to rely on private contracts as a tool for structuring basic economic relationships.” J.A. 332. In its brief, moreover, FERC invokes the need to be sensitive to the congressional policy decision, inherent in the NGPA, “to move toward a deregulated gas commodity market.” FERC Brief at 130-31. As the Supreme Court has recently stated, “To the extent that Congress denied FERC the power to regulate affirmatively particular aspects of the first sale of gas, it did so because it wanted to leave determination of supply and first-sale price to the market.” Transcontinental Gas Pipe Line Corp. v. State Oil & Gas Board, 474 U.S. 409, 106 S.Ct. 709, 717, 88 L.Ed.2d 732 (1986).
The Commission also invoked the closely related policy of holding pipelines accountable for their decisions in order to induce them to act more in the manner of firms in a competitive industry. J.A. 44-48. Although the context was a discussion of risk allocation as between pipelines and their customers, FERC’s reliance on it in that context underscores FERC’s commitment to the policy of nurturing a competitive wellhead market.
In essence FERC argues that the pipelines’ subjection to regulation is hardly, in itself, a reason why they should be able to escape contractual liability more readily than unregulated firms. Like those firms, pipelines are able to invoke the doctrines of impossibility and impracticability and to assert defenses under force, majeure clauses.29 They may also be able to persuade their suppliers to accept contract adjustments by appealing to the latters’ long-term business interests. These steps would be the only recourse open to an ordinary middleman in a competitive market who had entered into unlucky or improvident contracts. FERC suggests that pipelines should not enjoy an extra means of escape.
FERC invokes the same principle in distinguishing producer-pipeline contracts from those between pipelines and LDCs, which it proposes forcibly to adjust pursuant to the CD conversion and reduction options. First, FERC notes that, unlike a provision granting pipelines relief from uneconomical supply contracts, the CD con*47version/reduction option is essential if Order No. 436 is to carry out its fundamental purpose of providing all parties meaningful access to low-cost gas. See J.A. 414-25, 1119-21. See also Maryland People’s Counsel v. FERC, 761 F.2d 768 (D.C.Cir.1985) (“MPC I”); Maryland People’s Counsel v. FERC, 761 F.2d 780 (D.C.Cir.1985) (“MPC II’’); Maryland People’s Counsel v. FERC, 768 F.2d 450 (D.C.Cir.1985) (“MPC III"). Second, FERC highlights the fact that the troublesome pipeline-producer contracts are largely “straightforward commercial agreements” while pipeline-consumer contracts are heavily regulated “utility-type ‘service agreements.’ ” J.A. 1120.
FERC’s policy arguments in favor of subjecting pipelines to the pressures of a competitive market seem powerful and well grounded in the statutes it is authorized to enforce. Lest the Commission on remand place undue weight upon them, however, we must note some limits. Most important, producers’ access to transportation under Order No. 436 is itself dependent on government intervention. To condition that access on some producer cooperation in disposing of outstanding contracts is hardly identical to raw governmental abrogation of the contracts. Second, while the middleman buying and selling in competitive markets will bear the risk associated with his bad luck or improvidence, the pipeline’s market power allows it to pass some of that burden forward. As noted above, some portion of that ability will remain despite “open access” under Order No. 436 and despite FERC’s view that passthrough of buy-out costs is permissible only where they have been prudently incurred. Third, the pipelines have been caught in an unusual transition. They entered into the now uneconomic contracts in an era when government officials, berated pipeline management for failures of supply and constantly predicted continuing energy price escalations. Moreover, as sales and transportation were then wholly bundled, and unregulated pipeline gas trading affiliates were unknown, there was no way that a pipeline could generate direct profits on the gas-trading component of its business. Thus, their being abruptly and retroactively subjected to the downside risk is at least jarring. See Carpenter, Jacoby & Wright, Adapting to Change in Natural Gas Markets, in Energy, Markets & Regulation: Essays in Honor of M.A. Adelman 1 (1986) (tracing evolution of natural gas pipelines’ exposure to risk). For the reasons already noted, see supra part IV.B., the resulting equities are not enough to block necessary pro-competitive reforms. But they are not weightless.
3. Reasons for rejection of specific proposals.
a. Section 5 action against jurisdictional contracts. In rejecting the proposal that it invoke its § 5 power directly to set aside jurisdictional contracts with troublesome take-or-pay provisions,30 the Commission reasoned that because such a remedy would be incomplete, its use would be both inequitable and inefficacious. As a result of the NGPA, the proportion of wellhead sales that are jurisdictional is steadily declining. “It would be inequitable,” the Commission said, “to allow pipelines to reduce their takes only under contracts still subject to the Commission’s NGA jurisdiction, particularly because many of the problem contracts are not subject to the Commission’s jurisdiction.” J.A. 1121. If in fact jurisdictional contracts account for only a small proportion of the troubling price discrepancy, we can well understand the Commission’s viewing the invocation of § 5 as unpromising — at least unpromising enough to justify its proceeding with open access and deferring the abrogation approach. Elsewhere, however, the Commission states that “the bulk of high-cost ‘problem’ contracts without ‘market-outs’ may be those entered into prior to 1982 involving offshore gas regulated under NGPA section 102 \i.e., jurisdictional].” J.A. 1072. See NGPA § 601(a)(1)(B), 15 *48U.S.C. § 3431(a)(1)(B) (1982) (preserving NGA jurisdiction over gas from Outer Continental Shelf). On remand FERC must make clear the extent to which contracts for jurisdictional gas account for the basic economic problem.
In addition to the ambiguity in the factual underpinning of FERC's argument, we have some difficulty with the claim of inequity. In the NGPA Congress selected a particular device for bringing about a transition from the Phillips world, with all wellhead interstate sales for resale subject to Commission jurisdiction, to a quite different world, with only Outer Continental Shelf wellhead sales jurisdictional. The device was primarily one of perfectly standard grandfathering. Whatever inequity may flow from some producers suffering the consequence of falling on the wrong side of the line, while others fall on the right side, appears to be no more than the usual result of grandfathering.
The above discussion may not accurately capture the Commission’s concern. It has, rightly we think {see supra pp. 1026-27), placed great weight on the congressional determination that market forces should operate freely at the wellhead. Even for jurisdictional contracts, moreover, Congress has provided that prices at or below the NGPA ceilings are “just and reasonable.” NGPA § 601(b), 15 U.S.C. § 3431(b) (1982). Thus FERC is clearly barred from setting sub-NGPA ceilings on jurisdictional wellhead sales. Assuming that the Commission may find specific take-or-pay percentages to violate § 5, see Columbia Gas Transmission Corp., Opinion No. 204, 26 F.E.R.C. ¶ 61,034, at 61,120 (1984), aff'd in part and reversed in part sub nom. Office of Consumers’ Counsel v. FERC, 783 F.2d 206, 229 (D.C.Cir.1986), it may well fear that action against take-or-pay provisions without regard to price would be both ineffective and inequitable: ineffective because it would not reach many contracts that are deleterious only because very high price provisions combine with otherwise harmless take-or-pay clauses; inequitable because it would restrict contracts that are harmless despite high take-or-pay provisions.
The Commission has not expressed the meaning of its reasons for inaction on jurisdictional contracts clearly enough for us to determine the legality of its analysis. Accordingly, it should on remand reassess its refusal to act under § 5.31
b. Conditioning producer access. In refusing to condition producer access on cooperation in solving take-or-pay issues (or to allow pipelines to impose such a condition), FERC asserts that it would be unduly discriminatory for a pipeline to refuse transportation service under Order No. 436 “merely because the producer and pipeline cannot agree on take-or-pay relief in another contract.” J.A. 1073. It believes that such a refusal would discriminate not only against the producer-seller but also against the would-be buyer to whom the pipeline would have carried the gas but for the proposed condition. Id.
Insofar as the argument simply states that such a refusal would violate the anti-discrimination concept manifested in Order No. 436 itself, it obviously provides no justification for FERC’s having chosen to so delineate its anti-discrimination concept. Insofar as it interprets § 5’s prohibition of “undue discrimination,” the analysis seems very unpersuasive. Given that the Commission itself has identified the producer-pipeline contracts as a primary cause of the problem that Order No. 436 is intended to cure, J.A. 315, it eludes us why pipeline denial of access to producers that stand on the letter of their contract rights should be viewed as unduly discriminatory. It is clear that § 5 does not prohibit every distinction. See Saint Michaels Utilities Comm’n v. FPC, 377 F.2d 912 (4th Cir. 1967); cf. Texas & Pac. Ry. v. ICC, 162 U.S. 197, 218-19, 16 S.Ct. 666, 674-75, 40 L.Ed. 940 (1896) (one of long line of cases finding Interstate Commerce Act’s prohibi*49tion of discrimination not to bar every rate differential). FERC offers no reason why it views this one as beyond the pale.
The Commission’s response may arise out of a concern that any such condition would give pipelines a veto power over transportation of each producer’s gas. We think such a fear underestimates the Commission’s power to impose both procedural and substantive limits on pipeline use of any conditioning power. First, it could obviously require that the pipeline specify the allegedly offending contracts, thus preventing pipelines from transforming the condition into one allowing them unlimited discretion. Second, without even addressing the nuances of which price and percentage terms are truly objectionable, it could limit the condition rule to disputes over contracts entered into before some date; it could select the date, for example, as of which it had become reasonably clear that the NGPA’s premise — the notion that the market price of gas would never fall below NGPA ceilings — had ceased to be valid. Third, it could explore the possibility of developing rules that would identify the price and take-or-pay provisions that unduly thwart the Commission’s purpose in Order No. 436 and its duty under the NGA to protect consumers.
