Opinion PER CURIAM.
PER CURIAM:We review in this proceeding Federal Communications Commission (FCC or Commission) decisions focused on the future of United States interstate telephone services. In the Commission’s view, this case presents perhaps “the most difficult” and probably “the most important” problem ever to come before the agency. MTS & WATS Market Structure: Third Report and Order 11368, 93 F.C.C.2d 241, 340-41 (1983). The decisions at issue, we conclude, are within the Commission’s authority and, for the most part, are rationally grounded and sufficiently supported by evidence. We therefore affirm the FCC’s orders in all major respects., We remand to the agency for further, more careful, analysis only two portions of its orders: the segments dealing with party line service and small telephone companies’ election of "average schedule company” status.
I. Background
A. AT & T and the Separations Process
The multiple petitions consolidated for our review address facets of a controversial, compound question: Among telephone users, how should the costs of local telephone company equipment be divided. That equipment starts at every subscriber’s wall plug; it includes the line, or “loop,” between each subscriber’s premises and the local telephone company central office. Switching equipment at the office routes each incoming call out onto the local loop of the subscriber receiving the call, or out to another local office where the call may be switched onto the long-distance *399lines of AT & T or another long-distance carrier.
A large part of the cost of this local plant is nontraffic sensitive (NTS). Plant costs are nontraffic sensitive when they do not vary with the extent to which the facilities are used. The basic cost of installing and maintaining a local loop, for example, remains the same whether the subscriber, or “end user,” uses the loop to make one call or a hundred, and whether those calls are local or long-distance. Some switching costs, on the other hand, are traffic-sensitive. They in fact increase with usage; for example, as more calls pass through the equipment, heavier, more costly switches must be employed.
In the days when AT & T was the only interstate long-distance carrier, the recovery of telephone equipment costs was not the controversial matter it is today. At first, local telephone companies 1 recovered all local exchange plant costs, for the most part through flat per-month charges paid by local subscribers.2 The long-distance carrier (AT & T for interstate calls, perhaps another company for intrastate long-distance calls) recovered the costs of long-distance or “toll” lines and long-distance switching equipment through usage-sensitive charges imposed on the makers of long-distance calls — the more and farther one called, the more one paid. Where the local carrier owned long-distance property — for example, toll lines out to the city limits — the long-distance carrier reimbursed the local company for the use of that property through a share of long-distance revenues.3 See, e.g., Illinois Bell Telephone Co. v. Moynihan, 38 F.2d 77, 82-83 (N.D.Ill), rev’d sub nom. Smith v. Illinois Bell Telephone Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930); Re Indiana Bell Telephone Co., 1922C Pub.Util. Rep. 348, 368.
All long-distance calls, however, require the use of both local property and long-distance facilities. The calls begin on some subscriber's local loop; they then travel through local switches on their way out to long-distance lines; from the long-distance lines, they drop back into a local system at the receiving end and pass through local switches; finally, they pass onto the call recipient’s local loop. In Smith v. Illinois Bell Telephone Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930), the Supreme Court decided that, because local plant is used for interstate calls, an appropriate percentage of local plant costs should be placed within the jurisdiction of federal4 rather than state regulators. AT & T and the local companies then adjusted their cost-allocation system to accommodate Smith.
State regulators, after Smith, could authorize local companies to recover only the portion of local telephone plant costs allocated to the intrastate jurisdiction. AT & T recovered local telephone plant costs allocated to the interstate jurisdiction and passed those revenues back to the local companies.5 AT & T chose to recover local *400telephone costs assigned to the interstate jurisdiction in the same way it had all along recovered costs associated with interstate service — through usage-based charges imposed on the.makers of long-distance calls. Thus long-distance callers, charged on the basis of the frequency and distance of their calls, covered through their payments a significant portion of the costs of local subscriber plant. Revenues paid in by long-distance callers were shared by AT & T with the local companies through a process called settlements and division of revenues.
That basic system remains in effect today. The FCC, working with a Federal-State Joint Board established pursuant to 47 U.S.C. § 410(c) (1976), allocates local plant costs between the interstate jurisdiction (FCC controls recovery of costs) and the intrastate jurisdiction (state commissions control recovery of costs). This mode of allocation — the “separations process”— currently assigns roughly 26% of the costs of local exchange plant to the interstate jurisdiction. See Amendment of Part 67, 89 F.C.C.2d 1, 5, modified, 90 F.C.C.2d 522, recon. denied, 91 F.C.C.2d 558 (1982), petition for review pending sub nom. MCI Telecommunications Corp. v. FCC, No. 82-1237 (D.C.Cir. filed Mar. 4, 1982); see also MCI Telecommunications Corp. v. FCC, 712 F.2d 517, 523 & n. 4 (D.C.Cir. 1983).
Two key characteristics of the system bear emphasis. First, local charges do not cover the full costs of local telephone facilities. Today, local charges cover only 74% of the costs of the basic local network. The rest of the local plant costs are recovered from long-distance fees paid by long-distance callers on a traffic-sensitive basis. Second, subscribers who are heavy long-distance users, under the current, usage-based charges, pay a percentage of the costs of the local network wholly out of proportion to the costs of supplying them with service. These subscribers are heavy users of their local loops, but the basic cost of a local loop is nontraffic sensitive — that cost remains the same regardless of how many, or how few, calls a subscriber makes.
B. The Growth of Alternatives to Ordinary Long-Distance Service
Ordinary long-distance service is not the only way to arrange calls from one state to another. AT & T and other carriers offer a variety of “private line” arrangements. Private line services furnish to the large-scale user, for a flat rate, full-time private interstate circuits between specific points. Foreign Exchange (FX), for example, is one form of specialized private line service. Essentially, FX users have a private line connected to the local exchange at one end. A Washington business buying Washington-New York FX service with the “closed end” in Washington, under the current scheme, can call any telephone subscriber in New York, without paying an additional per-call charge, and any New York telephone subscriber can call the business in Washington for the price of a local call. Common Control Switching Arrangement (CCSA) is another specialized private line service. CCSA involves a network of private lines linked through switches at a local telephone company’s premises. See Bell System Tariff Offerings, 46 F.C.C.2d 413, 418 n. 5, aff'd sub nom. Bell Telephone Co. v. FCC, 503 F.2d 1250 (3d Cir. 1974), cert. denied, 422 U.S. 1026, 95 S.Ct. 2620, 45 L.Ed.2d 684 (1975).
The FCC, prior to the set of decisions before us for review, did not require users of private line services to contribute, as ordinary long-distance callers do, to local plant costs allocated to the interstate jurisdiction. Users of these services, however, may in fact place long-distance calls originating and/or terminating in the local exchange. For example, every FX call is an interexchange call originating or terminating in a local exchange. FX users now pay ordinary business line rates (as well as a flat fee for the private line) and thus contribute to local plant costs allocated to the intrastate jurisdiction; currently, however, they make no contribution to the local plant *401costs allocated to the interstate jurisdiction. A further illustration is the “leaky PBX.” The “leaking” becomes possible when a private line user hooks up its private line to an ordinary local loop through a switchboard, or “PBX,” at one or both ends. A business with a New York-Washington private line can thus dial up from any Washington exchange telephone to the Washington switchboard, dispatch the call to New York via the private line, and drop it back out into the New York local exchange through the switchboard on the other end. Here too, the caller currently makes no contribution to local plant costs allocated to the interstate jurisdiction.
Choice increased when other common carriers (OCCs) entered the market once served only by AT & T. The Commission’s decision in Specialized Common Carrier Services, 29 F.C.C.2d 870 (1971), aff'd sub nom. Washington Utilities & Transportation Commission v. FCC, 513 F.2d 1142 (9th Cir.), cert. denied, 423 U.S. 836, 96 S.Ct. 62, 46 L.Ed.2d 54 (1975), paved the way for private line market entry by carriers offering services in competition with AT & T. Subsequent FCC and appellate court decisions clarified that AT & T could not, through its subsidiaries, the Bell Operating Companies (BOCs), block competition in private line offerings, such as FX and CCSA, by denying OCCs loop service through which they could interconnect their private line circuits into the local exchange. See Bell System Tariff Offerings, 46 F.C.C.2d 413, aff'd sub nom. Bell Telephone Co. v. FCC, 503 F.2d 1250 (3d Cir. 1974), cert. denied, 422 U.S. 1026, 95 S.Ct. 2620, 45 L.Ed.2d 684 (1975).
In 1974 MCI filed tariffs for a new service it called Execunet. With Execunet, MCI linked together a nationwide network of intercity private lines and local interconnections to offer a service that could compete with AT & T’s ordinary long-distance service. An Execunet customer could enter the MCI network from any local phone in an area served by the network and, after entering a subscriber authorization code, dial an ordinary long-distance number. The call would travel over MCI private facilities to the appropriate local exchange, drop back into the local exchange, and be routed along local lines to its recipient. This court, in its Execunet I and II decisions, held that Execunet service was not beyond the scope of MCI’s private line authorizations; that the Commission had not made the explicit public interest findings necessary to bar OCCs from offering such service; and that, unless and until the Commission made such findings, local telephone companies were required to give the OCCs appropriate interconnections for the service. MCI Telecommunications Corp. v. FCC, 561 F.2d 365 (D.C.Cir.1977), cert. denied, 434 U.S. 1040, 98 S.Ct. 780, 54 L.Ed.2d 790 (1978) (Execunet I); MCI Telecommunications Corp. v. FCC, 580 F.2d 590 (D.C.Cir.), cert. denied, 439 U.S. 980, 99 S.Ct. 566, 58 L.Ed.2d 651 (1978) (Execunet II).
After the Execunet decisions, the FCC opened Common Carrier Docket No. 78-72 to determine “whether the public interest requires that interstate message toll telephone service (MTS) [i.e., ordinary long-distance service] and/or wide area toll telephone service (WATS), or their functional equivalents, should be provided ... free from direct competition[ ].” MTS & WATS Market Structure: Notice of Inquiry and Proposed Rulemaking, 67 F.C.C.2d 757, 757 (1978) (footnotes omitted). The Commission announced that it would consider, as part of its analysis,
what reimbursement interstate services should make to local operating companies for the use of local plant, on a cost causational basis; what additional charges, if any, should be levied on interstate services to support local exchange services; and whether and how these charges can be equitably imposed on all interstate services.
Id. at 759.
During the pendency of the Execunet litigation, the OCCs and their subscribers made no contribution to local plant costs allocated to the interstate jurisdiction. In 1978, however, the BOCs filed a new tariff *402with the FCC under which OCCs would pay for Exchange Network Facilities for Interstate Access (ENFIA) at rates designed to parallel the contribution AT & T made to the costs of local plant through the settlements and division of revenues process. The OCCs, the BOCs. claimed, were using local exchange plant to make interstate calls in the same manner as AT & T — yet the OCCs made no special contribution to the costs of that plant, as AT & T did. The ENFIA tariff was the remedy the BOCs proffered for this disparity.
The OCCs protested vigorously. Their interconnections into the local exchange, they argued, were simple FX interconnections to the “line side” of a local central office switch. FX users had never contributed to the interstate portion of local plant costs. AT & T’s interconnection to the “trunk side” of a toll office switch, the OCCs pointed out, was different and, at the time of the ENFIA tariff filing, technologically available only to AT & T. “Trunk side” connection, the OCCs maintained, was far superior to “line side” linkage. Because of the unique quality of AT & T’s interconnection, the OCCs argued, AT & T was the only true provider of ordinary long-distance services. Thus, in the OCCs’ view, AT & T should shoulder alone the costs of local exchange plant allocated to the interstate jurisdiction.
To cope with this dispute pending eventual action in Docket No. 78-72, the FCC in 1978 supervised several months of negotiations. The Commission’s objective was to achieve “some form of a ‘rough justice’ interim” solution. AT & T, 91 F.C.C.2d 1079, 1081 (1982). The negotiations culminated in December 1978 with the signing of the ENFIA agreement, a settlement due to expire, by its terms, upon the effective date of a superseding Commission order in Docket No. 78-72, or after five years, whichever came first. Under the ENFIA agreement OCCs providing Execunet-type services made “payments] to the local telephone company for use of jointly used subscriber plant ... in the provision of their interstate services.” ENFIA Agreement II9, reprinted in 43 Fed.Reg. 59,129, 59,-131 (1978). These payments, however, were far below those AT & T made through the settlements and division of revenues process.
Parties to the ENFIA agreement remained mindful of the disparate treatment of Execunet-type services in comparison to other interstate services, whose providers paid only local business rates for connections to local exchange facilities. They were mindful also that an attempt to arrive at a more encompassing settlement might have impeded negotiations. See MCI Telecommunications Corp. v. FCC, 712 F.2d 517, 524 (D.C.Cir.1983). The ENFIA settlement therefore stated that the parties reached their interim agreement on compensation of local telephone companies for use of local exchange facilities “as if” the charges they established “could, after appropriate consideration by the Commission, be ultimately applied to other services which also utilize the local telephone company exchange facilities.” ENFIA Agreement II 5, reprinted in 43 Fed.Reg. at 59,-131.
C. Proceedings in Docket No. 78-72
The Commission did not linger over the basic market structure issue in Docket No. 78-72; it concluded that an open market and free competition were “in the public interest and will further the goals of the Communications Act.” MTS & WATS Market Structure: Report and Third Supplemental Notice of Inquiry, 81 F.C. C.2d 177, 183 (1980) (Third Supplemental Notice). Turning to the further issues posed in the docket, the Commission, after four supplemental notices of inquiry, ultimately released MTS & WATS Market Structure: Third Report & Order, 93 F.C. C.2d 241 (1983) (Access Order), the first of the series of FCC orders presented for our review. The Commission twice modified the Access Order plan initially and most substantially on reconsideration, MTS & WATS Market Structure: Memorandum Opinion & Order, 48 Fed.Reg. 42,984 (1983) (Reconsideration Order), and again on further reconsideration, MTS & WATS *403Market Structure: Memorandum Opinion & Order, 49 Fed.Reg. 7,810 (1984) (Further Reconsideration Order). All three orders are before us in this appellate proceeding.
In its Docket No. 78-72 dispositions, the Commission sought to accommodate four goals that tugged in different directions. “[T]he continued assurance of universal service” appeared first on the FCC’s list. Access Order II122, 93 F.C. C.2d at 278. Section 1 of the Communications Act requires the FCC to “regulat[e] interstate ... commerce in communication by wire and radio so as to make available, so far as possible, to all people of the United States a rapid, efficient, Nationwide and worldwide wire and radio communication service with adequate facilities at reasonable charges.” 47 U.S.C. § 151. Our case law recognizes the prominence of this universal service objective. See, e.g., United States v. Western Electric Co., 569 F.Supp. 1057, 1120 (D.D.C.1983); see also Access Order ¶1¶ 74-84, 93 F.C.C.2d at 265-67.6 The other three objectives identified by the Commission were “the elimination of unjust discrimination or unlawful preferential rates,” as mandated by section 202 of the Communications Act; “the encouragement of network efficiency”; and “the prevention of uneconomic bypass.” Access Order 11122, 93 F.C.C.2d at 278.
“Bypass” occurs when end users develop and employ alternative local distribution technologies in place of the local exchange system. The Commission considered certain forms of bypass a grave risk. It feared, particularly, that the current system for recovery of subscriber plant costs might lead to “uneconomic bypass” — that heavy users might turn to bypass technologies priced lower than local exchange facilities, but in fact costing more to provide. The Commission explained that “uneconomic” technologies posed a threat to the local telephone network when, as under the current system of charges, access to the local telephone network, for heavy interstate users, is priced above cost. This ultimate concern influenced the FCC’s course: if large users left the network and turned to bypass technologies, the local companies would have to raise the rates paid by their remaining subscribers, thus jeopardizing universal service.