We do not mean to suggest that the Commission need consider any of the options discussed. See Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Ins. Co., 463 U.S. 29, 50-51, 103 S.Ct. 2856, 2870-71, 77 L.Ed.2d 443 (1983). Our point is simply that the Commission seems not to have made much of a case against a condition of the sort proposed. Given the serious threat to Order No. 436’s fulfillment of the Commission’s statutory goals, we think that rejection of such a proposal requires more explanation than the Commission has offered.
Finally, the suggestion that any access-conditioning rule would unduly discriminate against customers seems to us at best ill-developed. Generally speaking, one might expect customers and consumers to be unanimous in favoring any exercise of FERC power that would help to eradicate the supra-market contracts. Many customers and customer representatives join the brief demanding greater action on take-or-pay; no gas purchaser opposes it. Of course a purchaser that arranged for a bargain purchase of gas would be disappointed to see the deal founder on the producer’s take-or-pay recalcitrance. But, recognizing the reasons underlying the barrier, it would be likely to accept the occasional burden as a just one.
Some producers support FERC’s rejection of access-conditioning by invoking the concept that with such a condition the Commission would be doing indirectly what it is forbidden to do directly — i.e., regulate wellhead prices. The Commission itself at one time appeared to make an essentially similar argument, when it said “there is a serious question as to whether the Commission has the authority to require a nonjurisdictional party to give up a valid contractual right as a condition to obtaining service.” J.A. 381. See also J.A. 331 (“It is the Commission’s tentative judgment that Congress, through its decisions to ultimately decontrol all natural gas supplies (as old gas is depleted), determined that the work-ably competitive wellhead market would do a better job of setting such rates for the commodity than would a utility regulatory body.”). But by the rehearing stage FERC had ceased to invoke the position (at least in this form), see J.A. 1065-74, and by the time of briefing seemed to repudiate it, FERC Brief at 129. However, since some petitioners invoke the argument, we address it briefly.
That access-conditioning and wellhead price controls might have somewhat similar effects (lower wellhead prices) does not make them the same thing at all. Order No. 436 mandates access in order to solve unprecedented problems in the natural gas market. That access will give producers entirely new market opportunities. Order No. 436 would create these opportunities not because the statute expressly mandates that producers should enjoy them, but because the Commission believes that such access is a well-designed corrective for practices that it finds unduly discriminatory. If it is appropriate for the Commission *50to bar refractory producers from the benefits of Order No. 436 in order to fulfill the underlying statutory mandate, then the fact that this constitutes a form of government pressure to reduce prices is itself no barrier.
C. Conclusion.
In deciding on inaction, FERC (1) assessed the hazards created by Order No. 436, finding them likely to be nil, (2) invoked certain policy concerns militating against action, and (3) rejected as flawed a variety of alternative proposals suggested during the rulemaking proceedings. Recognizing that the Commission is expert in this area, we nevertheless find its stated legal premises questionable and its factual assessments utterly Panglossian. In contrast, its primary policy concerns seem to us forceful ones. We remand to the Commission to determine, in light of our discussion of its questionable factual and legal premises, to what extent these policy considerations may justify its inaction on the uneconomical producer-pipeline contracts. We do not require that FERC reach any particular conclusion; we merely mandate that it reach its conclusion by reasoned decisionmaking.
VI. Optional Expedited Certification
Order No. 436 sets out to ameliorate certain anticonsumer and anti-competitive consequences of the Commission's classical approach to certification under the NGA. Before a pipeline may acquire or construct any facilities or offer any transportation or sales service subject to the Commission’s NGA jurisdiction, it must obtain a § 7 certificate authorizing the acts. The Commission must issue such a certificate where it finds that the applicant is able and willing to do the acts and perform the service proposed, and the facility or service is or will be required by the present or future public convenience and necessity. NGA § 7(e), 15 U.S.C. § 717f(e) (1982). Historically, the Commission’s approach to certification has entailed an elaborate effort to predict the likely consequences of allowing the proposed service or act. This included the Commission’s foretelling — or, more realistically, attempting to foretell — the impact of certification on, for example, shareholders and customers of both the applicant firm and those of any firm whose sales would be displaced. See, e.g., Atlantic Seaboard Corp. v. FPC, 397 F.2d 753, 756 (4th Cir.1968); Tennessee Gas Pipeline Co., 22 F.E.R.C. ¶ 61,174, at 61,294-96 (1983); Natural Gas Pipeline Co., 20 F.E.R.C. ¶ 61,324, at 61,675 (1982), reh’g denied, 21 F.E.R.C. ¶ 61,223 (1982), amended 21 F.E.R.C. ¶ 61,231 (1982); Columbia Gulf Transmission Co., 37 F.P.C. 118, 131, reh’g denied, 37 F.P.C. 591 (1967), aff'd sub nom., Atlantic Seaboard Corp. v. FPC, 397 F.2d 753, 757-59 (4th Cir.1968). The process has been time consuming, see, e.g., Natural Gas Pipeline Co., 30 F.E.R.C. ¶ 61,017 (1984) (Commission approval on January 14, 1985 of application filed June 1, 1982), and costly; it also utterly stifles the sort of quick responsiveness to demand that is associated with competition.
Order No. 436 seeks to stimulate competition and to enhance consumers’ access to the wellhead market by making an optional expedited certificate (“OEC”) procedure available to pipelines seeking to provide “new service” {i.e., transportation or sales service of a different magnitude or kind than it currently provides, 18 C.F.R. § 157.-101(b)(2) or to acquire or construct facilities to do the same. To qualify for the OEC procedure, a pipeline must meet criteria designed by the Commission to assure that it assumes the full economic risk of the venture.32 Applicants who agree to these *51conditions receive in return the benefit of a rebuttable presumption that the facility or service meets the statutory prerequisites of certification, 18 C.F.R. § 157.104(c), and thus enjoy the possibility of escaping the usual cumbersome hearing, 18 C.F.R. § 157.104(b). Protesters who wish to challenge the certification in a full hearing must raise genuine issues of material fact as to compliance with the statutory criteria.
This facet of the Commission’s order has provoked the greatest concern among state commissions and LDCs. These parties fear that pipelines will be able to take advantage of the OEC procedures to obtain all of the facilities and certificates necessary to deliver gas directly to end users, bypassing the LDCs currently making such sales. Such bypass, if it occurs, would at least in the short run reduce the bypassed LDC’s volume and thus force either its shareholders or remaining customers to shoulder higher costs (the so-called “cost-shifting” problem). Accordingly, the state commissions and LDCs attack both the substance of the OEC rule and the procedures established for its application.
A. Ripeness.
So far as we know, no pipeline has yet applied for expedited certification, so that none of the rule’s substantive or procedural terms has yet been applied in a concrete setting. Indeed, we may not freely assume that any pipeline will ever invoke the option: the rule’s criteria may chill all possible applications. Accordingly, we face a preliminary question of whether the challenges to this part of Order No. 436 are ripe.
The functions of the ripeness doctrine are well-established:
ference until an administrative decision has been formalized and its effects felt in a concrete way by the challenging parties.
Abbott Laboratories v. Gardner, 387 U.S. 136, 148-49, 87 S.Ct. 1507, 1515-16, 18 L.Ed.2d 681 (1967). The Court in Abbott Laboratories went on to derive from these criteria its well-known test, requiring us to evaluate “both the fitness of the issues for judicial decision and the hardship to the parties of withholding court consideration.” Id. at 149, 87 S.Ct. at 1515.
Petitioners’ claims may for ripeness purposes be divided into two classes, loosely speaking the attacks on the rule’s substance and those on its procedures. We find the first ripe, the second not. Concerning the substance of the rule, petitioners claim that the Commission in adopting it disregarded values that Congress required it to consider (e.g., impact upon LDCs’ residential customers, the need to prevent wasteful investment), made unsupported assumptions (e.g., that under the rule pipelines would really bear the full economic risk and that LDCs were really in a position to compete with pipelines on an equal footing), and failed to supply adequate reasons (especially for its abolition of a prior policy against bypass). These issues seem fit for judicial review, and we believe that delay in resolving them will impose material hardship on the challenging parties.
In the OEC provisions of Order No. 436, the Commission clearly states that it will consider itself free to certify proposed facilities or service whenever (1) it receives a qualifying application and (2) no opposition is filed. The implication, indeed, is that this will ordinarily be the outcome. Special circumstances seem needed to defeat the application; one relevant factor, for example, is whether the applicant has committed itself to nondiscriminatory transportation under Order No. 436. See J.A. 568, 1362. *52Moreover, the Commission expressly states that in an OEC proceeding it will not allow onto the agenda certain issues formerly cognizable in § 7 hearings: duplication of facilities, the problem of cost-shifting in a bypassed distributor’s rates, or impact on local interests generally. FERC Supp. Brief at 6, 8.