The Commission determined that its various objectives could be balanced most effectively if it moved toward a system in which a substantial portion of the NTS costs of local telephone plant within the FCC’s jurisdiction would be recovered through flat per line charges billed to end users. The FCC targeted for elimination heavy long-distance users’ payment of a share of local telephone plant costs in excess of the actual cost of supplying those users with service. A critical factor, sometimes overlooked in discussions of the cost-allocation problem the FCC faced, was spotlighted by the Commission: “A subscriber who does not use the subscriber line to place or receive calls imposes the same NTS costs as a subscriber who does use the line.” Access Order ¶ 121, 93 F.C. C.2d at 278. It should be the main rule, the FCC decided, that subscribers bear responsibility for the local telephone network costs they actually cause. Explaining its position, the Commission stated: *404non-usage sensitive charges for recovery of access costs.
*403Economics teaches us that, except in certain circumstances involving market failure, prices equal to the cost of producing another increment of a good, i.e., equal to the marginal cost of production, are optimal. Provision of telephone services involves two marginal costs. One varies with the traffic level. The other varies with the number of access lines demanded. For this reason, efficient pricing requires both usage sensitive and
*404The cost imposed upon the nation’s telecommunications system, and ultimately upon the general public, by our present usage sensitive method of recovering these NTS costs pose[s] a substantial danger to the long term viability of our nation’s telephone systems. New technologies and radical improvements in older technologies make available alternatives to the traditional telephone network. Telecommunications is substitutable for a wide variety of other goods and services produced by our society. Prices based upon the true cost characteristics of telephone company plant are necessary both to make a decision on whether use of the alternative technologies is appropriate and to make a decision on whether to substitute telecommunications for other activities.
As telecommunications plays a larger and larger role in fundamental U.S. industries, the problems resulting from inappropriate pricing grow. Computer technology and communications have grown so similar that the Commission has redrawn its traditional definition of communications. Access pricing that does not reflect cost can turn computer technologies from directions that would enhance the productivity of this essential U.S. industry and all of the industries that depend on computers and communications toward simple avoidance of non-cost based telecommunications prices. Investment may be misdirected as a result.
Id. ¶¶ 27-29, 93 F.C.C.2d at 251-52 (footnotes and paragraph numbers omitted).
The FCC acknowledged that “uncertainty surround[ed] the precise size and threat of uneconomic bypass,” but found the bypass phenomenon real and growing. Id. if 31, 93 F.C.C.2d at 252. It would be unwise, the Commission concluded, to delay implementation of a cost-based pricing system until the effects of bypass became more pronounced. Delay risked losing “the luxury of the gradual transition needed to satisfy our objective of maintaining affordable service”; any further bypass “might mean higher long run costs for those who were required to remain on the network.” Id. Of prime importance in its decisionmaking, the Commission observed:
[W]ere we to delay instituting the smooth movement towards a rational pricing system until a significant number of large users had initiated constructing alternative bypass systems, it could well be too late for any remedial action. Usually uneconomic bypass is uneconomic only before the construction of bypass facilities starts. Once a large telecommunications user has committed significant capital to building a private bypass system, the maintenance of that system is no longer uneconomic. Consequently, we believe that prompt action is essential to preserve the public interest.
Id. ¶ 32, 93 F.C.C.2d at 252-53.
Under the FCC’s plan to shift most subscriber plant costs in the interstate jurisdiction onto end users, a $6 end user charge will be imposed on multi-line business customers starting in June 1984. The Commission now anticipates imposing some end user charges on residential and single-line business users, on a transitional basis, starting in June 1985. The initial Access Order contemplated recovery through flat-rate end user access charges of all NTS subscriber plant costs except those allocated to a carrier-supported Universal Service Fund, designed to keep down local rates in areas where exchange plant costs are especially high. Access Order H 134, 93 F.C. C.2d at 281-82. On reconsideration, the Commission chose to institute further decisionmaking on several matters: the magnitude of the end user charge after the transition; categories of low-income subscribers who should be exempted from end user charges; the shape of the transition; and mechanisms sensitive to the particular needs of small telephone companies in high cost areas. The residential and single-line business end user charge will not exceed $4 per line per month through 1990, and the FCC has announced its intention to monitor closely the impact of residential end user *405charges on universal service as those charges are imposed. Further Reconsideration Order ¶¶ 15-26, 49 Fed.Reg. 7,810, 7,812-13 (1984).
The FCC has allocated the jurisdictionally interstate NTS costs of local telephone plant that it is not shifting to end users— or, for the transitional period, that it is not yet shifting to end users — to interexchange carriers and to users of interstate private line services. Interexchange carriers are to pay a “carrier’s carrier” charge; a surcharge has been established for private line users. The interexchange carriers (most prominently, AT & T and the OCCs) will in turn pass their carrier’s carrier charge on to their customers through the long-distance rates they set. The modified final judgment settling the government’s antitrust suit against AT & T, see supra note 1, requires the BOCs to move, over the next few years, to provision of equal interconnection to the OCCs; to the extent that the same quality interconnection is unavailable in the transitional years, AT & T’s carrier’s carrier charge will be maintained at a “premium” above the charge paid by the OCCs. The interim surcharge set by the Commission is to be paid by the following category of users: private line and other users who may engage local exchanges for interstate calling without otherwise contributing to the portion of local telephone plant costs assigned to the interstate jurisdiction.
D. Review of the FCC’s Orders
Our review of the FCC’s multifaceted decision first considers arguments relating to the Commission’s imposition of flat-rate end user access charges. Under this main heading, we deal with (1) contentions that the imposition of end user charges exceeds jurisdictional limitations on the Commission’s authority implicit in the Communications Act and/or the Supreme Court’s decision in Smith v. Illinois Bell Telephone Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930);7 (2) assertions that the FCC’s decision is irrational, unsupported, or procedurally defective;8 and (3) attacks on the end user charges the Commission ordered for Centrex-CO subscribers9 and party-line users.10 We also place in this portion of the opinion review of the Commission’s changes in the prerequisites for small telephone companies’ election of “average schedule company” status.11
In the second main segment of the opinion, we address objections various petitioners raise to the carrier’s carrier charge and private line surcharge. These objections include claims that (1) the FCC’s plan perpetuates unlawful discrimination between OCCs and other entities using the local exchange for interstate calling;12 (2) the FCC’s prescription of a surcharge payable *406by private line and other users lacks evidentiary support or rational basis;13 (3) the FCC’s action regarding private communications systems is unjustified;14 and (4) errors warranting reversal infect the FCC’s treatment of FX users.15
II. Analysis
A. Flat-Rate End User Access Charges
1. FCC Authority to Impose the Charges
Petitioners16 present, elaborate, and recapitulate their core argument that the Commission lacks authority to impose flat-rate end user access charges. They derive their contentions from precedent and statutory provisions. We turn first to their insistent claim that the Supreme Court’s decision in Smith v. Illinois Bell Telephone Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930), see supra p. 1104, confirms the powerlessness of the federal agency to take the action they challenge. Petitioners misapprehend the Court’s holding in Smith. We here describe that case in sufficient detail to clarify the precise thrust of the High Court’s opinion.
Smith presented for Supreme Court review a district court decision striking down as confiscatory Chicago coinbox rates set by the Illinois Commerce Commission. The district court, like the state commission before it, had taken as a rate base all of Illinois Bell’s Chicago property, including both exchange plant and toll (long-distance) lines out to the city limits. In computing the revenue generated by that investment, the district court counted both the sums Illinois Bell received directly from local users and the share of interstate tolls AT & T paid over for the use of Illinois Bell’s long-distance lines in interstate calling.
The Supreme Court reversed. It held first that the state commission and district court erred in not separating out Illinois Bell’s intrastate and interstate property, revenue, and expenses. Chief Justice Hughes wrote:
The separation of the intrastate and interstate property, revenue and expenses of the Company is important not simply as a theoretical allocation to two branches of the business. It is essential to the appropriate recognition of the competent governmental authority in each field of regulation. In disregarding the distinction between the interstate and intrastate business of the Company, the court found it necessary to pass upon the fairness of the division of interstate tolls between the American and Illinois companies. The court held that the division was reasonable and the appellants contest this ruling. But the interstate tolls are the rates applicable to interstate commerce, and neither these interstate rates nor the division of the revenue arising from interstate rates was a matter for the determination either of the Illinois Commission or of the court in dealing with the order of that Commission.
282 U.S. at 148, 51 S.Ct. at 68 (emphasis supplied). The Court stated that it was the ICC, then charged with regulating interstate communication, see supra note 4, that had “authority to estimate the value of the property used in the interstate service and to determine the amount of the revenues and the expenses properly attributable thereto”; “[t]he proper regulation of rates can be had only by maintaining the limits of state and federal jurisdiction.” Id. at *407149, 51 S.Ct. at 69. Any review of the state commission's order, the Court continued, must therefore rest on an appropriate determination of “the value of the property employed in the intrastate business and of the compensation receivable for the intrastate service.” Id. The Court cited the Minnesota Rate Cases, 230 U.S. 352, 435, 33 S.Ct. 729, 755, 57 L.Ed. 1511 (1913), in which it had imposed a similar constraint on the states’ determination of intrastate rail rates.17
The requisite allocation of property between the interstate and intrastate services, the Court then stated, must be made with an eye to “the actual uses to which the property is put.” 282 U.S. at 151, 51 S.Ct. at 69. Figures Illinois Bell had submitted to the district court reflected treatment of the costs of exchange plant as wholly local. Id. at 150, 51 S.Ct. at 69. That allocation was impermissible, the Court declared, “for unless an apportionment is made, the intrastate service to which the exchange property is allocated will bear an undue burden.” Id. at 151, 51 S.Ct. at 69.18
One petitioner argues that Smith precludes FCC-imposed flat-rate end user charges because the imposition means “costs attributable to interstate usage [are not] allocated to and recovered from the interstate network.” Brief of Petitioner National Association of Regulatory Utility Commissioners at 28 (hereafter, NARUC Brief). Another asserts that the FCC’s plan “essentially repeals” Smith’s command to allocate exchange plant between the interstate and intrastate jurisdictions. Brief of Intervenor National Association of State Utility Consumer Advocates, et al. at 13 (hereafter, NASUCA Brief). See also Brief of the People of the State of California and the Public Utilities Commission of the State of California at 11 (hereafter, California Brief); Brief of Intervenor Florida Public Service Commission at 16 (hereafter, Florida Brief).
Petitioners confuse or blend two questions: (1) jurisdiction or authority to recover costs; (2) the manner in which costs are to be recovered. Smith dealt with jurisdiction; it held that a portion of the costs of local subscriber plant may be recovered only under the authority of a body with interstate regulatory powers. The Smith Court did not address the manner in which the federal agency was to perform its task. It did not hold that the FCC must order recovery of costs allocated to its jurisdiction through usage-based charges. The practical effect of the Smith decision in 1930, it is true, was a system under which subscriber plant costs in the interstate jurisdiction would be recovered on a usage basis. But nothing in Smith mandated that result; other plans under which those costs were subject to federal, rather than local, regulatory authority might have served as well.19
*408Coupled with their exorbitant reading of Smith v. Illinois Bell Telephone Co., petitioners assert that section 2(b) of the Communications Act,20 which denies Commission jurisdiction over purely intrastate communications, precludes the FCC’s imposition of flat-rate end user charges.21 Those charges, petitioners say, are in fact for intrastate, not interstate, service. They must be paid to receive any telephone service; even subscribers who neither make nor receive interstate calls in the billing period must pay. Thus, petitioners conclude, the charges should be regarded as a “local access” toll, an intrastate rate, a directive the FCC lacks jurisdiction to impose. See Brief of Intervenor New York State Department of Public Service at 9-10, 19 (hereafter, New York Brief); California Brief at 11; Florida Brief at 18.
Petitioners here lose sight of the Commission’s main theme. The end user charge reflects costs caused not by a subscriber’s actually making interstate calls, but by the subscriber’s connection into the interstate network, which enables the subscriber to make interstate calls. The same loop that connects a telephone subscriber to the local exchange necessarily connects that subscriber into the interstate network as well. Under Smith, a portion of the costs of that loop are assigned to the interstate jurisdiction, for recovery under the regulatory authority of the FCC, on the basis of a complex division taking into ac*409count statistical calling patterns. That separations decision, however, does not affect the cost of the loop. Local telephone plant costs are real; they are necessarily incurred for each subscriber by virtue of that subscriber’s interconnection into the local network, and they must be recovered regardless of how many or how few interstate calls (or local calls for that matter) a subscriber makes. The FCC may properly order recovery, through charges imposed on telephone subscribers, of the portion of those costs that, in accordance with Smith, have been placed in the interstate jurisdiction.
In North Carolina Utilities Commission v. FCC, 552 F.2d 1036 (4th Cir.), cert. denied, 434 U.S. 874, 98 S.Ct. 222, 54 L.Ed.2d 154 (1977) (NCUC II), the Fourth Circuit reviewed a Commission order setting conditions under which terminal equipment (including home telephones, answering devices, and switchboards) could be connected to local telephone company lines. The order covered, inter alia, subscribers who neither made nor received interstate calls. Petitioners in NCUC II argued that the Commission had impermissibly invaded the intrastate jurisdiction; the FCC, they objected, had set rules for use of equipment needed, and used dominantly — sometimes exclusively — for local calls. The Fourth Circuit rejected the argument that the FCC had trespassed on state territory. The terminal equipment in question, it noted, was used for both interstate and intrastate communication. “The withdrawal of [FCC] jurisdiction over one,” the court stated, “cannot be read to mean the withdrawal as to the other,” 552 F.2d at 1046, nor could the fact that the phones were necessary for local calling divest the FCC of its “paramount” interstate regulatory authority. Id. at 1043 (describing holding of North Carolina Utilities Commission v. FCC, 537 F.2d 787 (4th Cir.), cert. denied, 429 U.S. 1027, 97 S.Ct. 651, 50 L.Ed.2d 631 (1976) (NCUC I)). The court concluded:
Petitioners confuse the fact that almost all terminal equipment is and has been used predominantly for local communication, with the statutory division of decisionmaking power. We find it difficult to credit an argument which amounts to an assertion that Congress created a regulatory scheme that depends on the calling habits of telephone subscribers to determine the jurisdictional competence of the FCC versus state utility commissions.
552 F.2d at 1046.
We endorsed the Fourth Circuit’s NCUC I and II reasoning in Computer and Communications Industry Association v. FCC, 693 F.2d 198 (D.C.Cir.1982), cert. denied, 461 U.S. 938, 103 S.Ct. 2109, 77 L.Ed.2d 313 (1983) (Computer II). That case presented for review a Commission order detariffing terminal equipment (there described as customer premises equipment, or CPE), including home telephones, and preempting all state rate regulatory authority in the area. Petitioners in Computer II asserted that “the Commission’s decision to order the states to remove CPE charges from their tariffs is an unjustifiable invasion of the authority to regulate intrastate communications services reserved to the states by the Act.” 693 F.2d at 214. The NCUC II court had noted that it did not have before it an FCC attempt “to control the rates for exclusively local service.” 552 F.2d at 1047 (emphasis in original). We rejected an attempt to distance NCUC II from Computer II on that ground, and said we saw no reason to distinguish between “preemption principles applicable to state ratemaking authority and those applicable to other state powers.” 693 F.2d at 215, 216.
In Computer II the FCC asserted exclusive rate regulatory authority over CPE; we upheld its order, although the equipment in question is necessary to make local calls, and although some subscribers might use their CPE only to make local calls. Similarly here, the FCC asserts authority to condition end user access to the local telephone plant on the end user’s payment of plant costs in the FCC’s interstate domain, although the plant in question is necessary to make local calls, and although some subscriber’s might use it only to *410make local calls. In each case, the very same equipment is by its nature a key element of both interstate and intrastate calling. Every telephone subscriber is automatically connected through the same subscriber plant into both the local exchange and the interstate network. No subscriber can avoid “causing” those costs of its telephone line allocated to the interstate jurisdiction.
Counsel for petitioner NARUC indicated at oral argument that NARUC’s jurisdictional objection would evaporate if a mechanism could be utilized to allow end users who elected not to make or receive interstate calls to escape payment of local telephone plant costs allocated to the interstate jurisdiction. Oral Argument Transcript at 28. This argument would make some sense if a subscriber’s choice not to make interstate calls meant that certain fixed “interstate” costs would not be incurred; if that were in fact the case, it might well be unfair to ask a subscriber who neither made nor received interstate calls to pay those costs. A subscriber’s choice not to make or receive interstate calls, however, would not reduce the costs of that subscriber’s loop; the local telephone plant costs would remain unchanged, as would the need to recover those costs. If we indulged NARUC’s claim — that jurisdictional significance attends an individual subscriber’s decision to use its line entirely for intrastate calls — then, as NARUC’s counsel conceded, NARUC could hardly contest an allocation of all of such a subscriber’s line costs (previously divided between the interstate and intrastate domains) to the intrastate jurisdiction alone. See id. at 22-23. It is hard to see what significant benefit NARUC would gain under such an arrangement.22
The scheme advanced by the FCC simply requires all telephone subscribers to pay, on a per-line basis, for that portion of their necessarily-incurred local telephone plant costs assigned under Smith to the interstate jurisdiction. We cannot sensibly say that the FCC has overstepped the limits of its jursidiction in embarking upon such an arrangement.23
2. The FCC’s Decisional Process
Petitioners unleash a volley of argument attacking the FCC’s decisionmaking process as unreasoned or unreasonable. They claim that the FCC misjudged the threat of economic bypass, inadequately considered the potential harm to universal service, and failed to observe necessary or proper procedural requirements during the rulemaking.
a. Uneconomic Bypass
Most prominently, petitioners say the evidence before the Commission did not warrant an immediate response of the kind the FCC gave to the prospect of uneconomic bypass. If heavy users of interexchange *411services can cut costs through bypass technologies, petitioners concede, the risk of bypass is present. But the Commission should have held back, they contend. Technologically, petitioners maintain, bypass potential has not matured as a market-place force to be reckoned with. “[Regulation perfectly reasonable and appropriate in the face of a given problem,”, petitioners remind us, “may be highly capricious if that problem does not exist.” NARUC Brief at 39 (quoting Home Box Office, Inc. v. FCC, 567 F.2d 9, 36 (D.C.Cir.), cert. denied, 434 U.S. 829, 98 S.Ct. 111, 54 L.Ed.2d 89 (1977)).24
We hold that the FCC’s treatment of the bypass issue is adequately reasoned and indicate here why we have so concluded. We turn first to the staff effort preceding the Commission’s determination.
During the past few years, members of the communications industry have been bombarded with press and trade writings promoting the use and virtues of bypass technology. Status Report on Near-Term Local Bypass Developments 6 (attached to Access Order as Appendix F), reprinted in J.A. 3101, 3106 (hereafter, Appendix F). Many of these materials came to the Commission’s attention in the commenting process. The Commission staff did not swallow the materials whole. It prepared and submitted to the Commission a preliminary report based on its own survey of the industry and the current literature. Id. at 1, reprinted in J.A. 3101, 3101. The report documents the reality of both local and long-haul by-pass. We cite several examples.
Martin Marietta connects its Data Systems Center to its Orlando plant via fiber optic cable; it is currently installing an earth station to bypass the local exchange and make interstate calls directly via satellite. Southern Bell estimates its current gross revenue loss from Martin Marietta’s self-help at an annual $500,000, and its projected annual revenue loss at up to $3 million. Id. at A-3, reprinted in J.A. 3101, 3128. Southern Railway System uses a private microwave system in three states (Alabama, Georgia, Kentucky); South Central Bell estimates its resulting yearly revenue loss at $300,000. Id. Combustion Engineering completely bypasses the local network for long-distance calling; Southern New England Telephone estimates the revenue loss to it at $1.7 million per year. Id. at A-4, reprinted in J.A. 3101, 3129. Textronics operates a digital microwave system linking various of its facilities in Oregon and Washington; Pacific Northwest Bell estimates its revenue losses at $800,000 per year. Id. at A-5, reprinted in J.A. 3101, 3130. The federal government utilizes digital satellite links to connect its Seattle switching network with a California switching node; Pacific Northwest Bell estimates its revenue losses at $1.5 million per year. Id. Boeing uses a digital microwave system linking several of its facilities in the Puget Sound basin, causing Pacific Northwest Bell revenue losses estimated at $2 million per year; Boeing has interconnected that network via satellite to locations in Kansas, Virginia, Pennsylvania, and elsewhere, generating Pacific Northwest Bell revenue losses estimated at $400,000 per year. Id. at A-5, 6, reprinted in J.A. 3101, 3130, 3131.
From its investigation, the Commission staff concluded:
[W]hether or not local bypass is successful, operationally and economically, will be determined in the next three to five years. If the carriers are slow in responding to business users’ needs and bypass proves viable during this period, there could be an irreversible swing to local bypass by a large sector of the Fortune 500 — and by various government agencies as well.
Id. at 4, reprinted in J.A. 3101, 3104.
The FCC’s view was informed by its staff’s “effort[ ] to identify and understand *412bypass activities, technology and trends,” id. at 1, reprinted in J.A. 3101, 3101, and by comments “numerous participants” made. Access Order ¶ 107, 93 F.C.C.2d at 274. These included Comments of Southern Pacific Communications Co. at 17-19, reprinted in J.A. 1729, 1743-45 (OCC describes its usé of microwave radio facilities to bypass local telephone network for local portion of interstate calling); Comments of the Association of Data Communications Users at 18, reprinted in J.A. 1416, 1433 (“ADCU members are currently availing themselves of bypass technologies and will continue to do so as such technologies are further refined”); and Comments of the Western Union Telegraph Co. in Response to Fourth Supplemental Notice of Inquiry and Proposed Rulemaking at 23, reprinted in J.A. 1827, 1848 (“Western Union already has a growing investment of over $100 million in local distribution facilities as part of a specific program to reduce its dependency on the Bell System”) (hereafter, Western Union Comments). AT & T and the BOCs, we note, submitted an extensive study of bypass economics and technologies. Comments of BOCs and AT & T in Response to the Fourth Supplemental Notice of Inquiry and Proposed Rulemaking at 90-105, reprinted in J.A. 1436, 1525-40.
Petitioners characterize much of the bypass data the Commission gathered as anecdotal, and much of the discussion as merely theoretical. The FCC, it is true, had nothing in hand approaching a “definitive analysis.” See Appendix F at 1, reprinted in J.A. 3101, 3101. But Congress has charged the Commission with responsibility to regulate in “a field of enterprise the dominant characteristic of which [is] the rapid pace of its unfolding.” National Broadcasting Co. v. United States, 319 U.S. 190, 219, 63 S.Ct. 997, 1011, 87 L.Ed. 1344 (1943). “[P]ure factual determinations” were not, and could not have been, made; the FCC’s decision inevitably rested on “judgment and prediction.” FCC v. WNCN Listeners Guild, 450 U.S. 582, 594, 101 S.Ct. 1266, 1274, 67 L.Ed.2d 521 (1981). In light of the material available to the Commission, and the leeway the FCC has to make “deductions based on [its] expert knowledge,” FCC v. National Citizens Committee for Broadcasting, 436 U.S. 775, 814, 98 S.Ct. 2096, 2121, 56 L.Ed.2d 697 (1978) (quoting FPC v. Transcontinental Gas Pipe Line Corp., 365 U.S. 1, 29, 81 S.Ct. 435, 450, 5 L.Ed.2d 377 (1961)), we cannot indict the agency’s assessment of the bypass risk as arbitrary or unsupported.
Petitioner NARUC and intervenor Rural Telephone Coalition (RTC) wanted the Commission to deal with uneconomic bypass by denying potential bypassers permission to construct facilities posing the problem. Access Order ¶ 110, 93 F.C.C.2d at 274 & n. 38. The FCC rejected this proposal as unsound and unworkable. It stated:
We are simply not in a position to determine what constitutes an uneconomic “bypass” service and what is a wholly new service that will attract a new set of users and enhance the ability of all users to make full use of telecommunications service-potential. For example, some comments assert that cellular services constitute a bypass technology. We have concluded, however, that cellular radio is a distinct service that serves distinct needs and that cellular service could be complementary to existing wire-line service. Indeed, a given technology may be the efficient means of providing service to certain groups yet constitute uneconomic bypass for other services or groups.
Id. ¶ 111, 93 F.C.C.2d at 275. It was not unreasonable for the FCC to resist an approach that would curtail, and perhaps stifle, development and construction of new technology.25
*413According to intervenor Michigan, the Commission is disarmed by its own self-confessed inability to identify uneconomic bypass. Without any evidence of the existence of the phenomenon the FCC seeks to check, Michigan argues, the agency’s position rests on quicksand. See Reply Brief of Intervenors the State of Michigan and the Michigan Public Service Commission at 2.
We have already underscored the permissibly predictive, judgmental character of the Commission’s stance. See supra p. 1117. The FCC sufficiently established the existence of bypass as a real market force. It sufficiently established both 1) the inability of regulators, under a noncost-based pricing system, to determine accurately the economic efficiency or inefficiency of a given technology, and 2) the likelihood that market participants under such a system will exploit available bypass technology when they can reduce their costs by doing so. Cf. Western Union Comments at 23, reprinted in J.A. 1827, 1848 (“To the extent that access charges are set at excessive levels, there can be little doubt that interstate users will be encouraged to invest in unnecessary and duplicative facilities. This is particularly true in the case of private line users, and Western Union is a case in point.”). As a reviewing court, we would reach beyond our limited range of surveillance if we demanded more.26
Intervenor Florida Public Service Commission faults the FCC for failing to determine the degree to which the price of access to the local network can be set above cost before it will be in the interest of large-scale users to turn to uneconomic bypass. Florida Brief at 5-6. The Commission, however, is engaged in a continuing venture. It is now undertaking the investigation Florida presses. See Further Reconsideration Order ¶ 20, 49 Fed.Reg. at 7,812:
We will also develop and analyze additional information with respect to the extent and dangers of bypass____ [W]e intend to use this additional information to design the transition plan. The end user charges at particular points in the transition should be low enough to avoid any adverse effect upon the universality of service and high enough to produce toll rate reductions that are sufficient to deter uneconomic bypass.
We note finally that the threat of uneconomic bypass entered the Commission’s calculus not as a self-standing justification for the agency’s decision but as a particular— albeit important — feature of the FCC’s general concern that proper allocation of resources requires a pricing system consistent with the lessons “[economics teaches *414us.” Access Order ¶ 27, 93 F.C.C.2d at 251.27
b. Universal Service
We turn next to an objection of a different order. Intervenor RTC challenges the Commission’s conclusion that a cost-recovery plan involving flat-rate end user charges will not have a substantial adverse impact on universal service. The only FCC decision now ripe for our review, however, is the agency’s determination that some access costs should be recovered through end user charges. See Further Reconsideration Order ¶¶ 13-26, 49 Fed.Reg. at 7,812-13.
The FCC has deferred, pending further study, determination of the extent to which NTS subscriber plant costs should be shifted to residential end users. See id. ¶ 19, 49 Fed.Reg. at 7,812 (“additional information with respect to the elasticity of demand for local exchange service ... will ... assist us in determining the maximum charge that should be established at the end of the transition”). In ongoing supplemental proceedings, the Commission seeks “to devise an exemption for persons who cannot afford to pay any end user charge, reevaluate the transition plan for end user charges, and explore alternative mechanisms to assist customers of small telephone companies.” Id. ¶ 4, 49 Fed.Reg. at 7,811.
The FCC has thus far made two firm, correlative determinations: (1) relying on its determination that “demand for exchange telephone service ... is very inelastic,” it has rejected claims that no end user charge can be levied without driving many exchange service subscribers away from the telephone system, id. ¶ 13, 49 Fed.Reg. at 7,812; (2) it has concluded that “[m]ost residential and single-line business customers can ... afford to pay at least a portion of their Common Line costs through fixed charges,” id. ¶ 15, 49 Fed. Reg. at 7,812. We see no sound basis for assailing these determinations as unreasoned or unsupported.
RTC relies primarily on L. Perl, Economic and Demographic Determinants of Residential Demand for Basic Telephone Service (1978). It argues that if the FCC adopts what has been known in this proceeding as a “Pure 2” plan, and if state regulators adopt similar schemes with regard to intrastate long distance rates, then nationwide telephone penetration could decline by as much as eight percent. Brief of Intervenor Rural Telephone Coalition at 7. “Such a decrease,” RTC argues, “can hardly be said to be insignificant by an agency charged with maintaining service at reasonable rates ‘to all the people.’ ” Id. at 7-8.
A “Pure 2” approach, however, would involve immediate recovery of all subscriber plant costs through flat-rate end user charges without any Universal Service Fund, transitional period, or exemption mechanism. The FCC has rejected that approach; it credited arguments that “Pure 2 would constitute a substantial step away from universal service.” Access Order ¶¶ 120-22, 93 F.C.C.2d at 277-78.
In summary, the Commission at this juncture has decided no more than this: some end user charge for residential and single line business users should go into effect on June 1, 1985. The Commission had record support for the view that a *415charge could be imposed without substantial negative effect on universal service. See Analysis of Effects of Federal Decisions on Local Telephone Service, FCC 83-567 (released December 21, 1983). It has not decided the size of the charge, the shape of the transition, or the class of persons affected. When those decisions are made, RTC and other interested entities will be positioned to seek review of the precise plans the Commission adopts.
c. Alleged Procedural Deficiencies
Petitioners raise several procedural objections. NASUCA argues that notice-and-comment proceedings did not suffice. Trial-type hearings were required, NASUCA maintains, because the Commission “prescri[bed] rates and revenue requirements.” NASUCA Brief at 24.
In AT & T v. FCC, 572 F.2d 17, 21-23 (2d Cir.), cert. denied, 439 U.S. 875, 99 S.Ct. 213, 58 L.Ed.2d 190 (1978), the Second Circuit reviewed the FCC’s prescription of unlimited resale and sharing of private line services; it found in the Administrative Procedure Act (APA) and Communications Act no relevant trial-type hearing requirement.28 If AT & T v. FCC is the appropriate guide, we should end our inquiry here, for we have no authority to add to the procedural requirements ordered in the Communications Act and the APA. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978); see also Western Union Telegraph Co. v. FCC, 665 F.2d 1126, 1151-52 (D.C. Cir.1981).
NASUCA appears to urge separation of this case from AT & T v. FCC on two grounds. Even if the Act does not require trial-type hearings for Commission prescription of “practices” as in AT & T, NASUCA first suggests, it does require such hearings when “revenue requirements and rates” are at stake. See NASUCA Brief at 26 n. 2 (emphasis omitted). Second, even if the Act itself does not require trial-type hearings before the Commission may prescribe rates, NASUCA states, “well-settled agency procedures of long standing” do so require; Vermont Yankee, NASUCA observes, does not impede judicial correction of an agency’s “totally unjustified departure” from its own procedures. Joint Reply Brief of Petitioner NARUC, et al. at 27-28 (quoting Vermont Yankee, 435 U.S. at 542, 98 S.Ct. at 1210). NASUCA points to cases in which the Commission has investigated the legality of carrier-filed rate increases, see, e.g., In re AT & T, 57 F.C. C.2d 960 (1976), determined a carrier’s revenue requirements as part of an assessment of the legality of its charges, see In re AT & T, 9 F.C.C.2d 30, modified, 9 F.C.C.2d 960 (1967), or evaluated a carrier request to increase its prescribed rate of return. In re AT & T, 73 F.C.C.2d 689 (1979).
These arguments are not convincing. The Commission referred briefly at one point in its discussion to an “approximate[ ]” revenue requirement for interstate NTS exchange plant. See Access Order ¶ 25, 93 F.C.C.2d at 250. The rules it adopted “to determine the manner in which telephone companies will be compensated for the origination and termination of interstate and foreign communications services,” Further Reconsideration Order ¶ 1, 49 Fed.Reg. at 7,810, included “many of the steps that carriers must follow in order to compute access charges,” Access Order n 43, 93 F.C.C.2d at 256, and some limited direct prescription of charges — for example, the $25 surcharge on private lines. But the FCC prescribed no specific revenue requirement and did not determine the legality of specific rates. On the contrary, the Commission initiated broad-gauged rulemaking to develop policies for the interstate long-distance market and it set prospective, policy-implementing rules.