This is a reasonably specific, concrete proposition. Action under the OEC regulations appears likely to be virtually as automatic as under the policy statement at issue in Nader v. CAB, 657 F.2d 453 (D.C.Cir.1981), in which the CAB stated that it would not suspend any fare filing that fitted within a band surrounding the fare that a specific formula would produce. We found the policy statement ripe for review (without even considering the “hardship” component of ripeness), noting that the “mandatory language ... narrowly circumscribes the Board’s discretion to suspend air fares” within the prescribed bounds. Id. at 456. Though the Commission does not here commit itself to an automatic outcome (i.e., it does not employ mandatory language), it significantly circumscribes its discretion. It commits itself to grant applications that meet specific criteria, at least so long as offsetting factors do not appear. And while the Commission has not furnished an all-inclusive list of possible offsetting factors (the list is in a sense open-ended), it has expressly removed from the list many factors that have historically dominated § 7 proceedings. Thus one can state a range of hypotheticals and predict the Commission’s disposition with a fair degree of confidence. That some issues are unresolved does not in itself render unfit the ones that the agency has clearly determined. See Continental Air Lines v. CAB, 522 F.2d 107, 125-26 (D.C.Cir.1975).
In some respects, fitness is clearer here than in Nader. Not only has the Commission made clear it considers the regulations final and outcomes largely predetermined, see Continental Air Lines, 522 F.2d at 125, but the action takes the form of specific regulations adopted after a notice-and-comment rulemaking rather than a mere policy statement.
Moreover, the Commission’s decision is reasonably likely to have an effect on the primary conduct and day-to-day affairs of at least some of the petitioners. Cf. Toilet Goods Ass’n v. Gardner, 387 U.S. 158, 164, 87 S.Ct. 1520, 1524, 18 L.Ed.2d 697 (1967). The creation of the presumption, and the refusal to except instances of LDC bypass, increase the risk that such bypass will occur. That incremental risk clearly gives LDCs an incentive to make immediate adjustments in their business — for example, to attempt to renegotiate their contracts with end users, to lean up their operation in an effort to be more competitive, or to bring their rate structures more closely into alignment with costs. Indeed, in adopting the rule the Commission expressly noted its tendency to create economic incentives to promote economic efficiency. See J.A. 561; see also FERC Brief at 165. This likely impact on petitioners’ primary conduct, coupled with the fitness of the issue for judicial review, render the Commission’s adoption of the presumption ripe for review.
The same cannot be said of petitioners’ attacks on procedural aspects of the OEC presumption. These attacks revolve around the thesis that the Commission’s procedures may hamper their ability to mount effective attacks on proposed new services. Petitioners argue, for example, (1) that the Commission’s creation of a presumption illegally switches the burden of proof from the applicant to protesters (Brief of .the State Commission Petitioners on Bypass Issues at 18-22; Brief of United Distribution Companies at 28-29); (2) that the provisions for change of delivery points might in practice allow pipelines to deliver at a previously unauthorized location (thereby entailing bypass possibilities not suggested by the initial OEC application) (Brief of the State Commission Petitioners on Bypass Issues at 34-37); (3) that the Commission procedures would not allow protesters to ascertain facts essential to formulating their objections (Brief of the State Commission Petitioners on Bypass Issues at 37-41; Brief of the American Gas Ass’n at 23-24; Brief of Associated Gas Distributors at 36-37); and (4) that the *53Commission’s provisions for a “noncontested proceeding” might be triggered even in cases where a genuine contest existed33 (Brief of Associated Gas Distributors at 17-19).
All these attacks share a highly abstract and speculative character. On the first two issues the Commission asserts that petitioners have misread the regulations and are thus attacking phantoms of their own imaginations, see FERC Brief at 172-79; its view is certainly not implausible. All the claims seem likely to turn on individualized, fact-specific situations that are not before us. Some conceivable application — or misapplication — of the Commission’s “noncontested proceeding” regulations might unduly truncate a protester’s rights, but that possibility hardly justifies our issuing an advisory opinion on a vast range of possible situations. The claim about possible information gaps suffers the same flaw.
The character of these attacks seems even more abstract and speculative than that of the ones found unfit for review in Abbott Laboratories’s companion case, Toilet Goods Ass’n v. Gardner, 387 U.S. 158, 87 S.Ct. 1520, 18 L.Ed.2d 697 (1967). There the Court considered regulations allowing the Commissioner of Foods and Drugs to suspend certification service (a necessity for marketing the regulated commodities) whenever a subject of the regulation refused FDA inspectors “free access” to their manufacturing facilities. The Court reasoned that until the regulations were applied it could not know whether or when the Commissioner might order an inspection, or what reasons he might give for doing so. Id. at 163, 87 S.Ct. at 1524. Resolution of the legal issues would turn on the FDA’s enforcement problems, the need for supervision, and safeguards for protection of trade secrets. The Court believed that all could be resolved better in the setting of a specific application rather than “the framework of the generalized challenge made here.” Id. at 164, 87 S.Ct. at 1524. See also Association of National Advertisers v. FTC, 617 F.2d 611, 620 (D.C. Cir.1979) (finding a challenge to certain Federal Trade Commission procedures unfit for review as it would entangle the court “in an essentially abstract discussion of the [challenged procedures’] potential effects”).
Nor do petitioners demonstrate that they would suffer the slightest hardship as a result of our declining immediate adjudication of their claims. Again Toilet Goods controls. There the Court argued that the “free access” regulation was most unlikely to affect “primary conduct” — the challengers’ behavior in the day-to-day world of negotiating contracts, testing ingredients or compiling records. 387 U.S. at 164, 87 S.Ct. at 1524. It has been argued that in fact the Court overlooked a likely impact on primary conduct, namely, that drug manufacturers might reallocate resources away from development of drugs dependent on secret formulae. See Vining, Direct Judicial Review and the Doctrine of Ripeness in Administrative Law, 69 Mich.L. Rev. 1445, 1502-04 (1971). Here we see no hazard at all that uncertainty about these procedural flaws could materially alter any of the challengers’ decisions about how to conduct their business operations. The connection is far too iffy.
Accordingly, for want of both fitness for review and any material hardship to the challengers, we dismiss the procedural claims as unripe. We now turn to the merits of the petitioners’ challenge to the Commission’s creation of the presumption.
B. Legality of the Presumption.
The OEC aspect of Order No. 436 enables pipelines to make an occasional end-run around the costs and delays of ordinary certification. It establishes criteria aimed at assuring that the applicant will bear the full economic risk of the venture. The Order takes the position that, if the applicant agrees to those criteria, and no offsetting factors appear, the various public interest concerns explicit or implicit in § 7 will be satisfied.
*541. Alleged disregard of legally relevant factors.
Petitioners’ primary claims revolve around the idea that the Commission’s innovation disregards important factors that Congress intended the Commission to consider. Most prominent among these allegedly neglected factors are (1) the risk of waste through construction of duplicative facilities and (2) adverse impact on the customers of bypassed LDCs.34
It seems clear that among the congressional purposes in providing for certification was “to prevent uneconomic extensions and waste.”35 But it is equally clear that Congress has adopted no general policy against firms taking their (i.e., their investors’) money and applying it to projects that may prove unjustifiable in terms of the ultimate return. What generates special concern about wasted investment in the context of a regulated monopoly is that the ultimate bearers of the risk may be the consumers rather than the investors. For example, investment in a particular stretch of pipeline typically goes into the regulated firm’s rate base, its cost to be recovered out of rates charged all the firm’s customers. Since regulation supposedly holds the rates below profit-maximizing levels, the firm has no difficulty collecting the charge if the regulators authorize it. And it can do this even though the new investment proves a complete loser.
This risk-shifting under regulated monopoly contrasts sharply with conditions in either a competitive or an unregulated industry. Where competition prevails, a firm cannot compensate itself for losses on one venture by raising prices on other lines of business; if it tried to do so, competitors could profitably capture the business. And an unregulated firm, even if monopolistic, will have sought the profit-maximizing price and output for each of its goods or services; thus a price increase on some lines, aimed at recouping losses on another, would only aggravate its loss.
The challenge before the Commission was to see whether it could accomplish the congressional intent of protecting ratepayers from such risk-shifting and at the same time allow natural gas customers to benefit from investment and competition freed from the deadening effects of the system of certification that had evolved over the years since 1938. Its OEC procedures are thus an experiment designed to realize the congressional scheme, not to defeat it.
The losses possibly to be suffered by existing firms, whose business the applicant may capture, are another matter. But the Commission’s criteria also address that problem. Firms will apply only when their managements believe that they will be able to capture enough business to repay the investment. That estimate will be correct only under certain circumstances: there must either be undercapacity in the market (so that the applicant and incumbent firms can both enjoy an adequate return), or the incumbent firms must be performing so inefficiently that the applicant can under-price them, or the incumbent firm must be discriminating in price against the customer or class of customers that the new entrant seeks to serve. If undercapacity is the explanation, the investment entails no waste. If incumbent-firm inefficiency or price discrimination is the explanation, then certification will result in losses for a firm that, so far as appears, deserves to lose. (We address below the special case of an incumbent firm that loses business because *55the state regulatory agency has forced it to engage in price discrimination.)
Thus certification under the OEC rule seems likely to occur only when the risk of wasteful investment is remote. Of course the applicant pipeline may err; all of us do. But the Commission has at least implicitly estimated that the risk of error by a pipeline management, with its company’s solvency at stake, is no greater than the risk of error by the Commission itself. See J.A. 593-95. The Commission also anticipated that by lowering bureaucratic barriers to entry the rule would afford consumers the benefits of competitive (or at least more competitive) pricing. See J.A. 553, 561. As the Commission has given the applicant firms an incentive to make the decision correctly, we cannot gainsay its belief that the rule will in fact protect consumers from the adverse consequences of inefficient construction.
But where entry will inflict deserved losses on an incumbent LDC, some or all of those losses may ultimately fall not on its shareholders but on its remaining customers — most notably its captive residential consumers. Several petitioners strenuously argue that the Commission’s presumption violates congressional purpose to the extent that it creates such a risk. See Brief of the American Gas Ass’n at 15 — 19; Brief of Associated Gas Distributors at 15-16; Brief of the State Commission Petitioners on Bypass Issues at 24ff.
The Commission argues that the risk is slight, that state regulatory agencies have the necessary authority to keep it slight or reduce it even further, and that those agencies can protect captive consumers in the instances where it does occur. We find the Commission responses persuasive.
First, FERC notes that the record contains little to suggest that industrial consumers want to bypass their distributors. See FERC Brief at 158 n. 1. Indeed, it seems logical to suppose that what they want is gas at competitive prices. So long as the LDC will provide such gas — either as merchant or as transporter of gas bought at the wellhead or from a broker— the customer will have no incentive to try to induce pipeline investment in new facilities for a bypass. So long as the LDC is able to gain a return from the provision of transportation services, reduced purchases by the industrial buyer will normally have no more than a modest negative impact.
Second, state agencies have a variety of powers by which to control the risk of bypass. That risk seems to derive primarily — perhaps almost exclusively — from LDCs using industrial rates to subsidize residential consumers, perhaps under regulatory pressure from state commissions. The Commission so asserts, FERC Brief at 159 n. 1, and the record provides at least implicit support. Several commentators asserted that pipelines would “skim the cream” from among a distributor's customers, J.A. 559. See also Brief of Associated Gas Distributors at 32 (alluding to the potentially contested industrial user as a “lucrative customer”). “Cream-skimming” necessarily involves cross-subsidization: a customer is “creamy” simply because the rate structure allows the firm to extract from it a disproportionate share of costs. See, e.g., E. Gellhorn & R. Pierce Jr., Regulated Industries 275 (2d ed. 1987). Consequently, the Commission suggests, state agencies can head bypass off by adopting more strictly cost-based rate designs. See J.A. 1260.
State agencies may also control bypass by virtue of their jurisdiction over the bypassing transportation itself. The Supreme Court has upheld state power to require a pipeline to obtain from it a certificate of convenience and necessity before selling gas at retail (and bypassing an LDC), even where it recognized that the “end result” of the requirement might be “prohibition of particular direct sales.” Panhandle Eastern Pipe Line Co. v. Michigan Public Service Comm’n, 341 U.S. 329, 336, 71 S.Ct. 777, 781, 95 L.Ed. 993 (1951). The exact scope of this power is unplumbed, and is not before us now. But Panhandle clearly tends to undermine petitioners’ assault on the Commission’s refusal to except bypass situations from the OEC procedures, insofar as that assault *56turns on assumptions of total state incapacity to address the issue.
Similarly, the suggestion of our partially dissenting colleague that the OEC rule has “overridden the states’ ability” to subsidize residential consumers by charging higher rates to large industrial customers, see post at 1045, is correct only in a very limited sense. For example, even if we disregard the residual authority left in place by Panhandle, a state is free to tax gas sales to industrial customers and apply the proceeds to whatever subsidies it likes. This may be more awkward than the present arrangement, in which a state commission may attune its rate rulings to the price elasticity of each purchasing firm. But to say that is only to say that the present arrangement allows states to employ extraordinarily ad hoc means to finance subsidy programs. Panhandle may well allow persistence in such practices. If it does not, it is hard to see anything in the NGA that guarantees states the right to use such means, especially where the constraining factor is a Commission rule adopted in fulfillment of its duty under the Natural Gas Act to assure that natural gas be sold “at the lowest possible reasonable rate consistent with the maintenance of adequate service in the public interest.” See Atlantic Rfg. Co. v. Public Service Comm 'n, 360 U.S. 378, 388, 79 S.Ct. 1246, 1253, 3 L.Ed.2d 1312 (1959).36
Third, in the event that bypass does occur despite the potential competitive responses of LDCs and state commissions, FERC argues that the commissions may protect the captive consumers by modifying rate designs so that LDC shareholders, rather than ratepayers, bear the consequences of LDC inability to handle competition. See J.A. 1260. Where the LDC can be shown to have lost the business through imprudent judgments, that authority seems indisputable. That may dispose of the lion’s share of cases. The state commissions argue that under FPC v. Hope Natural Gas Co., 320 U.S. 591, 64 S.Ct. 281, 88 L.Ed. 333 (1944), they could not deny recovery of costs that were prudently incurred. We do not think Hopes so restrictive. If the investment was prudent when made, but the losses occurred because the LDC maintained an irrational rate design that drove away the industrial customers, it is hard to see anything in Hope’s modest constitutional requirements that would bar state commissions from curtailing return on investment thereby wasted. Of course the matter would be different if the state commission had forced the LDC to maintain the vulnerable rate structure.
We note in this connection that the Brief of Associated Gas Distributors contends both that the Commission has neglected its duty to protect consumers by “delegating” it to state commissions, Brief of Associated Gas Distributors at 15-16, and that it has intruded upon the regulatory jurisdiction of the states, id. at 33-35. The first arm of this paradoxical complaint seems to us to assume far too great a reach for FERC’s obligations. The Natural Gas Act was adopted in order to fill a “regulatory gap” created by Supreme Court decisions finding the states without power, as a result of negative implications of the commerce clause, U.S. Const. Art. I, § 8, cl. 3, to regulate interstate sales at wholesale. See H.Rep. No. 709, 75th Cong., 1st Sess., 1-2 (1937) (citing Missouri v. Kansas Natural Gas Co., 265 U.S. 298, 44 S.Ct. 544, 68 L.Ed. 1027 (1924), and Public Utilities Commission v. Attleboro Steam & Electric Co., 273 U.S. 83, 47 S.Ct. 294, 71 L.Ed. 54 (1927)). The Commission fulfills its mission under the Act if it establishes conditions by which the gas arrives in the hands of consumers or LDCs on terms that conform to the Act. That consumers may later suffer because of regulatory decisions made by state commissions, acting pursuant to their lawful authority, evidences no “delegation” by FERC.
The claim of intrusion on state authority appears to derive from the view that FERC authorization of transportation under OEC will lead to transactions that sharply affect *57state interests but which the state is unlikely to be able to control. The specific example offered is of gas purchased out of state by an industrial user and then transmitted through OEC facilities. Brief of Associated Gas Distributors at 35. The distributors argue that state regulation of such a sale may not be legal. They suggest that Panhandle Eastern Pipe Line Co. v. Michigan Public Service Commission, 341 U.S. 329, 71 S.Ct. 777, 95 L.Ed. 993 (1951), holding that the NGA left intact state authority to regulate direct pipeline sales to an industrial end user, might not encompass an end user’s purchase of gas from an out-of-state supplier.
Thus the contention is that the OEC procedures may present a state with a dilemma. We remain uncertain why the possibility of this dilemma should be characterized as an intrusion into state authority. In any event, the entire problem turns upon application of Panhandle to a set of facts that is not before us. Accordingly, we find the claim too hypothetical for disposition in this context.
Finally, we note that the presumption created by the OEC procedures is rebut-table. We cannot know at this point precisely what will be necessary to persuade the Commission that application of the presumption is inappropriate. But the existence of that backstop confirms our judgment that the OEC rules themselves can readily be applied in a manner consistent with the congressional purposes.
The Associated Gas Distributors claim that the OEC rules constitute a breach of the Commission’s procedural obligation to ferret out relevant information even when no party draws its attention to them. They cite Scenic Hudson Preservation Conference v. FPC, 354 F.2d 608, 619-22 (2d Cir.1965), cert. denied, 384 U.S. 941, 86 S.Ct. 1462, 16 L.Ed.2d 540 (1966), which faulted the Commission for failure to pursue a wide variety of issues that environmental intervenors had brought to its attention. The court said that the Commission had “an affirmative duty to inquire into and consider all relevant facts.” Id. at 620. This proposition seems unexceptionable but irrelevant. The Commission determined that in most instances compliance with the limited criteria would assure a properly certificable facility or service, and provided for intervention in individual cases where anyone thought otherwise. Further, the issues not directly addressed in the initial criteria — notably, impact on competitors — are precisely the ones that are likely to trigger intervention. Under these circumstances, the procedures seem likely to assure an adequate vetting of all relevant and persuasive contentions.37
2. Unsupported assumptions.
The state commissions argue strenuously that the Commission has made an unwarranted assumption that under its OEC rules the pipeline will bear all the risks of a venture. The commissions point to rules allowing the applicant to charge a reservation fee for firm transportation. See § 157.103(d)(3). Such a charge of course involves risk-sharing with the particular customer who arranges for the service. But that form of risk-sharing seems wholly compatible with the Commission’s fundamental reasoning: no risk is shifted onto any natural gas consumer against its will.
Many LDCs argue that the Commission has wrongly assumed that LDCs have the capacity to compete with pipelines for industrial customers on an equal footing. They point specifically to the LDCs’ being subject to legal service obligations (including duties to continue service even to customers that fail to pay their bills) and to restrictions on selective discounts.
*58We find no basis for believing that the Commission acted on false premises. The existence of service obligations alone does not distinguish LDCs from pipelines. Both are subject to them; the pipelines’ are discussed in this opinion’s section on CD modification. That some states may impose greater burdens on LDCs hardly seems a sufficient basis for the Commission to refrain from its efforts to stimulate the availability of a supply of gas at competitive prices. If states choose to require LDCs to ■ continue service to non-paying customers, those states must address the consequences. They can extract the cost from price-inelastic customers, primarily the solvent residential consumers; they can seek to extract the cost from industrial customers, at the peril of driving them off the system; they can fund the expense out of tax revenue; they can use their power under Panhandle to thwart the possible bypass, accepting the economic consequence that their industrial gas users may be unable to compete with firms in other states. All of these choices may involve some pain — like all true choices. But that hardly requires the Commission to abandon its effort, required under the NGA, to facilitate the flow of competitively-priced gas into the hands of gas consumers everywhere.