*416Nor does longstanding FCC procedure chart the Commission’s course in this extraordinary matter. There is no Commission case closely in point establishing a trial-type hearing pattern.29 As the Commission observed five years ago:
We have noted claims that evidentiary hearings will provide the most effective mechanism for accurate fact-finding. However, that type of multi-party proceeding could become almost interminable, given the complexity of the issues which we must resolve. We are endeavoring to establish an entry policy for the 1980’s. A procedural option which would preclude us from reaching a final decision at or near the beginning of that decade would not be consistent with that goal.
MTS & WATS Market Structure: Supplemental Notice of Inquiry and Proposed Rulemaking, 73 F.C.C.2d 222, 232 (1979).
NASUCA stands on somewhat firmer ground in complaining that the Commission did not give the parties an immediate opportunity to comment on .the staff study of bypass. The study was not released until the Commission reached its initial Access Order decision. We have more than once cautioned agencies “that even in an informal [proceeding] parties have a right to be informed of and comment on staff positions.” Independent United States Tanker Owners Committee v. Lewis, 690 F.2d 908, 925-26 (D.C.Cir. 1982) (“[WJhere an agency’s analytic task begins rather than ends with a set of forecasts, sound practice would seem to dictate disclosure of those forecasts so that interested parties can comment on the conclusions properly to be drawn from them.”) (emphasis in original); see United States Lines, Inc. v. Federal Maritime Commission, 584 F.2d 519, 534 (D.C.Cir. 1978) (“[W]e have required information in agency files or reports identified by the agency as relevant to the proceeding to be disclosed to the parties for adversarial comment____ Such requirements ... ensure that parties to agency proceedings are afforded the opportunities guaranteed by statute meaningfully to participate in those proceedings ____”). Disclosure of staff reports allows the parties to focus on the information relied on by the agency and to point out where that information is erroneous or where the agency may be drawing improper conclusions from it. An agency’s denial of a fair opportunity to comment on a key study may fatally taint the agency’s decisional process.
But automatic upset should not attend a fault of the kind that occurred here. See FCC v. WNCN Listeners Guild, 450 U.S. 582, 591 n. 22, 101 S.Ct. 1266, 1272, n. 22, 67 L.Ed.2d 521 (1981); see also Air Transport Association of America v. CAB, 732 F.2d 219 at 223-224 (D.C.Cir. April 20, 1984). Because the Commission gave this case reconsideration and further reconsideration, interested entities eventually had ample opportunity to address the staff study, and the FCC ultimately collected reams of comment. The bypass study should have been released earlier. The Commission’s reconsideration decisions, however, were framed with adversarial comment in full view. Essentially, the error has been rendered harmless. See Independent United States Tanker Owners Committee, 690 F.2d at 926 (importance of allowing parties to comment on staff report “augmented” because of absence of opportunity for reconsideration); United States Lines, 584 F.2d at 535 (“While [disclosure of data and reports relied on by the agency] would ideally appear appropriate at the earliest stage of the agency proceeding, at the very least it is clear that it must come in the final decision so that reconsideration *417may be sought and judicial relief meaningfully afforded.”).30
3. Centrex-CO Service
Centrex-CO is a tariffed service sold by local phone companies. It offers such telecommunications functions as intercom calling, conference calling, direct inward dialing, and automatically identified outward dialing. Reconsideration Order ¶ 43, 48 Fed.Reg. at 42,991. These electronic switching services are performed at the exchange carrier’s central office switch, and have traditionally been offered principally to large organizations and government agencies as an alternative to purchasing or leasing a private branch exchange (“PBX”), which is a smaller switch located on the subscriber’s premises used to provide the same services. Because CentrexCO switching is executed off-premises, each Centrex-CO subscriber station must be connected by a loop to the central office switch. Consequently, Centrex-CO service requires approximately six times more loops than PBX service, which primarily uses inside wiring to link together subscriber stations. Brief of Public Service Commission of the District of Columbia at 12.
The Commission ruled that Centrex-CO lines are generally subject to flat-rate per line access charges, as are other local loops between the subscriber’s premises and the local switch used jointly in exchange and inter-exchange traffic. This ruling effectively imposed substantially higher total access charges on Centrex-CO service, potentially rendering it significantly less economical to subscribers than its principal competitor, PBX. Several parties now seek to overturn the Commission’s decision. Some, such as the Public Service Commission of the District of Columbia (“D.C.P.S. C.”) argue that the Commission treated Centrex-CO subscribers too harshly, threatening the very existence of the service; others, representing the PBX industry, claim that the Commission granted an unlawfully discriminatory preference to Centrex-CO service.
The D.C.P.S.C. claims that the Commission should not have applied a full per line access charge to existing Centrex-CO systems. It argues that the per line fee should be set at only sixteen percent of that levied on other local loops in order to maintain competitive parity with PBX. Id. at 28-29. Failure to levy the access charge in this manner will allegedly cause a large number of subscribers to abandon CentrexCO service, which in turn could threaten universal service since higher phone rates would have to be charged to the remaining exchange subscribers to recover the cost of the prematurely abandoned Centrex-CO plant remaining in the rate base. The D.C. P.S.C. charges that the Commission inadequately considered this problem of “stranded investment.”
However, on the record before us we conclude that these arguments are without merit. The Commission thoroughly considered the objections of the D.C.P.S.C. and other similarly aligned parties below. Although it was generally unwilling to set access rates at a level which would not recover the full cost of Centrex-CO loops, it accepted the claims of state commissions that in many exchanges Centrex-CO rates may have been set above cost in order to subsidize local rates. Since a sharp rate increase in those circumstances could de*418prive the telephone companies of a fair opportunity to adjust local Centrex-CO rates downward to compensate, depriving customers of any benefits that might be derived from efficiently priced, economical Centrex-CO service, the Commission decided to adopt partially the suggestions made by the D.C.P.S.C. It ruled that the full multi-line business access rate will be applied only prospectively, to newly laid Centrex-CO lines. Existing Centrex-CO plant will be subject to transitional access charges, not to exceed $2.00 in 1984. Reconsideration Order ¶¶ 47-49, 48 Fed.Reg. at 42,991-92. These charges will be reexamined in supplemental proceedings and will presumably eventually be equivalent to the multi-line business access rate, but the Commission does not expect the access charge to exceed $3.00 in the 1985-1986 access period. Further Reconsideration Order ¶ 40 n. 24, 49 Fed.Reg. at 7815.
The D.C.P.S.C. claims that this solution did not adequately address the potential threat to universal service since there was no record support in favor of the Commission’s minimal predictions of stranded investment. In particular the D.C.P.S.C. argues that the rate structure prescribed within its jurisdiction provides subsidies to Centrex-CO from other classes of subscribers, not vice versa. However, the Commission did not base its conclusions on a finding tKat all local Centrex-CO exchange rates were subsidized, nor was it required to do so.
The Commission made other findings supported by record evidence which rationally buttressed its conclusion that there is no “threat to universal service so substantial and so imminent that we must depart from this approach____” Id. at ¶ 47, 49 Fed.Reg. at 7816.
First, the Commission recognized that as a practical matter most Centrex-CO users would be unable to terminate Centrex-CO service immediately. Several parties before the Commission indicated that it would take at least three years to move to alternative services. During that interim period more costs of existing Centrex-CO plant could be recovered.
Second, testimony before a state commission cited by the parties indicated that a large percentage of Centrex-CO plant could be reused in other local service. Id. at ¶ 46. This further diminished the likelihood that universal service would be greatly impaired by full flat-rate per line access charges on Centrex-CO subscribers.
Third, the Commission took affirmative steps to monitor the threat posed by increased Centrex-CO rates. In the Reconsideration Order it requested the Joint Board in Docket 80-286(1) to assist the Commission in monitoring “the nature and magnitude of any stranded investment problem,” and (2) to recommend solutions which could be adopted at the state or federal level to avoid a threat to universal service. Reconsideration Order ¶ 49, 48 Fed.Reg. at 42,992. These were reasonable responses to the claims of the D.C.P.S.C. In view of the Commission’s decision to refer this problem to the Joint Board, arguments of the D.C.P.S.C. that local rate payers will inevitably bear the full brunt of obsolescent Centrex-CO plant are premature.
The D.C.P.S.C. does not even offer a reasonable alternative to this decision. Permanent depression of Centrex-CO access charges to reflect only sixteen percent of the standard flat-rate end user charge in order to achieve parity with PBX users is inconsistent with the basic principles of the Commission’s orders, which generally require end users to bear the cost of the interstate NTS charges attributable to their use of the exchange. The price disparity between Centrex-CO and PBX service is directly related to the cost of providing the exchange access. Because Centrex-CO requires approximately six times the number of local loops as PBX, the interstate share of local NTS costs is correspondingly higher. Access charges that rationally reflect that increased level of costs are not discriminatory. Id. ¶ 45, 48 Fed.Reg. at 42,991. Pricing telecommunications services based solely on mainte*419nance of comparative competitive positions, regardless of the cost of providing the services, would fundamentally contradict the primary rationale underlying the access charge proceedings. The Commission reasonably determined to apply these principles consistently to all end users, including Centrex-CO subscribers; we may not arbitrarily overturn that decision in our appellate review.
As might be expected, the PBX industry, as the principal competitor of Centrex-CO service, supports the Commission’s decision to impose full flat-rate end user access charges on Centrex-CO service. However, represented by the North American Telecommunications Association (“NATA”), a trade association comprised of manufacturers, distributors, retailers, and installers of customer premises equipment, the PBX industry challenges the Commission’s authority to grant any transition to Centrex-CO users. NATA’s argument is two-pronged.
First, it asserts that the Commission’s response to the threatened loss of competitive Centrex-CO equipment was not supported by record evidence. NATA argues that instead of relying on representations made by the state commissions and other commenting parties the Commission should have amassed the relevant underlying facts, and then independently analyzed them to develop its own conclusion on the level of intrastate Centrex-CO costs, and the concomitant effect on local rates and universal service of immediate Centrex-CO rate hikes. Brief of Petitioner/Intervenor NATA at 58. Failing to complete these steps, NATA charges that the Commission has not presented “the minimum of evidence upon which an agency can rely to support agency action in an informal ... rulemaking.” Reply Brief of Petitioner/Intervenor NATA at 12.
Second, NATA charges that the lower transitional rates charged to Centrex-CO subscribers unreasonably discriminate against all other classes of business end users since the price discrepancies are not cost based. The proper remedial action, NATA urges, is to “direct the Commission” to eliminate the transitional preference for Centrex-CO users. Brief of Petitioner/Intervenor NATA at 45. However, neither of NATA’s arguments would justify such an order to the Commission.
An agency decision arrived at through informal rulemaking must have a rational basis in the record and be based on a consideration of the relevant factors under its statutory mandate. Almay, Inc. v. Califano, 569 F.2d 674, 681 (D.C.Cir.1977). Consequently, when an agency undertakes a thorough, primary, evaluation of all relevant facts, it is highly desirable that the agency: independently amass the raw data; verify the accuracy of that data; apply that data to consider several alternative courses of action; and reach a result confirmed by the comments and submissions of interested parties. But the Commission’s failure to take or complete some of these steps does not fundamentally prejudice its decision. The paradigmatic scope of agency expertise is often pragmatically circumscribed internally by limited agency resources and restricted externally by the need to respond to complex, growing regulatory problems within a reasonable period of time. Notice and comment procedures are partially designed to overcome this problem. They permit parties to bring relevant information quickly to the agency’s attention. A degree of agency reliance on these comments is not only permissible but often unavoidable. Thus, although an agency must consider and analyze the factual materials gathered during the informal rulemaking process, see Action for Children’s Television v. FCC, 564 F.2d 458, 471 (D.C.Cir.1977), we have never held that an agency must conduct this analysis without relying on the comments submitted during the rulemaking.
In a lengthy and complicated rule-making such as this one it could very well be impossible to conduct elaborate independent verification proceedings on each factual comment submitted to the agency and still conclude the proceeding within a reasonable period. The Commission implicitly recognized the difficulty of immediately *420conducting a complete assessment of the local tariffs for Centrex-CO- service when it chose “to delay the full impact of the new rate structure until state commissions could reevaluate the rate structures for Centrex-CO services ... and to allow time for the Joint Board to study the subject.” Further Reconsideration Order ¶ 50, 49 Fed.Reg. at 7817. The Commission’s decision to rely on the comments of the state commissions that some Centrex-CO services are priced above cost, pending a more thorough assessment of the problem, was a reasonable response.
The representations made by various parties, including the Bell Operating Companies and state commissions, contained specific allegations about the impact on local Centrex-CO rates that could cause tens or hundreds of millions of dollars of equipment to be idled. See Brief of Intervenors Bell Operating Companies at 69 nn. 81-82. State commissions and other eommenters also argued that Centrex-CO rates were set above cost to subsidize residential rates. The Commission was entitled to rely on these representations by parties who were uniquely in a position to know the level of current subsidization and the impact on local rates of nongraduated Centrex-CO access charges.
This is not a case in which party submissions were accepted uncritically by the Commission. On the contrary, the FCC rejected a number of petitions as “unsupported speculation” and criticized other more substantially grounded estimates of revenue loss. Further Reconsideration Order ¶ 46, 49 Fed.Reg. at 7816. However, the Commission was unwilling totally to discount the problems which could be posed by stranded Centrex-CO investment. It requested the Joint Board to monitor the situation. The Commission also understood several responsible petitioners to assert the existence of above-cost Centrex-CO pricing, supporting the conclusion that an interim period would be necessary to permit some state commissions to reevaluate their Centrex-CO tariff policies. The decision to accommodate that period of reevaluation, in conjunction with the establishment of a Joint Board, was the method chosen by the Commission to balance the risks and benefits alleged to follow its regulatory action. That balancing, based on the Commission’s expert evaluation of the comments placed before it, is reasonable and rationally supported by the record.
NATA’s argument that the Commission has unreasonably discriminated between Centrex-CO and PBX subscribers is also without merit. The Commission is quite properly concerned about facilitating the capacity of state regulators to respond meaningfully to its orders, and minimizing unnecessary disruptions in service. A transition period to full implementation of Centrex-CO access charges is rationally related to those goals and is in no way unlawfully discriminatory.
The Commission very carefully threaded its way through the opposing claims raised by local exchange companies and the PBX industry, evaluating the sufficiency of those comments, and responsively adjusting its access orders to achieve what it determined in its expert discretion to be the optimum balance among its statutory goals. The decision to impose flat-rate Centrex-CO access charges on a per line basis after a period of transition is supported by the record and reasonably related to those goals. We affirm the FCC’s conclusions on this issue.
4. Party Lines
In addition to prescribing special rules for end users of Centrex-CO systems, the Commission also considered the proper structure of party-line end user access rates. Unlike ordinary subscription service, in which one loop serves only one subscriber, party lines permit several subscribers to share a single loop, reducing the cost of the telephone service to each subscriber. The service is often used in rural areas where the average cost of laying each loop may be much higher than in urban exchanges.
In its Access Order the Commission directed that the access charge for individual *421party-line subscribers be computed by dividing the standard access fee for a single-subscriber loop by the number of party-line subscribers sharing the party line. Access Order, App. A., 47 C.F.R. § 69.104(c), 93 F.C.C.2d at 349. On reconsideration, several parties claimed the rule would be too burdensome to administer on a line-by-line basis. In response the Commission adhered to its decision to compute the charge by dividing the single-line rate by the number of party-line subscribers, but permitted the number of subscribers to be calculated based on the average level of subscription, or “fill,” in each class of party-line service, such as two-party, four-party, or six-party lines, rather than individually tallying the number of subscribers to each party line. See Reconsideration Order, App. A., 47 C.F.R. § 69.104(c), 48 Fed.Reg. at 43,018.
The Rural Telephone Coalition (“RTC”) charges that the Commission acted unreasonably in adopting this system. It claims that this system inaccurately reflects the cost of providing party-line service, since party lines do not cost less in direct proportion to the number of subscribers sharing the line. Record evidence relied upon by RTC suggests that a multiple line system serving several subscribers with separate loops may cost only 25% more per subscriber than a party-line system, although the Commission’s rules would impose an access charge on a single-line subscriber up to 250% greater than that paid by the party-line subscriber. Petition for Reconsideration and Clarification of the RTC, J.A. 2926, 2941-42, 2958. Because the proposed system thus “misstate[s] the relationship between the cost of providing party line and single party service,” Brief of Intervenor RTC at 19, single-line subscribers effectively subsidize much of the cost of providing the inferior grade party-line service. The RTC argues that this distortion between cost of party-line access and the assigned flat-rate charge would create an artificial economic incentive for customers to switch from single-line to party-line service. Besides artificially encouraging the widespread use of a lesser grade of service, the RTC asserts that converting existing single-party plant to party-line plant would cost more than building single-line plant initially.
The Commission failed to respond adequately to these charges. Its Reconsideration Order recognized that the RTC objected to the “uneconomic incentives for choosing party-line over single-line service.” Reconsideration Order ¶ 38, 48 Fed.Reg. at 42,990. The Commission also agreed with the RTC that end-user party-line charges are intended to reflect the actual cost of providing the service. However, the Commission never responded to the RTC’s comments and evidence suggesting that the party-line plan established by the Commission contradicts these very goals.
The Commission made two modifications to its party-line plan on reconsideration. First, it required computation of the fill ratio for each class of party-line service, rather than for each line. However, far from “largely ameliorat[ing] the problems perceived by” the RTC, id. ¶ 39, this amendment addresses only the method of administering the plan, not the cost-price relationship between single and party-line service. The Commission’s second change, waiver of its party-line allocation rules in those rare cases when party-line service may cost more to provide than single-line service, is similarly nonresponsive to the core problem identified by the RTC: as a general rule the cost of party-line access per subscriber will not be recovered by that subscriber’s access charges, even when that party-line costs less than a single-subscriber line.
The only record evidence on the issue of party-line costs was that submitted by the RTC which suggested that interstate party-line access charges would not fully reflect costs. The Commission never referred to this evidence, never explained why it was flawed or unreliable, and never offered any alternative explanation for its plan. It is thus entirely likely that the Commission unwittingly adopted a party-line plan totally inconsistent with the driving principle *422behind its access charge decisions, which is to align the level of access charges, as far as possible, with the actual cost of access.
Certainly the Commission would not be required to maintain full cost-based access pricing for end-user party-line charges, but we should expect to find either a clear explanation for its departure from the general principle followed elsewhere in the rulemaking, or a responsive rebuttal of charges that its party-line access charges do not accurately reflect costs. The Commission offered neither. It concluded its discussion in apparent agreement with the RTC that party-line charges should permit rational economic choices, again ignoring the RTC’s claim that the Commission’s plan would reach the opposite result.31 This unexplained and unsupported conclusion is the antithesis of rational rulemaking.
We must therefore remand this portion of the Commission’s orders for meaningful consideration of claims that its party-line access charges are uneconomically subsidized by single-line access rates. Further analysis by the Commission, accompanied by a well-articulated explanation of its response to these claims will, we are confident, clear up the confusion that now exists about the extent to which end-user party-line access charges are intended accurately to reflect the cost of providing the service and the extent to which party-line access fees actually achieve their intended result.
5. Average Schedule Company Status
The last major part of the Commission’s rules dealing with end-user access charges challenged in this proceeding concerns not who will bear the charges, but the method by which smaller exchange companies will identify the level of their costs incurred in providing access. Before an exchange carrier can recover its access costs, it must first determine what they are. Precise determination of á local company’s costs in all relevant areas may require extensive data collection, analysis, reporting, and auditing, which can be a difficult and costly burden for small telephone companies. As a result, the Commission’s rules have traditionally allowed smaller exchange carriers to estimate some or all of their costs through use of an “average schedule” which adopts generalized industry data to reflect the costs of a hypothetical exchange company. According to the RTC, many of the small exchange carriers which elect to use the average schedule, known as “average schedule companies,” use the average schedule to compute every element of their costs except the expense'of connecting local central offices to toll offices, or “line haul” costs. Because line-haul cost studies are relatively less expensive and less complicated than procedures necessary to identify other types of costs, a large number of smaller exchange companies prefer to conduct their own rather than rely on the average schedule.
The FCC, in the access orders, changed current practice relating to average schedule company status in two ways. First, the Commission ruled that a company may no longer elect to use the average schedule if it is affiliated with, or owned by, a carrier computing its costs directly, often referred to as a “cost company.” See Reconsideration Order, App. A., 47 C.F.R. § 69.-605(c)(1), 48 Fed.Reg. at 43,022. Second, the Commission required that companies electing average schedule status participate in all of the general exchange carrier association tariffs, effectively precluding a carrier from using independently generated cost data to demonstrate line-haul costs while relying on the average schedule to establish all other costs. Id. ¶ 193, 48 Fed. Reg. at 43,014; Access Order, App. A., 47 *423C.F.R. § 69.605(c)(2), 93 F.C.C.2d at 362. The RTC challenges both of these rules.
In reviewing the reasoning used by the Commission to support its decision to preclude partial reliance on average schedule status, it is again apparent that the Commission agreed with the fundamental objectives advanced by a commenting party, but promulgated a rule which could cause drastically different consequences, without record support for the discrepancy between means and end. The Commission clearly expressed its intent “to avoid imposing the burden of developing cost information upon companies” which may be too small to perform the necessary cost studies, Reconsideration Order ¶ 194, 48 Fed. Reg. at 43,014, a goal also advocated by the RTC. The Commission’s decision is a substantial change from prior practice; the Commission recognized that many independent companies will be forced to abandon average schedule status. Id. ¶ 193. However, there is nothing in the record which suggests that parent, subsidiary, or affiliate status accurately distinguishes between those companies which would be prohibitively burdened by the cost, and those which could easily bear it.
In defense of its rule the Commission reasoned:
Some companies that are large enough to compile cost information undoubtedly also participate in average schedule settlements. We could not reasonably defer the implementation of access charges to identify such companies, but we did infer that companies or affiliated groups of companies that are partly in and partly out of average schedule settlements are not too small to perform cost studies.
Id. ¶ 194. Yet, the mere inference that affiliation alone indicates ability to bear the cost burdens of the affiliate is not always reasonable since, as the RTC argues, a small cost company affiliated with a second, small, average schedule company would not necessarily have sufficient financial resources to bear the full expense of dual, full-fledged cost studies. Reply Brief of Intervenor RTC at 7-8. More importantly, the Commission imposed this “compulsory pooling requirement,” Reconsideration Order ¶ 193, 48 Fed.Reg. at 43,014, between affiliates without inquiring into the regulatory or corporate barriers which may prohibit such cross-pooling; in particular, the RTC argues that state commissions may not allow revenues of locally regulated exchange companies to be diverted to fund the operations of affiliated companies in other jurisdictions. Reply Brief of Intervenor RTC at 8.
The Commission dismissed similar arguments below without any explanation, asserting that the average schedule companies have “not presented any reason for concluding that they should be entitled to” the benefits .accruing from their current average schedule status. Reconsideration Order ¶ 195, 48 Fed.Reg. at 43,014. This minimal consideration is inadequate, especially since the Commission came close to accepting the weight of these arguments when it decided to delay the start-up date of the new rule for two years.32 However, a transition period alone does not mitigate the difficulties identified by the RTC, since it fails to address the reasonableness of the Commission’s. inference that cost companies are generally both financially prepared and administratively authorized to bear the cost burdens of affiliated average schedule companies. .We are therefore remanding this aspect of the Access orders to the Commission for further study and consideration on these issues.
Given the Commission's willingness to defer the effective date of this rule for two years, our remand to the Commission for a brief period to enable the Commission to conduct a more responsive and meaningful evaluation of small carriers’ claims does not run counter to the Commission’s stated desire to avoid unreasonably deferring the implementation of access charges pending *424the identification of those carriers who are currently able to compile individualized cost information. It will ensure that the Commission, otherwise spurred by the necessity to design and establish a comprehensive access charge plan at the earliest reasonable date, does not give short shrift to the complaints of small exchange companies that the benefits historically derived from the existing average schedule company status rules would be lost by the unconsidered decision to abandon the prior rules in favor of a less reasonable system not immediately necessary.
For similar reasons we must also remand this proceeding to the Commission for a reasoned evaluation of the parties’ challenge to the Commission’s refusal to permit use of average schedules for less than all access costs. The Commission announced this new rule in its Access Order without any prior public notice or opportunity for participation by the affected parties. Its reevaluation of the rule in light of the substantial comments subsequently submitted in petitions for reconsideration demonstrates a failure to consider seriously the legitimate objections of small exchange companies.
The RTC has identified at least one significant reason which could justify softening of the new rule: forcing exclusive reliance on the average schedule could gravely affect companies with exceptionally high line haul costs. Brief of Intervenor RTC at 17. Computation of these costs, unlike other categories of access costs, can be done through relatively simple accounting procedures. Reply Brief of Intervenor RTC at 10. In its Reconsideration Order the Commission merely added conclusory language without addressing these contentions. The FCC’s explanation for the rule change reads in full:
Some petitioners also note that the existing average schedule system gives a company the option of participating in an average schedule for only a portion of its costs and suggest that the access charge rules be modified to include such an option. The average schedules do not appear to correspond with access elements we have defined and accordingly could not be easily adapted even if we found that such a system would be desirable.
Reconsideration Order ¶ 196, 48 Fed.Reg. at 43,014.
However, it may be possible that the newly reorganized categories, or “elements,” see id. ¶ 4, 48 Fed.Reg. at 42,985, of access costs can be harmonized with the established industry practice in this area, or that the average schedules can themselves be adjusted to reflect the new access cost elements identified by the Commission. Reply Brief of Intervenor RTC at 9-10. The Commission never adequately considered these possibilities.
By ignoring these alternatives and the potential advantages to continuing the status quo, the Commission failed to discharge its special responsibility to balance the policies embodied in the Communications Act, selecting the regulatory course of action most likely to meet the public interest. Accordingly, the Commission’s orders prohibiting election of partial average schedule company status are remanded to the Commission for further consideration.33
B. Access Charges to Carriers and Private Line Users
Not all of the interstate share of local exchange costs will be recovered through flat-rate end user access fees. Traffic sensitive exchange costs will be recovered from interexchange carriers on a usage sensitive basis. See Access Order ¶¶ 197-249, 93 F.C.C.2d at 297-315. The Commission established an additional component of the carriers’ access charges which is designed to recoup costs assigned to the “carrier common line element,” including (1) the *425NTS costs associated with inside wiring and customer premises equipment, (2) the gradually declining balance of local NTS exchange costs not recovered from end users pending the transition to more fully cost-based end user access charges, and (3) the cost of the Universal Service Fund established to help subsidize certain NTS costs on a permanent basis. Only the Universal Service Fund is expected to be a nontransitional component of the carrier common line element. The inside wiring and customer premises equipment costs will be gradually removed from the rate base, and the balance of NTS exchange costs not recovered from end users or through the Universal Service Fund will probably drop to zero. Brief for FCC at 30 n. 46, 31 n. 47.
The costs allocated by the Commission to the carrier common line element will be recovered on a usage sensitive basis from most interexchange carriers in the form of a carrier common line charge. See id. at 30. A private line surcharge intended to approximate the carrier usage charges will be levied on private line users and a few other classes of specialized carriers. See Reconsideration Order ¶ 81, 48 Fed.Reg. at 42,997; id. App. A., 47 C.F.R. § 69.-115(b), 48 Fed.Reg. at 43,019. In this section of the opinion we discuss claims raised by the affected carriers and private line users that the Commission acted unlawfully when it assigned these access charges to the various categories of exchange users.
1. Unlawful Discrimination Against Carriers Subject to the Carrier Common Line Charge in Favor of Other Interstate Users of Exchange Facilities
The advent of competition in the provision of interexchange telephone services, and the rapidly proliferating varieties of services offered through the medium of telephone transmissions, demonstrated to the Commission the need for a coherent, uniform method of compensating exchange carriers for the interstate costs of providing exchange access. The correction of disparities in exchange access rates charged to the various classes of interexchange carriers was a primary goal in Docket No. 78-72. See MCI Telecommunications Corp. v. FCC, 712 F.2d 517, 529, 531-32 (D.C.Cir.1983). The Commission asserts that it accomplished this result through the imposition of the usage-based carrier common line charge, Access Order ¶ 51 n. 20, 93 F.C.C.2d at 258, and the private line surcharge. MCI disputes this claim. Before evaluating the merits of MCI’s arguments we find it helpful to review briefly the effect of the Commission’s orders on the previous system of exchange access compensation.
a. The Previous System of Exchange Access Compensation and the Commission’s Response
In addition to AT & T, the predominant interexchange common carrier, several types of communications carriers and telecommunications consumers require access to the local exchange.
One category, composed primarily of commercial, specialized common carriers, offers to transport subscribers’ telephone traffic between the originating and terminating local exchanges. Because this general type of service exists to provide interexchange carriage, virtually all interstate calls carried are accessed into the originating and terminating exchanges. Carriers offering this service include the OCCs, which use privately owned facilities to carry interexchange traffic for a basic fee; resellers, which lease OCC or AT & T interexchange facilities in order to offer high volume discounts to subscribing groups of moderately heavy interstate callers; and sharers, who join together in a nonprofit arrangement to use jointly the services and facilities of a single underlying carrier, paying a pro rata portion of costs based on relative use. FX service is another type of interexchange offering in which subscribers terminate all of their interexchange calls by switching them into the exchange, in this case by integrating *426the open end of the system into the distant exchange.
A second grouping of interexchange telecommunications users generally makes much more limited use of exchange access. These are the access users which the Commission identified as primarily responsible for the “leaky PBX” problem. They lease or own dedicated private lines used to provide point-to-point service, but terminate those lines at PBX facilities or other private switches that can patch an interstate private line call into the local exchange. Although the caller is thus able to originate and terminate jurisdictionally interstate calls, incurring interstate exchange costs, those costs are not currently recovered from the private line subscriber since local exchange companies generally are not equipped to distinguish between the local and jurisdictionally interstate traffic routed to or from the PBX. In addition to private lines, some not-for-hire privately owned telecommunications systems, used by large corporations to provide voice and specialized data transmissions, are also capable of accessing the local exchange.
These private line arrangements are technically capable of switching the same volume of traffic into the local exchange as the exchange carriers in the first grouping. However, less intensive use of exchange access can probably be inferred, if only because the first grouping of services exists primarily, and is commercially offered, for the purpose of originating and terminating interexchange traffic on the local exchange, while the second grouping of services has traditionally been offered for the primary purpose of completing intrasubscriber telecommunications.
Of course, the lines between these two groupings of exchange access users are often blurred. The Commission recognized that “[djepending upon the nature of its operation, a given private line ... user may or may not make significant use of local exchange service for interstate access.” Reconsideration Order ¶ 78, 48 Fed.Reg. at 42,996. One class of carriers which does not fit conveniently into either of these groupings is that of the enhanced service providers. These carriers use interexchange facilities to transport subscribers’ data transmissions between computer or data terminals in different exchanges, and may rely heavily on access to the local exchange to originate or terminate their transmissions.