3. Alleged failures of reasoned decisionmaking.
The state commissions fault the Commission for failing to explain the shift from a former policy against allowing bypass. Supporting its view of the existence of such a policy, the commissions cite three cases that may be so read. See Panhandle Eastern Pipe Line Co., Opinion No. 539, 39 F.P.C. 581 (1968); Panhandle Eastern Pipe Line Co., Opinion No. 274, 13 F.P.C. 301 (1954), aff'd 232 F.2d 467 (3d Cir.1956), cert. den., 352 U.S. 891, 77 S.Ct. 129, 1 L.Ed.2d 86 (1956); Southern Natural Gas Co., 25 F.P.C. 925 (1961). While it is true that the Commission has not, so far as we can discover, explicitly addressed the fact of this policy change, its overall reasoning provides ample explanation for modifying the policy.
The only ground ever given by the Commission for any prior anti-bypass policy — so far as we are able to determine— was its concern over the wasteful construction of duplicative facilities. See Southern Natural Gas Co., 25 F.P.C. 925, 927 (1961). In the absence of a mechanism for preventing such waste, the policy is quite understandable. In its deliberations leading to Order No. 436, however, the Commission believed it had discovered a mechanism for solving that problem, namely, its risk-to-applicant rules. As we believe the Commission adequately justified its view that this form of certification would prevent the waste that the NGA’s certification procedure sought to avoid, the Commission was not obliged to supplement the discussion by express consideration of the demise of its former policy.
VII. Miscellaneous Arguments
In addition to the broad attacks on Order No. 436 discussed above, the parties raise a number of relatively minor concerns.38
A. Pipeline Sales Service Issues.
A number of parties assert that FERC acted unlawfully in promulgating Order No. 436 without addressing various problems presented by pipeline sales service. The challenged omissions include: failure to completely bar minimum commodity bills (Brief of Columbia Gas Transmission Corp.); refusal to extend Order No. 436 to the transportation component of sales service (Brief of Peoples Gas Light & Coke Co. and North Shore Gas Co.); failure to eliminate sole supplier provisions from pipeline sales tariffs (Brief of Central Illinois Light Co.).
As the Notice of Proposed Rule-making made clear from the start, however, FERC intended Order No. 436 to *59be a comprehensive resolution only of transportation issues. In order to accomplish its goal of uncoupling transportation services from sales service effectively and in conformity with this court’s holdings in the MPC cases, FERC addressed certain issues presented by sales service, notably CD conversion/reduction and the producer-pipeline purchase contracts. As we noted above, consideration of these issues was in all likelihood essential to FERC’s task. Whenever possible, however, FERC has left sales issues for resolution outside of Order No. 436. This necessarily causes a certain incompleteness. But all natural gas issues are related to each other. That the web is seamless does not require an agency to reweave the whole whenever it reweaves a part. An agency is entitled to the highest deference in deciding priorities among issues, including the sequence and grouping in which it tackles them. Cf. Heckler v. Chaney, 470 U.S. 821, 831-32, 105 S.Ct. 1649, 1655-56, 84 L.Ed.2d 714 (1985). We see no abuse of discretion in FERC’s decisions to exclude the specified items from this page of its agenda.
B. Notice Provisions.
The Industrial End User Group objects to the notice and protest provisions applicable to § 7 transportation, 18 C.F.R. § 284.223, contrasting them with the far less onerous reporting requirements applicable to § 311 transportation, 18 C.F.R. §§ 284.106, 284.-126. The Group argues that the discrepancy constitutes an unjustifiable discrimination, thus violating the Commission’s statutory mandate and its stated purposes in embarking on the rulemaking. The distinction has bite for the industrial users because they are ineligible as direct beneficiaries of § 311 transportation, which must be on behalf of an LDC or another pipeline.
We agree with FERC that there is both factual and legal support for the distinction. As FERC notes, petitioners’ main objection is that such provisions make it more difficult to effect a bypass of an LDC, a situation FERC regards as meriting special scrutiny. The industrial users suggest that even on that basis no special notice is needed, for, they say, the gas will be transported either (1) to an existing direct customer or through an LDC, in which case no injury occurs, or (2) through a new connection, the construction of which would require Commission approval under § 7. But we do not understand these to encompass the universe of possible cases. An industrial user now supplied in part by a pipeline and in part by an LDC, for example, might shift away from reliance on the LDC and towards gas carried under a § 7 blanket certificate.
Moreover, the Commission in the rulemaking called attention to the value of fulfilling § 311’s purpose of integrating the formerly segregated interstate and intrastate markets. J.A. 1225-26. Though articulated in a different context, the point supports less stringent rules for § 311 transportation. In § 311 Congress gave FERC broad authority to prescribe terms of transportation thereunder; it surely did not contemplate that FERC would use this authority to duplicate the regulatory scheme in place under the NGA, including its elaborate certification mechanism. See NGPA § 601(a)(2), 15 U.S.C. § 3431(a)(2) (1982) (Commission’s NGA jurisdiction not to apply to transportation provided under § 311). The distinction drawn by the Commission is completely congruent with the one drawn by Congress in the two statutes.
C. Transportation Policies.
The Industrial End User Group also argues that FERC acted unreasonably in refusing to require pipelines not electing to operate under Order No. 436 to publish their transportation policies in their tariffs. This issue is currently pending in an individual rate case now on review before the Commission from an administrative law judge's initial decision, Transcontinental Gas Pipe Line Corp., 33 F.E.R.C. ¶ 63,035 (1985). The Group has not met the heavy burden of establishing the inadequacy of resolution in that forum. See NLRB v. Bell Aerospace Co., 416 U.S. 267, 294-95, 94 S.Ct. 1757, 1771-72, 40 L.Ed.2d 134 (1974).
*60D. Construction of Facilities for Use in § 311 Transportation.
Maryland People’s Counsel asserts that Order No. 436 violates the NGA by leaving in place a prior regulation making clear that construction of facilities for use solely in providing § 311 transportation does not require § 7 certification. See 18 C.F.R. § 284.3(c) (“The [NGA] shall not apply to facilities utilized solely for transportation authorized by section 311(a)____”). The crux of MPC’s argument is that Order No. 436 expands § 311 transportation to the point where it is indistinguishable from § 7 transportation. Relying on legislative history indicating that Congress adopted § 7 in order to assure that consumers were protected from having to bear the costs of uneconomic construction, MPC contends that FERC must, in effect, extend the construction certification requirements of § 7 to § 311.
MPC’s claim turns on the premise that Order No. 436 makes § 311 transportation a clone of transportation under § 7. We find this exaggerated. As Congress specified, § 311 transportation is limited to transportation on behalf of an LDC or a pipeline. Other petitioners call our attention to instances where that requirement appears to be fulfilled more in name than reality.39 We are told that in one case the contact with an LDC consists only of transportation through a six-foot stretch of pipe. Brief of Southern Indiana Gas & Elec. Co. at 13. Despite this episode, we do not believe that petitioners have shown a sufficient identity between § 311 and § 7 transportation to call FERC’s judgment into question.40 First, we are shown no evidence that § 311 transportation frequently nominal connection with an LDC or other pipeline. Second, § 7(h) gives the holder of a § 7 certificate a power of eminent domain to acquire right-of-way to implement the Commission-approved plan; no such power accompanies authorization under § 311.
Thus, MPC would have us order the Commission to graft regulatory burdens mentioned only in the NGA onto § 311 merely because transportation under the two sections overlaps to some degree. This Congress did not intend; in the NGPA it declined to require a system of certification akin to that of § 7. Congress clearly recognized that consumer protection purposes can be achieved without control over entry. It selected for § 311 transportation a system paralleling its choice for oil pipelines: control over price but not entry. See NGPA § 311(a)(1)(B), 15 U.S.C. § 3371(a)(1) (B) (1982) (providing power over rates for service); 42 U.S.C. § 7172(b) (1982) (transferring to FERC regulatory power to establish rates and charges of oil pipelines). Congress underscored this distinction in NGPA § 601(a)(2), providing that transportation in interstate commerce under § 311 shall not trigger the Commission’s NGA jurisdiction. This seems an unequivocal expression of intent that NGPA regulation should not replicate the burdens of the NGA.
E. Grandfathering Decisions.
In Order No. 436 FERC grandfathered a number of existing transportation arrangements for varying periods of time. These are attacked as too short or too long, or as being unjustified in comparison to the treatment of other arrangements.
All grandfathering decisions necessarily involve a trade-off between the values of stability on one side and of promptly securing a new rule’s benefits on the other. FERC undoubtedly has power to make such adjustments; no petitioner questions it. Although FERC must exercise the power in a reasoned manner, such exercises are “necessarily arbitrary to some extent.” *61Capital Cities Communications, Inc. v. FCC, 554 F.2d 1135, 1139 (D.C.Cir.1976).