Regardless of the amount of use, each of these classes of carriers and private line users relies on technologically similar access facilities to originate or terminate calls in the local exchange. Prior to the Commission’s orders under review there was no uniform scheme through which these exchange users compensated the local exchange for the interstate share of NTS costs associated with their access. Through the settlements and division of revenues process the Commission has traditionally authorized a fairly high charge for exchange access on AT & T, which AT & T recovered through usage-based charges for ordinary long distance phone service. OCCs were subject to ENFIA rates, which included a negotiated discount from the access charge rate paid by AT & T. FX users paid only local exchange rates for their open end access, but have not borne any of the cost burden attributable to their interstate use of the exchange. Users of private lines capable of switching into the local exchange similarly paid only local rates. Thus, although the Commission determined that these private systems “originate and terminate vast quantities of interstate and intrastate toll traffic through the use of exchange telephone service,” id. H 80, under the access scheme in effect prior to its orders, “no charge whatsoever [was generally] assessed for interstate use of local services.” Id. ¶ 81.
In surveying that system, the Commission found,
[N]one of the participants has attempted to demonstrate that there is any reasonable or rational relationship to justify the wide disparities among the charges for access that are directly or indirectly levied upon users of the various interstate services that might satisfy the require*427ments of Section 202(a). It is readily apparent that it would be impossible to do so. Indeed, the current methods of recovering costs of jointly used non-traffic sensitive subscriber plant ... are totally different and produce widely differing results even though each service uses the same plant in the same manner .... Since no one has attempted to justify the disparate rates charged for like access services in this proceeding, we must find them to be unlawfully discriminatory.
Access Order ¶ 51, 93 F.C.C.2d at 258.
To remedy this discrimination the Commission ruled that users of exchange access generally must pay either a carrier common line charge or a private line surcharge to contribute towards the access costs not recovered from end users. The carrier common line charge applies to those carriers whose previous access rates were governed under the ENFIA tariffs. Carriers in this category include the OCCs, and many resellers. Reconsideration Order ¶ 83, 48 Fed.Reg. at 42,997; id. App. A, 47 C.F.R. § 69.105(a), 48 Fed.Reg. at 43,018. In addition, the Commission determined that FX users would also be subject to the carrier common line charge. The private line surcharge applies to most other users of interexchange telecommunications facilities which are capable of using local exchanges to originate and terminate these calls. Carrier groups subject to this “special access” surcharge are most private line users, sharers, and enhanced service providers. Reconsideration Order ¶ 83, 48 Fed.Reg. at 42,997; id. App. A., 47 C.F.R. § 69.115, 48 Fed.Reg. at 43,018, as amended by Further Reconsideration Order, App.A., 49 Fed.Reg. at 7829. One narrow class of exchange users, private communications systems, were not placed in either category, but are subject to reasonable non-discriminatory tariffs for exchange access which may be developed by local exchange companies in the future. See Further Reconsideration Order ¶¶ 130-33, 49 Fed.Reg. at 7826.
Both the carrier common line charge and the private line surcharge are based on relative use, but different methods are used to determine the amount which the carrier uses the local exchange to originate or terminate interstate calls. The carrier common line charge is tied to measured usage. OCCs will pay a flat, per line rate based on the projected relative use attributable to the average OCC line. Reconsideration Order, App. A., 47 C.F.R. § 69.105, 48 Fed.Reg. at 43,018, as amended by Further Reconsideration Order, App.A., 47 C.F.R. § 69.105(b), 49 Fed.Reg. at 7829. Resellers will be similarly charged. FX users will pay a nonaveraged per minute/per line charge according to the actual use of each FX line. See Further Reconsideration Order ¶ 98, 49 Fed.Reg. at 7922. In a subsequent proceeding the Commission exempted FX lines whose use of the exchange cannot currently be measured. Pending development of measurement capability, these FX subscribers will pay only the local exchange rate or an alternative usage surrogate proposed by exchange carriers. See Investigation of Access and Divest. Related Tariffs, 49 Fed.Reg. 9174, 9185-86 (1984); Brief for FCC at 82 n. 108.
Those subject to the private line surcharge pay a flat monthly fee of twenty-five dollars based on the estimated average per line use of exchange facilities. This surcharge is based on estimated rather than measured use primarily because local exchanges are not physically equipped to monitor the interstate usage of these special access subscribers. See Reconsideration Order, App.A., 47 C.F.R. § 69.115, 48 Fed.Reg. at 43,019, as amended by Further Reconsideration Order, App. A., 49 Fed.Reg. at 7829.
b. Failure to Cure Pre-existing Discrimination
Petitioner MCI claims that the Commission merely perpetuated the unlawful discrimination which the Commission found to exist in its Access Order, without curing the disparate charges for functionally simi*428lar access. MCI estimates that it will pay approximately $355 for each access line, while private line users will pay only the $25 surcharge. Because the two classes of telecommunications users utilize “the same plant in the same'manner” but continue to pay widely divergent per line charges, MCI argues that the Commission merely continued the “widely different” rate results which originally caused the Commission to reach a finding of unlawful discrimination. See Access Order ¶ 51, 93 F.C.C.2d at 258. MCI seizes on the Commission’s characterization of the plan as “designed to achieve a rough equity among access service users,” Reconsideration Order ¶ 77, 48 Fed.Reg. at 42,995, as proof that the Commission has breached its admitted obligation to develop a permanent solution to the “rough justice,” MCI Telecommunications Corp. v. FCC, 712 F.2d 517, 524 (D.C.Cir. 1983), in access rates established in the ENFIA agreement. Brief for Petitioner MCI at 45-46. MCI demands that the unlawful discrimination be promptly remedied, relying on authority which prohibits the establishment or continuation of charges which have been held to be discriminatory or unlawful. MCI Telecommunications Corp. v. FCC, 712 F.2d 517, 535 (D.C.Cir.1983); National Ass’n of Motor Bus Owners v. FCC, 460 F.2d 561, 568 (2d Cir.1972); American Trucking Ass’ns, Inc. v. FCC, 377 F.2d 121, 130 (D.C.Cir. 1966), cert. denied, 386 U.S. 943, 87 S.Ct. 973, 17 L.Ed.2d 874 (1967).
MCI’s general attack on the Commission’s allocation of access charges between carriers and private line users is ill-founded. The Communications Act prohibits unjustifiably different rates for the same service. 47 U.S.C. § 202(a) (1976). But when there is a neutral, rational basis underlying apparently disparate charges, the rates need not be unlawful. For instance, when charges are grounded in relative use, a single rate can produce a wide variety of charges for a single service, depending on the amount of the service used. Yet there is no discrimination among customers, since each pays equally according to the volume of service used.
This is the neutral, reasonable principle underlying the disparities in per line access rates charged to OCCs and private line users. Although OCC’s will pay many more dollars per line for access charges, they require and use a correspondingly higher volume of exchange access on each line. Between OCCs and leaky PBX users the size of the per line access charges has a definite relation to the volume of access minutes attributable to each line. There is essentially no difference in the per minute rate charged to either class of access users.
Slight variations in rates identified by MCI are generally attributable (1) to the Commission’s decision to compute the individual line charges based on the classwide volume of access, or (2) to the difficulty of measuring the precise number of access minutes used by the PBX-private line subscribers. These choices are rationally necessary to minimize difficulties in administering the access charges and are within the range of the Commission’s discretion. See infra B.1.c., B.2. The resulting rates do hot rise to the level of “unjust or unreasonable” discrimination. 47 U.S.C. § 202(a) (1976).
The Commission held that the previous potpourri of access compensation arrangements was unlawful because there was no “reasonable or rational relationship to justify the wide disparities in charges.” Access Order ¶ 51, 93 F.C.C.2d at 258. The different rates could not be explained on the basis of functional differences in the type of access service offered, variations in the cost of providing that service, or the dissimilar amounts of access usage. Id. Correction of the illegality could be accomplished through the imposition of a single, neutral principle, such as relative use, without necessarily equalizing the amount paid by each access user for exchange access.
[23] The Commission’s decision to recover NTS access costs from carriers on a usage sensitive basis is not inherently discriminatory. Cf. MCI Telecommunications Corp. v. FCC, 675 F.2d 408, 414-16 *429(D.C.Cir.1982). Usage-based rates have long been imposed in the telecommunications industry. See, e.g., In re: AT & T, 9 F.C.C.2d 30, 93 (1967). Under the prior system most NTS costs were recovered through usage-based toll charges levied by AT & T for long distance telephone traffic, but different classes of carriers and private line users paid unequal rates. Now that the rates for each minute of exchange access have been equalized on an average basis there is no justification for a holding that the continuation of usage-based rates perpetuates the previous discrimination found to exist.
Although the Commission’s decision can be explained based on relative use, MCI claims that the circumstances of this proceeding make the Commission’s choice unreasonable, and therefore unlawful. MCI and other petitioners also contend that the plan unlawfully discriminates among the various classes of carriers and private users on factors extraneous to usage. We consider each of these objections in turn.
c. The Reasonableness of Adopting Usage-Based Charges for Carriers and Private Line Users Requiring Exchange Access
MCI contends that the Commission’s decision to recover NTS costs through usage sensitive carrier charges is unlawful on several grounds. First, the Commission itself determined that NTS costs, which by definition do not vary with usage, should in principle be recovered on a flat-rate basis from end users; its departure from that principle without adequate explanation is assertedly arbitrary and capricious. MCI also argues that a usage-based plan could not properly be-chosen to avoid extreme rate increases which could have caused disruptive service impacts. Finally, MCI contends that the plan would be unreasonably difficult to administer and should therefore not have been adopted.
These claims are not frivolous, Telocator Network of Am. v. FCC, 691 F.2d 525, 537 (D.C.Cir.1982), but inevitably misconstrue the breadth of the Commission’s statutory discretion to balance the multiple goals embodied in the Communications Act. They may suggest that the Commission could have reasonably elected to implement a nonusage-based scheme for recovering exchange access costs from carriers and other private line users, but they hardly prove error in choosing a usage-based recovery plan.
Recovery of the balance of NTS costs through the usage-based carrier common line charge and special access surcharge is not inconsistent with the decision to impose flat-rate end user charges. The Commission reasonably determined that NTS exchange costs are caused by subscribers, and should economically be recovered from the customers incurring those costs. Reconsideration Order ¶ 10, 48 Fed.Reg. at 42,987. The portion of costs which temporarily or permanently will not be borne by the end users, which forms the basis for carriers’ access charges, is essentially a subsidy. This subsidy is not logically attributable to a particular class of carriers. The Commission’s decision to recover NTS costs from end users on a flat-rate basis therefore does not require it also to assess the subsidized balance of those costs on a flat-rate basis from the interexchange carriers.
Moreover, the Commission carefully considered allegations that its usage-based recovery scheme was inconsistent with the general decision to promote economic efficiency by imposing flat rates to recover NTS costs:
Several petitioners seek ... complete elimination of the [carrier usage charge]. SBS argues that the charge is inefficient since it is not cost-based ....
[These petitioners] are legitimately concerned about the possible uneconomic effects of the Carrier Common Line charge. Such charges do reduce the relationship between rates and costs causation. Nevertheless, as the Access Charge Order stresses, for the most part the Carrier Common Line charge is a transitional charge. We explicitly recognized the economic costs of such a *430charge but viewed these costs as acceptable consequences of a gradual and certain transition. None of the petitions were able to suggest any alternative mechanism to produce such a transition. We, therefore, reject any suggestion that the carrier usage or Carrier Common Line charge be abandoned.
Id. ¶¶ 65-66, 48 Fed.Reg. at 42,994 (footnotes omitted). Other aspects of the Commission’s orders make it reasonably clear that a usage -based transition plan would help avoid disruptions in private line usage and other burgeoning interexchange telecommunications services which would have contradicted the very goals of a measured transition. Id. ¶¶ 77, 83, 90, 48 Fed.Reg. at 42,995-96, 43,000. Harsh results would have attended adoption of a flat-rate carriers’ charge such as that now proposed by MCI. That risk was evident even under the usage-based recovery plan adopted by the Commission, see id.; it would likely have been magnified under a flat-rate approach.
The choice of a usage-based scheme for the largely transitional carrier common line charge and private line surcharge was properly influenced by the need to avoid the disruptions in service which could have accompanied a flat-rate carriers’ charge. Our conclusion today that the Commission may lawfully impose flat-rate end user access charges on a gradual basis in order to preserve universal service is premised on the holding that rates may be structured to avoid disruptive service impacts. When necessary to avoid excessively burdening carriers, the gradual implementation of new rates and policies is a standard tool of the Commission. E.g., Uniform System of Accounts, 85 F.C.C.2d 818, 828 (1981); ENFIA, 71 F.C.C.2d 440, 455 (1979). See also National Association of Independent Tel. Producers & Distributors v. FCC, 502 F.2d 249, 253-55 (D.C. Cir.1974). The Communications Act authorizes the Commission to impose reasonable charges to promote a rapid, efficient, and modern telecommunications network in which technological innovations are encouraged in order to permit the development of facilities adequate to provide this service. See 47 U.S.C. § 151 (1976). MCI has itself benefitted from transitional rate structures implemented to avoid potentially fatal rate increases both under the ENFIA agreement 34 and under the Commission’s access charge scheme, in which OCCs pay a lesser rate for their access connections than does AT & T although the equivalent costs of providing that OCC and AT & T access may not be fully reflected in the rate differential.35 The shift from one type of nondiscriminatory rate structure to another may certainly be accomplished gradually to permit the affected carriers, subscribers and state regulators to adjust to the new pricing system, thus preserving the effi*431cient operation of the interstate telephone network during the interim.
Other claims have been made that a usage-based access charge plan would be unreasonably difficult to administer. Those arguments generally revolve around the difficulty of accurately measuring the relative use, or the problems associated with distinguishing between classes of carriers, such as resellers and sharers. However, it is apparent from our review of the record that the Commission thoroughly reviewed these problems and made a reasonable determination that other alternatives posed even graver risks to its articulated goals. See National Indus. Sand Ass’n v. Marshall, 601 F.2d 689, 699-700 (3d Cir.1979).
Moreover, the Commission made every attempt to mitigate these perceived problems. It made the most precise measurements of average relative use possible. It expects that carriers will devise even more accurate methods of determining PBX leakage, and will revise the measurements when that becomes possible. The Commission identified means of distinguishing between resellers and other classes of exchange users, and cautioned that the difficulty of drawing those lines should not become a pretext for discrimination. Reconsideration Order ¶ 84 n. 61, 48 Fed. Reg. at 42,997. It recognized that although “certain resellers cannot be identified, ... the surcharge which applies to private lines will apply to them as well, and will similarly serve as a temporary surrogate for interstate access charges.” Id. ¶ 84. No party proposed a coherent, feasible alternative scheme which would have better served the diverse and delicately balanced goals identified by the Commission.
We have certainly been unable to divine the existence of such an alternative plan; and we fail to see how a remand for further reconsideration could possibly improve on the Commission’s.thorough assessment of these issues. The FCC’s choice ■ of a usage-based system of carrier common line charges and private line surcharges is reasonable under the circumstances.
d. Discrimination in the Uniform Access Compensation Plan
Several parties allege that the Commission implemented its plan for carriers access charges in an illegally discriminatory manner. The primary claim appears to be that the Commission impermissibly discriminated against OCCs and resellers in favor of the closely related enhanced service providers and some sharers. Although the latter carriers may, at times, heavily use exchange access, they are subjected only to the lesser private line surcharge. The access charges paid by the sharers and enhanced service providers may thus not fully reflect their relative use of exchange access.
To analyze the lawfulness of that classification, we begin with the language of the Communications Act:
It shall be unlawful for any common carrier to make any unjust or unreasonable discrimination ... for ... like communications service, ... or to make or give any undue or unreasonable preference or advantage to any particular person, ... or to subject any particular person, ... to any undue or unreasonable prejudice or disadvantage.
47 U.S.C. § 202(a) (1976) (emphasis added).
The Communications Act thus does not prevent all discrimination — disparities in prices for similar service — but only unreasonable discrimination. Associated Press v. FCC, 452 F.2d 1290, 1300, 1301 (D.C.Cir.1971). The reasonableness of the price disparity must be judged by the circumstances in which it is assessed.
The Commission justified its decision to impose the private line surcharge on enhanced service providers and sharers as necessary to preserve their financial viability, and hence avoid adverse customer impacts. Reconsideration Order ¶ 83, 48 Fed.Reg. at 42,997. That conclusion represents its considered judgment that the benefits to be achieved through preservation of an efficient telecommunications network are more important than adhering to an inflexible access charge plan. It is difficult *432to quarrel with this judgment, since the Commission anticipated that the general level of access fees may fall as the usage-based system of charges is perfected. Another plan could unnecessarily pretermit the existence of these valuable telecommunications services.
Moreover, this plan merely extends to the enhanced service providers and some sharers the benefits of a graduated transition which was previously granted to those carriers subject to ENPIA tariffs. These carriers were not generally regulated under ENFIA and the Commission recognized that they had not yet had time to adjust their business operations to compensate for higher access charges. Id.
On balance, the Commission’s decision to avoid unnecessary “customer impact or market displacement” reasonably justifies any slight rate disparities'implemented under the new access charge plan. Id. Neither the Commission’s rulings36 nor our cases have ever held that all pricing disparities which may fail to recover full costs from the customer — however temporary or necessary to achieve the statutory policies of the Communications Act— are invariably banned by the antidiscrimination sections of the Act.37 The Commission thus warned in its initial order that it would be impossible to implement a perfect plan:
An ideal access charge plan would eliminate all discrimination or preferences within or among services, create incentives for the most efficient utilization of all telecommunications facilities, discourage all uneconomic bypass, ensure that no local exchange service subscriber cancels that service, and establish full and fair competition in the interexchange services market. All of those objectives could not be fully accomplished simultaneously and immediately even if we had perfect knowledge. Therefore, we necessarily must exercise judgment and discretion in devising an access charge plan that takes all of those objectives into account.
Neither the language of the Act nor past court or Commission opinions preclude this Commission from striking a reasonable balance. On the contrary, Congress undoubtedly anticipated that an exercise of judgment would be required when it déclared that it was creating this Commission in order to achieve multiple purposes “so far as possible.”
Access Order ¶ 88-89, 93 F.C.C.2d at 268.
In requiring the enhanced service providers and sharers to pay the private line surcharge rather than the carrier common line charge the Commission acted to end existing discrimination as far as possible. We uphold its choice under these circumstances.
Resellers also object to their treatment under the Commission’s plan. However, they have generally been paying EN-FIA rates and have thus had the benefit of a gradual transition. When offering resale of long-distance services they use the local exchange facilities in a similar manner and amount as OCCs. See Brief for FCC at 79 n. 103; AT & T—Applicability of the ENFIA Tariff to Certain OCC Services, 91 *433F.C.C.2d 568, 569, 575-77 (1982), aff'd mem., U.S. Telephone Communications, Inc. v. FCC, 725 F.2d 126 (D.C.Cir.1984). The Commission rationally held that carriers reselling private line services to provide long distance service are situated similarly to OCCs, and should thus be subject to the carrier common line charge.
MCI asserts that the Commission unlawfully discriminated by not requiring exchange carriers to file tariffs for privately owned telecommunications systems, although these systems are capable of accessing the local exchange. Brief for Petitioner MCI at 69. However, the Commission did permit carriers to file tariffs. Its decision not to make filing mandatory is reasonable since it is not yet clear whether exchange carriers have the measurement and other technical capabilities to develop a surrogate surcharge which could sufficiently approximate usage and satisfy the statutory limits on tariffs. See Further Reconsideration Order ¶ 133, 49 Fed.Reg. at 7826. Agency action which fails to require that exchange carriers perform the impossible is hardly unlawful.
We have thoroughly considered the parties’ remaining claims of discrimination, which, although numerous, are without any substance; these appear to be set out as hollow decoys to distract attention from the expert manner in which the Commission thoroughly balanced the complex, often contradictory, policies of the Communications Act. The Commission has, to the best of its ability, attempted to reconcile existing disparities in charges levied against interexchange carriers and private line users. Its plan to recover specified access costs through a combination of usage-based carrier common line charges and private line surcharges is appropriately tailored to minimize the difficulties inherent in any transition to a new system of rates; it is designed to guarantee the widest possible participation among all classes of exchange access users; and it is neutrally imposed on the basis of relative use. Certainly, therefore, this plan falls within the broad zone of expertise and discretion which must be granted to the Commission in a proceeding which touches the very core of the rapidly developing telecommunications industry.
2. Private Line Service
We next turn to consider the FCC’s action with regard to private line /PBX service (“PBX”). The Commission recognized in its Second Supplemental Notice that these communications services posed special problems for any new regulatory regime that sought to move the industry to a more efficient, cost-based footing.38 These problems were rooted in the fact that the large portion of PBX users possessed the capability to circumvent the conventional long-distance network and yet achieve interstate connections beyond those envisioned by the private line service. These connections — their frequency and duration — are not measured, nor will they be, at least in the short-term future. Thus PBX use posed a classic “free-rider” dilemma: PBX users enjoy the benefits of non-traffic sensitive equipment, but do not contribute to the cost of maintaining the interstate equipment.
The FCC has determined that some of the costs associated with such “leaky” PBX use must be recovered. To this end the Commission devised a surcharge:
[T]he most reasonable interim approach to reducing the discrimination in rates for MTS [long distance] users and ... other persons would be to develop a surcharge on private lines and the closed ends of WATS lines. We determined that a surcharge would be imposed on the closed ends of all interstate WATS lines as well as all jurisdictionally interstate private lines not falling within specifically enumerated exceptions. These exceptions included: (1) private lines subject to carrier’s carrier charges; (2) private lines that cannot leak; and (3) cer*434tain private lines that can but probably do not leak (i.e., Telex and programming facilities). We established a flat surcharge of $25 per voice grade line for 1984 only because insufficient time remained for the telephone companies to develop a system of surcharges more precisely reflecting actual leakage before the October 3 deadline for filing access service tariffs. See Reconsideration Order at para. 88. We expressed our expectation that telephone companies would act to replace the $25 surcharge by such a system of surcharges as soon as possible.
Further Reconsideration Order, 11112, 49 Fed.Reg. at 7824. The Commission explained that this surcharge concept was the best alternative open to it: “The imposition of a modest surcharge that is not based upon actual usage measurements will reduce discrimination or preferences to the maximum extent possible without imposing costly and difficult measurement procedures.” Id., ¶ 116, 49 Fed.Reg. at 7824. The choices open to it were few, the Commission explained, and it described the difficulties attendant to each possible course. Id., ¶¶ 117-119, 49 Fed.Reg. at 7825. It is a remarkably candid review of the options open to the Commission. The computation by which the $25 per line figure was arrived at is contained in paragraph 88 of the Reconsideration Order.39 It is by admission an estimate based upon assumptions drawn from the collective experience of the Commission. It appears to be a conservative estimate. It evidences the Commission’s judgment based upon the agency’s expertise and experience collected over a great many years. It also represents, we conclude, an example of exactly the kind of difficult judgment call for which expert agencies have been created.
As is to be expected with any decision that departs from the status quo, the beneficiaries of the status quo have objected to the Commission’s plan. Aeronautical Radio, Inc. (“ARINC”), a heavy subscriber to private line service, summarized its objections to the $25 charge this way:
The Commission ... erred in prescribing a special access surcharge to be imposed upon the terminations of private lines which bears no relationship either to the amount of usage alleged [to] be made of the local exchange by such lines or the cost thereby imposed upon the exchange. In attempting to develop a value for the surcharge, the FCC relied upon unsupported and facially unreliable estimates of interstate private line usage which, in any event, are wholly irrelevant to the amount of private line traffic which might “leak” into the local exchange through private switching arrangements. The Commission’s manifest failure to articulate a rational basis for its determination of the special access surcharge, and its admitted reliance upon unrelated and irrelevant data in determining the value of that charge, clearly requires remand to the agency.
Brief for ARINC at 30.
The FCC responds by noting that “[n]o data are presently available on the percent*435age of private lines that are connected to PBXs or other switching machines.... Because private lines are not routinely metered, the FCC also had no data on the extent of ‘leakage’ from such private lines or on the percentage of leakage that is jurisdictionally interstate____ Rather than abandon entirely the possibility of recovering a fair part of interstate local exchange costs from private line users, the FCC undertook to develop a reasonable surrogate for the carrier charges.” Brief for the FCC at 97-98. The Commission adds that “[n]o one seriously challenges the more general proposition that private lines generate extensive traffic for local exchanges; and none of these who challenge the 80 percent figure offers a different estimate.” Id. at 98 & n. 130. AT & T adds in support of the Commission action that “[o]nce the use of NTS plant by special access line users is recognized, the only remaining question is the amount the user should contribute.” Brief for AT & T at 49.
In our judgment both the concept of a surcharge and the rate are lawful exercise of the statutory discretion vested in the FCC. The objections thereto are founded on faulty premises regarding the role of this court. We are not a policy-making body. This court instead patrols the perimeters of an agency’s discretion. If an agency in the course of an informal rulemaking does not attempt either to close itself off from informed opinion or to extend its reach beyond the scope of permissible authority, then it is our duty to accept that judgment if it is rational and not unreasonable. The fact that an agency’s decision is a difficult one, or that the decision rests on a set of evidentiary facts less desirable or complete than one which would exist in some regulatory utopia does not alter our role. We remain here to insist upon a necessary minimum: “[When an agency] is obliged to make policy judgments where no factual certainties exist or where facts alone do not provide the answer, [it] should so state and go on to identify the considerations [it] found persuasive.” Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467, 476 (D.C.Cir.1974). We cannot here conclude that “the FCC failed to take a sufficiently careful look at the problem presented, and failed to engage in reasoned decision making with respect to the issue....” Aeronautical Radio, Inc. v. FCC, 642 F.2d 1221, 1231 (D.C.Cir.1980), cert. denied, 451 U.S. 920, 101 S.Ct. 1998, 68 L.Ed.2d 311 (1981). The agency has not given us “[p]ious hope and speculation [in the place] of evidence.” Natural Resources Defense Council, Inc. v. EPA, 655 F.2d 318, 346 (D.C.Cir.) (Robb, J., dissenting), cert. denied, 454 U.S. 1017, 102 S.Ct. 552, 70 L.Ed.2d 415 (1981). Rather, it appears from the record that the FCC has spent considerable time and resources in dealing with the PBX problem, and it has indicated that it will continue the search for a better resolution. “[T]he Commission must be expected to make use of the experience it has gained through years of dealing with the problem....” City of Chicago, Illinois v. Federal Power Commission, 458 F.2d 731, 747 (D.C.Cir.1971), cert. denied, 405 U.S. 1074, 92 S.Ct. 1495, 31 L.Ed.2d 808 (1972). “[Effective regulation requires that the Commission bring to bear the full range of its knowledge, garnered from whatever source, in making the interpretation on which it bases important policy decisions.” Id.
The critical choice confronting the agency was either to do nothing about the “hidden” access enjoyed by PBX users, or to do something. The fact that PBX users could and do use the local networks dictated that the FCC seek some form of cost contribution. Such was not only a rational decision, it is in fact the only decision that is rational given the broader context of the FCC’s plan for the nation’s communication system.
Given this choice, the next critical question was how to cope with a problem about which no reliable data was available. The FCC chose to rely upon its historical experience and expertise to employ a system of conservative estimates. As discussed supra, some parties have assailed this choice because of their belief that it demands of *436free riders only a minimal fare. ARINC and others criticize the decision for demanding too much. We view it as one of a number of choices, any one of which would have been within the agency’s broad discretion. When “the figure selected by the agency reflects its informed discretion, and is neither patently unreasonable nor ‘a dictate of unbridled whim,’ then the agency’s decision adequately satisfies the standard of review.” WJG Telephone Co., Inc., v. FCC, 675 F.2d 386, 389 (D.C.Cir.1982). We note also that this rate is not one for the ages. It is an interim charge, imposed because of necessity. “The $25 surcharge applies only in 1984, 47 C.F.R. § 69.115(c), and the FCC said that it would reassess the level of the surcharge after 1984 in a separate proceeding.” Brief for the FCC at 100, n. 135.
The great part of the attack on the surcharge centers on the method the FCC employed in arriving at the $25 figure. See Brief for ARINC at 43-45; Brief for API at 32-34. ARINC and others would prefer that this court’s focus be confined to the particulars of this calculation and not the broad contours of the dilemma facing the Commission. It did not furnish the Commission, and it does not furnish the court, with any alternate methodology, much less one that is a persuasive substitute. Each step in the FCC’s calculation is to some extent guesswork, but it is reasoned guesswork. It is plausible. It is tethered to the Commission’s general experience with long-distance use. We remind ARINC that “some legislative judgments cannot be anchored securely and solely in demonstrable fact.” Industrial Union Department v. Hodgson, supra, 499 F.2d at 476. Of course we would prefer a more precise equation. But “[sjound principle bids us accompany any further judicial review — of the specifics of these approaches and tracing methodologies—with diffidence and restraint.” National Association of Greeting Card Publishers v. United States Postal Service, 607 F.2d 392, 401 (D.C.Cir.1979), cert. denied, 444 U.S. 1025, 100 S.Ct. 688, 62 L.Ed.2d 659 (1980). “The thoroughness and persuasiveness of the explanation we can expect from the. agency will, of course, vary with the nature of the prediction undertaken.” Natural Resources Defense Council, Inc. v. EPA, supra, 655 F.2d at 329. Here the prediction is one concerning the amount of use made of PBX facilities to connect with local loops. We cannot say that the prediction is unreasonable or unsound. The Commission “must be free, within the limitations imposed by pertinent constitutional and statutory commands, to devise methods of regulation capable of equitably reconciling diverse and conflicting interests....” Permian Basin Area Rate Cases, 390 U.S. 747, 767, 88 S.Ct. 1344, 1360, 20 L.Ed.2d 312 (1968).
ARINC argues that the FCC has not met its obligation to justify its policy choice, that the agency has somehow failed to carry its burden of proof in establishing the $25 figure. This is not a novel tactic. Nor is our response original: “Without embarking upon an extended discussion of burdens of proof, we think this view is manifestly unsound.” Telocator Network of America v. FCC, 691 F.2d 525, 539, n. 114 (D.C.Cir.1982). It is not the Commission’s chore to convince us that what it has done is the best that could be done, but that what it has done is reasonable under difficult circumstances. Here the unique nature of an unmeasurable but real problem of hidden access assists the FCC in justifying what it has ordered.
We cannot divorce the difficulty of the regulatory dilemma from the reasonableness of its resolution. The manifest equity of demanding some contribution to the upkeep of the system from PBX users combined with the current impossibility of more precision renders our conclusion unavoidable. Repetition of the phrase “arbitrary and capricious” may have dulled it and left less than obvious its primary meaning. But when an agency makes rational choices from among alternatives all of which are to some extent infirm because of a lack of concrete data, and' has gone to great lengths to assemble the available facts, reveal its own doubts, refine its ap*437proach, and reach a temporary conclusion, it has not acted arbitrarily or capriciously. “The Commission, as the expert agency entrusted by Congress with the administration of the crucial, dynamic communications field, requires and deserves some latitude in carrying out its substantial responsibilities.” Action for Children’s Television v. FCC, 564 F.2d 458, 482 (D.C.Cir. 1977). We thus conclude that the $25 rate was validly imposed.
3. Private Communications Systems
Clearly connected to the discussion of private line/PBX service is that small portion of the Commission’s order concerning private, not-for-hire communications systems. Regarding these systems the Commission commented:
[W]e shall allow exchange carriers to develop reasonable, nondiscriminatory surcharges on interconnected use of exchange services by carriers’ publicly offered interstate services using radio and other facilities (e.g. DTS), and privately-owned microwave relay systems, satellite transmission systems, and other interstate private facilities that would otherwise not be subject to either the surcharge or carriers access charges (that is, that will not employ any end links obtained from the exchange carriers to which private line surcharges would apply). In such cases, we are prepared to consider the carriers’ proposals for a surcharge to the individual exchange telephone lines which can be connected to such systems. Such a surcharge would have to be filed in tariffs with this Commission.