Michigan Consolidated Gas Co. (“Mich-Con”) complains that the Commission’s decision to grandfather blanket certificates previously authorized under § 311 and § 7 is an unreasoned one. It attacks both the decision to grandfather these transactions at all and the duration of the exemptions, which in the case of some § 7 certificates can run as long as 10 years from their start (the latest of which presumably originated shortly before promulgation of Order No. 436). Most acutely, MichCon argues that termination of these arrangements will not interrupt anyone’s supply of gas. As the pipelines are more than able to satisfy any displaced sales from their supplies, the functional effect is merely to provide the presently-favored parties continued access to low-priced gas and to increase the probability that others will not secure such access. (The grandfathered authorizations tend to relax the pressure on pipelines to accept Order No. 436.) FERC conceded as much, in rejecting the high-priority users’ pleas for a permanent exemption. J.A. 506-07 (“there is no need to retain special procedures for a limited group of shippers”). So far as appears, however, FERC grandfathered these transactions with no more than references to the desirability of a smooth transition to a new order, J.A. 273, to the reliance the parties placed on this transportation, J.A. 491-95, and to the authorized duration of the transactions and the need to avoid undue infringement on them, J.A. 1183.
These phrases do not seem to us to meet the modest standard implicit in the concept of reasoned decisionmaking. See Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 42-43, 103 S.Ct. 2856, 2866-67, 77 L.Ed.2d 443 (1983). Thanks to the delays of the legal process (for which we acknowledge our share of responsibility), the grandfathering that expires October 9, 1987 is nearly moot and scarcely worth mentioning. Matters are different for transactions whose unlawfulness is extended into the mid-1990s. Perhaps these are rare enough, or involve so little volume, that the grandfathering will only trivially encourage pipeline resistance to the concept of nondiscrimination. Or perhaps the equities of the participants are stronger than appear. We cannot, however, “discern the path” by which the Commission has reconciled such grandfathering to its acknowledged mandate to stamp out discrimination. Cf. Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 285-86, 95 S.Ct. 438, 441-42, 42 L.Ed.2d 447 (1974).
The intrastate pipelines attack from the other side, challenging FERC’s failure to preserve all preexisting § 311 authorizations for their full authorized terms. Order No. 436 permits previously authorized §311 transactions to continue until the earlier of their expiration or two years after the Order’s effective date. The two-year grandfather period allows previously authorized § 311 transactions to come to a natural end, as FERC generally limited such transactions to two-year terms. 18 C.F.R. § 284.102(b)(l)(i) (pre-Order No. 436 version). In some instances, however, FERC provided for longer terms, see id. § 284.102(b)(2); the two-year grandfathering limit will cut some of these short. The petitioners contrast this with FERC’s decision to grandfather existing § 7 certificates for their full terms.
There are two elements to the intrastate pipelines’ attack on FERC’s distinction between § 311 and § 7 grandfathering. First, they implicitly attack the existence of any line between the two at all, and here we cannot agree. In imposing the two-year ceiling on these § 311 extensions, FERC relied on its regulation, adopted at the outset of § 311 authorization, expressly making such certificates subject to prospective change. See Order No. 46, 44 Fed. Reg. 52,179, 52,181-82 (1979) (codified in pertinent part at 18 C.F.R. § 284.5); J.A. 1180-81. In its brief (though not in the rulemaking itself) the Commission contrasts that explicit warning with its absence in the § 7 context, suggesting greater likelihood of justifiable reliance in the latter. We think the point strong enough to explain non-identical treatment.
*62Insofar as the intrastate pipelines challenge the size of the discrepancy in treatment, we must agree. Above we found inadequate FERC’s explanation for allowing the continuation of § 7 transportation for five- and ten-year terms. The discrepancy in treatment of § 311 authorizations deepens our mystification.
We can find no merit in petitioner Tenneco’s lament that the Commission did nothing to allow continuation of the sales underlying the SMPs invalidated in Maryland People’s Counsel v. FERC, 761 F.2d 768 (D.C.Cir.1985) (“MPC I”). The entire thesis of MPC I was that the Commission had failed to assure that transportation for those purposes did not unduly discriminate. Order No. 436 is the Commission’s answer. It allows those sales to continue, so long as the transportation is provided on a nondiscriminatory basis. We can detect no reason why the Commission should even consider finding an especially soft berth for the SMP gas. Such a view would nullify this court’s time limit on continuation of the SMPs. See Maryland People’s Counsel v. FERC, 768 F.2d 450, 455 (D.C.Cir.1985) (“MPC III”).
F. System Supply Limitation.
Prior to Order No. 436 a FERC regulation limited § 311 transportation to gas delivered to an LDC, an intrastate pipeline, or an interstate pipeline for “its ‘system supply for resale.’ ” J.A. 504 (referring to 18 C.F.R. 284.102(b)). As part of Order No. 436 FERC amended this regulation to permit transportation under § 311 in any case where it is “on behalf of” an intrastate pipeline or LDC, J.A. 505, thus incorporating the standards of the statute itself, NGPA § 311(a)(1)(A), 15 U.S.C. § 3371(a)(1) (A) (1982). Southern Indiana Gas and Electric Company contends that this change impermissibly opens “the way for interstate pipelines to invade the core markets of LDCs without prior FERC approval or the approval of the state regulatory commission.” Brief of Southern Indiana at 12.41 As noted above, it points to a case where the LDC on whose behalf the gas is transported itself carries the gas for only six feet. Id. at 13. Southern Indiana argues that elimination of the test is a “perversion” of § 311 because it will enable pipelines to bypass an LDC.
We disagree with Southern Indiana’s contention that § 311 was “expressly intended to benefit LDCs.” Id. The purpose of § 311 — so far as we are able to discern — is not to bestow special benefits on any of the parties on whose behalf it authorizes transportation {i.e., interstate pipelines and intrastate pipelines as well as LDCs), but to reduce the barriers between the interstate and intrastate gas markets. See Public Service Comm’n of the State of New York v. Mid-Louisiana Gas Co., 463 U.S. 319, 342, 103 S.Ct. 3024, 3037, 77 L.Ed.2d 668 (1983); Process Gas Consumers Group v. United States Dep’t of Agric., 694 F.2d 728, 764 (D.C.Cir.1981) (other portions vacated and reconsidered en banc, 694 F.2d 778 (D.C.Cir.1982)), cert. denied, 461 U.S. 905, 103 S.Ct. 1874, 76 L.Ed.2d 807 (1983). The many comments received on this issue, including those of LDCs, indicate that the system-supply test frustrated this purpose by placing LDCs at a severe disadvantage as against pipelines serving industrial end users directly under § 7 certificates. J.A. 504 & n. 119. The Commission’s decision to eliminate the test in favor of a rule more closely tracking the language of the statute — and thereby rendering the legal barrier far more permeable — was well within its powers.
*63G. Transportation for Fuel-Switchable End Users.
Laclede Gas Company contends that by adopting a program that permits fuel-switchable end users access to self-implementing transportation, FERC has impermissibly undermined the incremental pricing provisions of Title II of the NGPA, 15 U.S.C. §§ 3341-48 (1982). We reject the claim.
Congress sought to achieve two purposes in providing for incremental pricing. The first was to protect high-priority end users from a portion of NGPA-permitted price increases by shifting those costs to industrial users. See Ohio Ass’n of Community Action Agencies v. FERC, 654 F.2d 811, 813 (D.C.Cir.1981). The second — the “market-ordering” purpose — was to maximize pipelines’ incentive to bargain with producers. Congress recognized that the system of pricing tiers (low ceilings for old gas, higher ceilings for new and none for certain high-cost gas) could lead pipelines to bid more for new or high-cost gas than their estimates of the long run market-clearing price. (Simplifying, pipelines would use the savings on old gas to finance purchases of new or high-cost gas; though paying supra-market prices for the latter, it would be able to recover the cost by selling at average prices. See National Regulatory Research Institute, State Regulatory Options for Dealing with Natural Gas Wellhead Price Deregulation 40-51 (1983). But see Niskanen, Natural Gas Price Controls: An Alternative View, Regulation 46 (Nov./Dec.l987).) Incremental pricing would supposedly cure this by forcing the pipelines to bid no more for high-cost or new gas than the price-elastic segment of the market would bear. See, e.g., Rule Required Under Section 202 of the Natural Gas Policy Act; Incremental Pricing, 45 Fed.Reg. 31,622, 31,623 (1980); Natural Gas Policy Act; Incremental Pricing; Phase II, 48 Fed.Reg. 55,294, 55,295-96 (1983). The principle has largely collapsed. Falling oil prices have made it impossible to load any significant portion of NGPA-permitted increases onto fuel-switchable consumers. 48 Fed.Reg. at 55,296.42
Laclede’s thesis is that elimination of the system supply test will facilitate industrial end users’ efforts to purchase gas directly from producers rather than through pipelines or LDCs. This, it suggests, will evade the intent of Congress in establishing incremental pricing, as incremental pricing only operates on pricing by pipelines and LDCs, not on direct producer sales to end users.
Laclede’s theory extrapolates far too much from the incremental pricing provisions. If accepted, it would seem to require invalidation of all blanket-certificate transportation for users possibly subject to incremental pricing, perhaps even case-specific certification. Further, Order No. 436 constitutes in itself an alternative mechanism for achieving the purposes of incremental pricing. First, it will give LDCs direct access to wellhead market prices considerably lower than pipelines’ embedded contract gas costs. See supra part I. Second, by affording all parties access to a competitive wellhead market it prevents pipelines from passing through — and therefore from bidding — prices exceeding their estimates of the long run market-clearing price. Where the Commission has devised methods for effecting Congress’s goals in Title II of the NGPA, it would be extraordinary for a court to spin out a remote inference from those provisions to strike down a portion of its handiwork.
Moreover, Congress was deeply concerned lest incremental pricing cause fuel-switchable end users to bolt the system, exposing high-priority customers to higher prices as a result of load loss. See Ohio Ass’n of Community Action Agencies v. FERC, 654 F.2d at 821-22. Order No. 436 in general, and elimination of the system *64supply test in particular, respond to that interest. By promoting fuel-switehable end users’ use of pipeline transportation service, they make it possible for captive customers to bear a smaller portion of the pipelines' fixed costs.
VIII. Conclusion
As a general matter we uphold the substance of Order No. 436 and the procedures the Commission employed in adopting it. However, we have found problems in a few of the Order's components and must remand the matter to FERC for further proceedings. We now summarize these defects briefly.
The CD adjustment conditions suffer from a want of both legal authority and reasoned decisionmaking. The problem with legal authority stems from the Commission’s misplaced reliance on § 7(e); the lack of reasoned decisionmaking lies in its failure to trace the path that led it from the statutory authority and the facts to its adoption of the CD reduction option. On remand, FERC is free to attempt to remedy these problems. If, however, it elects to go forward with the Order without CD conversion, it must, of course, explain why it feels that such a condition is no longer necessary.
The Commission’s decision not to take any affirmative action to solve the problems posed by the uneconomic producer-pipeline contracts also fails to meet the standards of reasoned decisionmaking. On remand, FERC must more convincingly address the magnitude of the problem and the adverse consequences likely to result from the nondiscriminatory access and CD adjustment conditions. It is not our place to say whether FERC should take measures to shift these losses from where they now lie; we ask only that it explain its course.
Finally, FERC has failed to adequately explain some of its grandfathering decisions. Even though a certain amount of arbitrariness goes with this territory, the Commission must reveal a reasoned basis for its decisions.
The parts of Order No. 436 are interdependent. The nondiscriminatory access and CD adjustment provisions aggravate the pipelines’ jeopardy from take-or-pay liability. When coupled with take-or-pay liability, those provisions may bring about a wasteful imbalance between pipeline sales and unbundled transportation service. Thus the Commission’s apparent insouciance on take-or-pay taints the package. And CD adjustment, which the Commission has identified as essential to remedying problems deriving from the pipelines’ market power, suffers from independent vulnerabilities. Accordingly, we vacate Order No. 436 and remand for further proceedings consistent with this opinion.
It is so ordered.
. References to the Commission also include its predecessor, the Federal Power Commission, where appropriate. See Executive Order No. 12,009, 42 Fed.Reg. 46,267 (1977) (implementing portions of Department of Energy Organization Act, Pub.L. No. 95-91, 91 Stat. 565 (1977), terminating the FPC and transferring its duties to FERC, an independent agency within the Department of Energy).
. A lawyer in the natural gas industry once urged that the only beneficiaries of the Phillips decision were lawyers. See C. Hawkins, The Field Price Regulation of Natural Gas 205 (1969). The fallout from that decision appears still to provide a rich lode.
. Except where indicated otherwise all citations to the Code of Federal Regulations are to the 1986 edition.
. Attacks on the Order for failure to take account of preferences based on pipeline contracts with particular customers are discussed below in consideration of problems of capacity allocation. See infra part II.C.2. Insofar as the pipelines claim that prohibitions of undue discrimination must be based upon individualized findings, the cases on which they rely, American Smelting & Rfg. Co. v. FPC, 494 F.2d 925, 939-41 (D.C.Cir.), cert. denied, 419 U.S. 882, 95 S.Ct. 148, 149, 42 L.Ed.2d 122 (1974); Louisiana Power & Light Co. v. FPC, 526 F.2d 898, 905-07 (5th Cir.1976), are plainly inapposite. See Wisconsin Gas Co. v. FERC, 770 F.2d 1144, 1165-68 (D.C. Cir.1985), cert. denied, — U.S. -, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986).
. "Any authorization granted under this section shall be under such terms and conditions as the Commission may prescribe." Id.
. Thus we reject attacks predicated merely on a view that pipelines will prefer to give up § 311 transportation altogether rather than submit to the conditions that Order No. 436 imposes. See, e.g., Brief of the American Gas Ass’n at 37-41.
.Section 5(e) provides:
Rights-of-way through the submerged lands of the outer Continental Shelf, whether or not such lands are included in a lease maintained or issued pursuant to this subchapter, may be granted by the Secretary for pipeline purposes for the transportation of oil, natural gas, ... upon the express condition that oil or gas pipelines shall transport or purchase without discrimination, oil or natural gas produced from submerged lands or outer Continental Shelf lands in the vicinity of the pipelines in such proportionate amounts as the Federal Energy Regulatory Commission, in consultation with the Secretary of Energy, may, after a full hearing with due notice thereof to the interested parties, determine to be reasonable taking into account, among other things, conservation and the prevention of waste.
43 U.S.C. § 1334(e) (1982).
. H.Conf.Rep. No. 95-1474, 95th Cong., 2d Sess. 87, reprinted in [1978] U.S.Code Cong. & Admin. News pp. 1450, 1674, 1686.
. The same petitioners object to the Commission's allowing a pipeline to offer blanket certificate transportation under § 311 without offering it under § 7 as well. See 18 C.F.R. part 284, subparts B and G. This allows a pipeline to provide open access to pipelines and LDCs without offering it directly to end users. In the absence of evidence that this decision will seriously impede access for the sort of end user that can buy gas for itself in the wellhead market, we decline to interfere with the Commission's distinction between transportation under the NGPA and transportation under the NGA — a distinction that the pipelines assert has eroded too much.
. These sections govern permissible rates for transportation by interstate pipelines under § 7 blanket certificates and under § 311. Rates charged by intrastate pipelines under § 311 are required to be “fair and equitable.” See NGPA § 311(b)(2)(A), 15 U.S.C. § 3371(b)(2)(A) (1982); 18 C.F.R. §§ 271.101-1106 (1985). Order No. 436 also requires intrastate pipelines to adhere to the restrictions on reservation fees that Order No. 436 imposes on interstate pipelines' transportation. See 18 C.F.R. § 284.123(b).
. See also supra part II.B.3.
. See E. Gellhorn & R. Pierce, Regulated Industries 185-89 (1987).
. See Tye & Leonard, On the Problems of Applying Ramsey Pricing to the Railroad Industry with Uncertain Demand Elasticities, 17A Transportation Research 439 (1983); Tye, Ramsey Pricing and Market Dominance Under the Staggers Rail Act of 1980, 24 Transportation Research Forum 667 (1983); Meyer & Tye, Toward Achieving Workable Competition in Industries Undergoing a Transition to Deregulation (March 19, 1987) (unpublished). We do not understand these critics to attack rate differentials where application of some apparently egalitarian principle, such as an equal revenue-to-variable-cost ratio, would result in prices for some customers or commodities above what the profit-maximizing monopolist would charge. See Tye, Ramsey Pricing and Market Dominance Under the Staggers Rail Act of 1980, 24 Transportation Research Forum at 669-70; Henderson, Price, Discrimination Limits in Relation to the Death Spiral, 7 Energy Journal (No. 3) 33, 37 (1986).
. Order No. 436 largely retains this two-tiered rate structure, with firm transportation customers paying a "reservation” fee for the guaranteed right to call on a certain amount of the pipeline's capacity and a "volumetric” fee to cover the variable cost of providing the service actually called for. See supra part III.
. A pipeline becomes subject to the provisions of Order No. 436 for these purposes when it accepts a blanket certificate, provides service under the § 311 regulations after a certain date, or commences or continues transportation of OCS gas under a blanket certificate. 18 C.F.R. § 284.10.
.A customer may elect both to convert and to reduce, but the total adjustment to CD cannot exceed the customer's accrued entitlement. 18 C.F.R. § 284.10(e).
. We see no procedural objection to the Commission’s identification of circumstances, in an otherwise valid rulemaking, which automatically trigger its approval of abandonment (i.e., establish a system of "pre-granted" abandonment approval). Cf. Wisconsin Gas Co. v. FERC, 770 F.2d 1144 (D.C.Cir. 1985), cert. denied, — U.S.-, 106 S.Ct. 1969, 90 L.Ed.2d 653 (1986) (approving modification of tariffs in a generic § 5 proceeding).
. In some cases these commitments were not freely entered into, but are the product of Commission orders under NGA § 7(a) that the pipeline extend service to a particular customer.
. Panhandle and Trunkline seek to supplement the reliance argument by noting that the investments were made on the assumption that "rolled-in" pricing {i.e., charges by pipelines based on average gas cost rather than marginal cost) would enable the pipelines to recover the costs. The Commission approved the investments, and in doing so acknowledged that average pricing was essential to cost recovery: no customer would buy the LNG at its full price. Opinion No. 796-A, 58 F.P.C. at 2940. The point does not enhance their claim; rolled-in pricing remains. The pipelines’ problem is that even when prices are rolled in, their embedded contract costs for gas exceed what the current market will bear.
. The Commission rightly acknowledged that the circumstances justifying the extraordinary remedy of unilateral contract adjustment were transitional. It afforded the conversion right only to firm sales customers on the date a pipeline becomes subject to Order No. 436. 18 C.F.R. § 284.10(b); J.A. 440-41. Any customer entering into a sales agreement thereafter will have had the option of nondiscriminatory access by virtue of Order No. 436; accordingly, it will be held to such a commitment. See id.
. The spot market is, of course, not adequate as an exclusive source of gas for an intermediary selling to non-fuel-switchable consumers. But even a price of $2.50 for gas under long-term contracts (the price available as of mid-1986, see supra part I) would represent a saving over the pipelines’ embedded contract price. The latter includes not merely the average price being paid by pipelines at the wellhead (about $2.50 as of mid-1986, accordingly to FERC, see supra part I), but also the build-up in take-or-pay liability deriving from reduced takes of higher-priced gas.
. MPC does not challenge the phase-in of CD reduction.
. We note that the Commission never explicitly responded to the proposals by the Department of Public Service of the State of New York that LDCs be required to give notice of their future gas purchase plans, with some sort of sanctions for deviation from projections. The Commission may have believed that its discussion of the need for customer flexibility addressed this issue. Arguably it did; the cursory response in FERC’s brief did not address the issue of what the Commission had considered in the rulemaking (as opposed to what counsel later thought). FERC Brief at 104-05 n. 1. On remand explicit consideration of the suggestion would eliminate a potential problem.
. The Phillips decision found wellhead sales of gas for resale in interstate commerce to be within the Commission's NGA jurisdiction. The NGPA removed some instances of existing wellhead sales from the Commission’s jurisdiction, and provided, in effect, that its jurisdiction would extend to no future ones except for sales from the outer continental shelf. See NGPA § 121, 15 U.S.C. § 3331 (1982).
. No party here presents any argument for a view that FERC could exercise its § 5 power directly to modify nonjurisdictional wellhead contracts.
. A broad coalition of pipelines and LDCs also advances two less prominent proposals made and rejected during the rulemaking. Brief of Indicated Petitioners and Intervenors on Take- or-Pay Contracts Issue at 31 & n. 1. First, petitioners contend that certain take-or-pay provisions in wellhead sales contracts might be deemed in violation of the NGPA. For example, the Commission might find that a contract that provided for sale at the NGPA ceiling and allowed a producer to keep take-or-pay prepayments even after it has resold the gas to another party, was in effect a contract for sale at a price in excess of the NGPA ceiling. Id. We find this premise somewhat improbable. In enacting the NGPA ceilings, Congress must have been aware that producers and pipelines would incorporate these ceilings into long-term contracts, and that the contracts would include remedies for producers. Obviously the remedial rights would constitute value. If any such value put into breach of the NGPA a contract nominally at the NGPA ceiling, the NGPA would provide a most uncertain guide. Congress may well have supposed that state rules against penalty clauses would suffice to prevent exorbitant remedial provisions. Even assuming arguendo that clauses restricting recoupment would violate NGPA ceilings, we do not believe that FERC was obliged to consider the issue in this proceeding in the absence of reason to believe that the economic value of these obligations forms a major portion of the discrepancy between pipeline average contract prices and the wellhead price for newly available gas. See Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 51, 103 S.Ct. 2856, 2870, 77 L.Ed.2d 443 (1983).
Second, petitioners suggest that the Commission require pipelines to file tariffs expressing a "cutback policy," pursuant to which the pipeline would not buy gas priced above some specified figure. Brief of Indicated Petitioners and Intervenors on Take-or-Pay Contracts Issue at 31 n. 1. If such tariffs were to be legally binding on producers, they would, as the Commission argued, see J.A. 332, essentially substitute FERC price controls for the wellhead market, a move clearly forbidden by the NGPA. See Transcontinental Gas Pipe Line Corp. v. State Oil & Gas Board, 474 U.S. 409, 106 S.Ct. 709, 716-17, 88 L.Ed.2d 732 (1986). To the extent that such *43tariffs would not bind producers, it is hard to see what they would accomplish.
. FERC expanded § 2.76(e) to provide that interstate pipelines participating in a take-or-pay settlement shall be deemed to have waived any objection to the producer’s application for abandonment.
. We reject the American Gas Association's attacks on the expedited abandonment provisions. These provisions merely outline circumstances in which FERC will proceed in an expedited manner. They do not change FERC's policy as to when abandonment is in the public interest, . and we have already upheld FERC’s power to streamline its procedures in this manner. See supra part IV.A.2.
. FERC has asserted that nothing in Order No. 436 "is intended to abridge the rights and obligations regarding take-or-pay liabilities under contracts between those pipelines and their suppliers, including any right to raise force majeure as a defense in an appropriate case.” J.A. 1231.
. The Commission has no power under § 5 to set aside or modify clauses in producer contracts relating to nonjurisdictional gas. See Pennzoil Co. v. FERC, 645 F.2d 360, 380-83 (5th Cir.1981), cert. denied, 454 U.S. 1142, 102 S.Ct. 1000, 71 L.Ed.2d 293 (1982).
. We note that the effectiveness of a solution may be a relative matter. If the Commission on remand should adopt some access condition that realistically promises to solve the problem of troublesome producer-pipeline contracts, that would alter the context of its determination whether specific take-or-pay clauses in jurisdictional contracts are unjust or unreasonable.
. See T.A. 553. The regulations adopted in Order No. 436 provide that projections of units of service may never be decreased, § 157,-103(d)(4), i.e., the pipeline cannot secure a rate increase by demonstrating that service has fallen below expectations; that costs originally allocated to the new service may be shifted only on a showing that they are being incurred for the benefit of other services, § 157.103(d)(8); and that the pipeline may not, for sales service (firm or interruptible), impose any charge tending to guarantee revenue, such as a demand charge or a reservation fee, § 157.103(e).
If the applicant seeks a certificate to provide transportation for others under the OEC procedure it must also apply for a blanket certificate under § 284.221, thereby exposing its entire *51[The doctrine’s] basic rationale is to prevent the courts, through avoidance of premature adjudication, from entangling themselves in abstract disagreements over administrative policies, and also to protect the agencies from judicial inter-transportation network to the conditions of Order No. 436, such as nondiscriminatory access and CD adjustment. Once a pipeline has obtained a blanket transportation certificate, it would no longer need case-specific certification.
As a practical matter this appears to render the OEC route useful only in obtaining a certificate to provide sales service, or, more importantly, to construct or acquire new facilities.
. Compare § 157.104(b) with § 385.802.
. Petitioners appear to make no claim that the Commission was not procedurally free to use rulemaking to identify in advance the criteria on which it would grant or deny certification. It is, of course, far too late in the history of administrative law for any such claim to succeed. In both FPC v. Texaco Inc., 377 U.S. 33, 84 S.Ct. 1105, 12 L.Ed.2d 112 (1964), and United States v. Storer Broadcasting Co., 351 U.S. 192, 76 S.Ct. 763, 100 L.Ed. 1081 (1956), the Court upheld the use of rulemaking to identify necessary (but not sufficient) criteria for licensure. Here the Commission has identified criteria that are conditionally sufficient for certification. We see no difference in principle, and indeed we upheld such a provision in National Tour Brokers Ass'n v. ICC, 671 F.2d 528 (D.C.Cir. 982).
. See S.Rep. No. 948, 77th Cong., 2d Sess. 7 (1942); see FPC v. Transcontinental Gas Pipe Line Corp., 365 U.S. 1, 13 n. 9, 81 S.Ct. 435, 442 n. 9, 5 L.Ed.2d 377 (1961).
. The quoted language, taken from the original version of NGA § 7(c), refers to prices for gas sold interstate for resale, but the purpose presumably also applies to the pricing of interstate gas transportation.services.
. ANR Pipeline Company and Colorado Interstate Gas Company object to 18 C.F.R. § 157.-103(d)(4), which provides that a pipeline offering OEC service may never decrease the units projected for ratemaking purposes below those projected in the original filing. They say that this defies standard principles of ratemaking, under which projections are altered to fit changed circumstances in order to prevent unlawful impairment of the regulated firm's capital. At least where the service is initiated under special rules promulgated clearly in advance, and seeking to mimic market constraints on firm behavior, we see no illegality.
. Those challenges not discussed in the opinion appear to us duplicative of others or frivolous and are dismissed.
. The only elaboration the rule provides on the exact nexus required to satisfy the "on behalf of standard is a statement that physical receipt is not necessary, but that the gas must be transported "for [the] account" of the benefited entity. J.A. 504-05 (reaffirming Order No. 46, 44 Fed.Reg. 52,179, 52,180 (1979)).
. In this connection we note that it is the practical effect of the "on behalf of distinction that prompts the Industrial End User Group's objection to the more exacting notice provisions applicable to § 7 transportation.
. Laclede Gas Co. also attacks the elimination of the system supply requirement. The attack rests in part on the ground that this action will undermine the incremental pricing provisions of the NGPA, which we reject for the reasons given infra part VII.G. It further rests on an asserted negative inference from the Public Utilities Regulatory Policies Act, Pub.L. No. 95-617, § 608, 92 Stat. 3117 (1978), which was adopted the same day as the NGPA and amended § 7(c) to authorize the Commission to use a rulemaking to permit transportation for high-priority end users. That Congress especially authorized such an approach for high-priority users may reflect no more than that it thought of them as the most likely recipients of unbundled transportation; the case for applying expressio unius est exclusio alterius seems weak in this context.
. For December 1986, alternative fuel price ceilings ranged from a low of $1.23 in Washington to a high of $1.95 in Minnesota, Nebraska, North Dakota and South Dakota. Alternative Fuel Price Ceilings by State, by Month, Natural Gas Monthly 64 (Jan. 1987). For the same period, the pipelines’ average cost for gas purchased from producers was $2.16. Selected National Average Natural Gas Prices, Natural Gas Monthly 10 (Jan. 1987). Thus it appears impossible for the pipelines to apply a surcharge in today’s market.