Reconsideration Order, ¶ 86, 48 Fed.Reg. at 42,998. The Commission was obviously concerned with the possibility that private communications systems, such as satellite transmission systems, might be enjoying or might soon enjoy the kind of hidden access to the nation’s telephone network which the PBX users currently possess. It thus acted to open the floor to proposals, in the form of carrier-proposed tariffs, on how to assess the costs of such access. Predictably, the users or potential users of in-place or projected systems reacted as though possible future levies had already been laid and collected.
Intervenors American Petroleum Institute,' the Association of Data Communications Users, and the Utilities Telecommunications Council (hereinafter referred to collectively as “API”) argue that:
Those provisions [of the FCC’s order] which authorize telephone companies to file the surcharges on private communications systems in federal access tariffs were issued without notice and opportunity for comment. Not once during the five-year history of this rulemaking proceeding has the Commission provided notice that it intended to impose surcharges upon private communications systems that engage local exchange facilities.
Brief for API at 17-18.
The FCC responds by raising three arguments: 1) that API is barred from seeking court review of its arguments because it failed to seek agency reconsideration of the issue; 2) that parties in API’s position “should have been aware that the Commission might consider” access charges for those systems, and 3) “[i]n any event, the Commission took no action with respect to private systems that could be considered a ‘rule’ ... and thus the notice and comment requirements ... were not triggered.” Brief for FCC at 117. API has failed to raise any issue here that could be considered ripe for review.
The basic components of the ripeness doctrine were laid out by the Supreme Court nearly two decades ago in Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967), when the court stated:
Without undertaking to survey the intricacies of the ripeness doctrine it is fair to say that its 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 interference until an administrative decision has been *438formalized and its effects felt in a concrete way by the challenging parties. The problem is best seen in a twofold aspect, 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 148-149, 87 S.Ct. at 1515-1516. While ripeness questions frequently recur, this formulation of a court’s inquiry remains the guide to any particular controversy. See Baltimore Gas and Electric Company v. ICC, 672 F.2d 146 (D.C.Cir. 1982). Applied here it is easy to see that the issues pressed by API are not currently fit for review, nor does API yet bear any burden as a result of the Commission’s announcement.40
What the FCC has done is to serve notice that it considers the hidden access— if any — enjoyed by private communications systems to be a subject worth studying and possibly a problem requiring a remedy. It has invited the exchange carriers to conduct such studies and, if warranted, to propose a method of cost recoupment for any hidden access that is discovered to exist. The invitation is not, as we read it, an exclusive one. Subscribers to private systems would no doubt be well-advised to assist the exchange carriers in their assessment of the problem. Nor does the wording of the announcement preclude an extended and in-depth agency consideration of any proposals if and when they should materialize. But the agency has as yet decreed nothing, and parties such as API are as yet paying nothing. Likewise, we have nothing concrete to review.
We emphasize that the agency’s announcement and our comments on it in no way preclude or in any way limit the right of a user of a private system to later challenge the lawfulness of any charges developed by the exchange carriers. We remind the Commission that the agency cannot limit any future proceeding in this area by reference to its decisions in this docket. We also caution the Commission that it cannot, of course, cede to private parties such as the exchange carriers either the right to decide contests between themselves and their opponents or even the opportunity to narrow the margins of the debate regarding access charges for private systems. API seriously contends that the Commission is unlawfully delegating agency authority. See Brief for API at 42, n. 40, citing Sierra Club v. Sigler, 695 F.2d 957, 963, n. 3 (5th Cir.1983). Such argument is typically presented in the context of a transfer of legislative authority from the Congress to agencies, but the difficulties sparked by such allocations are even more prevalent in the context of agency delegations to private individuals.41
We need not examine the problem because we divine no such abdication of the Commission’s role as disinterested arbiter to any interested party. .Had the Commission so acted and had the Congress so intended it to act, that would amount to a “legislative delegation in its most obnoxious form; for it is not even delegation to an official or an official body, presumptively disinterested, but to private persons whose interests may be and often are ad*439verse to the interests of others in the same business.” Carter v. Carter Coal Co., 298 U.S. 238, 311, 56 S.Ct. 855, 873, 80 L.Ed. 1160 (1936). Since the FCC has here retained its final authority over these possible surcharges, which cannot go into effect unless and until the Commission approves them, and since the Commission has not prescribed any formula for their composition, it is premature to accuse the agency of an unlawful delegation. Nor will it be open to the charge if, upon presentment by any exchange carrier of its plan for imposing such charges on private system in the future, it takes due note and careful consideration of the views of all interested parties. For the present, however, there is no case that presents any issue that is ripe for review.
4. Foreign Exchange (‘FX”) Service .
Foreign exchange (“FX”) service is another component of the nation’s communication system.42 “Foreign exchange service is a combination of a local telephone access with a dedicated intercity private line which permits the public to call a local number to reach an ... office in a distant city.” Brief for ARINC at 6, n. 3. “For example, if an airline with a central reservation office in Kansas City acquires FX service between Kansas City and Denver, a potential Denver customer can dial a 7-digit local phone number in Denver — the ‘open end’ — to reach the Kansas City reservation bureau — the ‘closed end’.” Brief for the BOCs at 66, n. 75. Every call going out through an FX number is an interexchange long-distance call. Thus every FX call uses the local loop’s non-traffic sensitive equipment. This is undisputed. Under the old regime, however, FX users paid only the cost of a local business line and made no contribution for access to the “foreign exchange” similar to that assessed users of MTS/WATS or equivalent OCC-provided long distance services. FX users enjoyed a uniquely privileged spot in the rate structure.
The FCC determined that this special preference within the rate structure had to be eliminated. The Commission commented in its Access Order that
[T]he current methods of recovering costs of jointly used non-traffic sensitive subscribed plant for MTS, open-end FX, CCSA and WATS service and the ENFIA services are totally different and produce widely differing results even though each service uses the same plant in the same manner. The FX and CCSA services pay local exchange rates for open end access, the MTS/WATS equivalent services must pay the higher ENFIA rates, and MTS and WATS pay even higher access compensation through the settlements and divisions of revenue process.
Access Order, 93 F.C.C.2d at 258. As discussed, supra, the FCC, as one of its ultimate goals, projected a single-tier rate structure within which all providers of interstate service would pay the same carrier access charges. As a way station on the road to equal rates, however, the FCC established a two-tier system that recognized the reality of premium access enjoyed by some carriers, primarily AT & T. The OCCs thus will temporarily pay less in recognition of their inferior access. The FCC also chose to place the FX users into this second tier and, recognizing that while most FX users depend heavily upon the service, others use it only lightly, the Commission also determined that carriers would recoup the cost of access on a minutes-of-use basis rather than through a flat charge.
While ARINC initially applauded the FCC’s proposal to move the industry to a cost-based footing for access, See Comments of Aeronautical Radio, Inc. at J.A. 1387, its enthusiasm quickly waned when it became apparent that the new plan would require of FX users that which it would require of all other equivalent services— payment of a fair share of their attributable costs. See Opposition of ARINC at J.A. 2469. ARINC’s stake in the treatment of FX service is considerable:
*440The airlines have established a system of some 12,000 foreign exchange (“FX”) circuits to provide the public with toll-free access to airline information services. Use of FX lines enables the airlines to absorb the cost of intercity calls when members of the' public contact an airline office in a distant city for flight information, reservations, and other services.
Brief for ARINC at 6. The FCC’s plan presented the airlines with an “increase from a typical $30 per month charge for business service at FX open ends to, as a minimum, the $235-plus per month non-premium carrier charge, some eightfold increase.” Id. at 32 (emphasis in original). ARINC’s most recent estimate of the total cost increase the airlines will incur is approximately $30 million annually. Reply Brief for ARINC at 18, n. 20. We have no reason to doubt the accuracy of ARINC’s cost forecasts.
The FCC’s response to ARINC’s complaints appeared first in its Further Reconsideration Order. The Commission noted that “FX customers pay substantially less for each minute of use of the foreign exchange’s local network than would an MTS customer or the customer of an OCC using that carrier’s MTS-like service.” Id., H 99, n. 46, 49 Fed.Reg. at 7822. The FCC then responded to ARINC’s comments:
The ARINC petition apparently- requests that initial access charges for open end FX access be established at a level that is lower than the access charges for MTS/WATS equivalent services in order to provide FX users with a more gradual transition from the existing charges.... The rules that we adopted in the Access Charge Order were designed in part to remedy the longstanding discrimination between rates charged FX and MTS users, thus satisfying one of the central purposes of this proceeding. As early as the Second Supplemental Notice we presented a plan to establish parity between FX and MTS rates. 77 F.C.C.2d 224. The Access Charge Order achieved this parity in a way that differed somewhat from the plan described in the Second Supplemental Notice. That Order required that carrier’s carrier charges, including Carrier Common Line charges, be assessed in the foreign exchanges for each minute of use that FX service made of that local network. Petitioners have not demonstrated that FX charges that are lower than OCC charges would be warranted.
Id.
ARINC’s argument on appeal is relatively simple. It argues that throughout the FCC’s orders there runs the recurring theme that the Commission has been on guard against potentially devastating impacts of the new rate structure upon certain classes of communications consumers. This concern has been embodied in a variety of transitional devices designed to introduce consumers of specified services to the new access charges in stages. “In sharp contrast to its solicitude for the impact of its decision on other entities,” ARINC argues, “the agency totally ignored the impact on subscribers of the largest rate increase it prescribed ... ”, namely, FX users. Brief for ARINC at 32. “[T]he Commission failed to take into account the manifest threat to the viability of an industry already under severe financial strain—the airlines,” continued ARINC. Id. at 33. It concluded:
The Commission’s complete indifference to the plight of FX users facing massive rate increases, in contrast to its oft-repeated solicitude for all other entities facing even a fraction of such an impact, flies in the face of its obligation to deal evenhandedly with different parties who are similarly situated.
Id. at 35. ARINC refers the court to a well-known series of cases which undoubtedly do establish the proposition that an agency must “take pains to reconcile, an apparent difference in the treatment accorded litigants circumstanced alike.” Garrett v. FCC, 513 F.2d 1056, 1060 (D.C. Cir.1975). See also Melody Music, Inc. v. FCC, 345 F.2d 730, 732-33 (D.C.Cir.1965).
*441The FCC’s response to ARINC’s objections is twofold. First, the agency repeats what it has earlier stated in its orders: it must treat FX users as it does the users of MTS/WATS or similar OCC-provided services in order to avoid violating its statutory command of non-discrimination. Second, the Commission points to various aspects of its plan which do take into account concerns for the position of FX subscribers:
In fact, the Commission did consider the transitional impact on FX customers in designing its access plan. First, FX customers pay only non-premium access rates, even though it costs virtually the same to provide the FX line-side connections as it costs to provide the premium trunk-side connections____ Second, the FCC decided to assess the FX non-premium charge on a per minute rather than on a per line basis in order to protect those FX customers with relatively low usage____ Third, the Commission has recognized the “substantial hardship” that low-volume FX customers might suffer in areas in which no FX usage measurement capability now exists____ The FCC refused to allow carriers in such areas to charge a flat FX rate based on a monthly minute count of 4,076, which the FCC viewed as too high. Instead, the Commission permitted carriers to charge the equivalent of the local business rate, or to develop some other reasonable surrogate charge... These measures show that while the FCC may not have gone as far as ARINC might have liked, ARINC is certainly wrong to assert that the Commission entirely ignored the impact on FX customers of the new access rates.
Brief for the FCC at 106-108.
We begin by noting that the FCC’s first order in this matter appeared on March 11, 1983. The new plan will take effect, at the earliest, in June of 1984. Thus every subscriber to any service which will be assessed an increase in cost due to access charges will have in effect enjoyed a minimum period of transition of more than a year. This does not even include the far longer period of time beginning with the Second Supplemental Notice in April of 1980 when notice was first given that the FCC intended to remedy the discrimination in rates between MTS, OCC and FX users. Second Supplemental Notices, 77 F.C. C.2d 224 (1980). Charges that the FCC has acted precipitously with regard to any class of user simply do not wash. Exclamations of “rate shock” are similarly undermined when the period since first notice has itself become a protracted delay.
Second, the court need not accept AT & T’s characterization of FX users as that category of users “who, more than any other class of users, have escaped responsibility for massive interstate use of local facilities,” Brief for AT & T at 54, to nevertheless conclude that FX subscribers have been enjoying very advantageous benefits for many years. The size alone of the airlines’ FX network is eloquent testimony of advantages enjoyed by FX users over consumers of other means of interstate connection. FX service appears to be the functional equivalent of standard long-distance and like services. Nothing in the record indicates that FX customers using FX service deserve any better treatment than customers of those other services who will ultimately be required to bear the cost of the new access charges. FX customers have in the past enjoyed what amounted to a subsidy paid for by other long distance users.
Third, we are impressed with those features of the FCC’s plan that do cushion the blow for FX users. Indeed, we are hard pressed to understand why FX users are not being charged the premium rate for access as MTS customers, unless the FCC in fact gave some consideration to the rate shock which FX users will feel when introduced to the new order.
We must reject ARINC’s attempt to reclassify FX as something other than the equivalent of MTS (long distance) service. ARINC shrewdly attempts to obscure obvious differences between FX and other services such as PBX and Centrex-Co and then *442to argue that it deserves every consideration shown these categories of services. There are substantial differences. Throughout this proceeding the FCC has taken great pains to draw a great number of careful and often subtle distinctions. It has done so here as well. ARINC urges the court to redraw the lines and shuffle the categories — a chore for which we are manifestly not suited. While the court is most certainly “not a hostile stranger to the office of first instance”,43 neither is it the draftsman of intricate new plans.
We must also point out that it is possible both to accept the accuracy of ARINC’s portrayals of the cost impact about to hit the airlines and to dismiss that consideration as irrelevant. Woven into ARINC’s brief is the theme that, somehow, the airlines are different. But, so far as this appeal is concerned, they are not. AR-INC could not seriously argue that the impact upon a single industry of the FCC’s vast and ambitious reworking of the communications industry’s rate structure could in any way affect the outcome of our review. There are winners and losers galore as a result of the FCC’s plan which will eventually place the cost of services provided upon those who use the facilities. It is certainly not within the scope of this court’s review to seize upon the impact on a single industry as a reason for returning the FCC to its drawing board. Consideration of the FXfConnected question places us in the center of the FCC’s ambit of discretionary authority. Accordingly, we affirm this segment of its plan.
III. Conclusion
Over five years’ study of complex and far-ranging issues set the foundation for the FCC’s orders in this case. The Commission predicted industry trends and the impact of rate changes, reconciled widely diverse policy goals and numerous competing interests, and devised solutions to novel and difficult problems. Confronting rapid and fundamental changes in the telephone industry as a result of competition and fast-moving technology, the FCC acted for the most part with flexibility and care. The Commission’s work fell short of the ideal at several turns, but our review does not and cannot require perfection. See MCI Telecommunications Corp. v. FCC, 627 F.2d 322 (D.C.Cir.1980); VOLTAIRE, DICTIONNAIRE PHILOSOPHIQUE (Dramatic Art) (1794) (“Le mieux est l’ennemi du bien” — the best is the enemy of the good).
We do not find unreasoned the Commission’s determination that, to the extent practical, telephone prices should be “based upon the true cost characteristics of telephone company plant.” Access Order ¶ 28, 93 F.C.C.2d at 251. Nor, with limited exceptions, can we quarrel with the FCC’s exercise of judgment in deriving workable formulas, estimating numbers that cannot be mathematically proved, and making equitable assessments of the speed with which changes should be introduced.
Further refinements to the FCC’s plan could be debated endlessly. However, the Commission has reached a point in its deliberations where “a month of experience will be worth a year of hearings.” American Airlines, Inc. v. CAB, 359 F.2d 624, 633 (D.C.Cir.), cert. denied, 385 U.S. 843, 87 S.Ct. 73, 17 L.Ed.2d 75 (1966). The FCC has not achieved a plan of pristine quality. But “[i]n the words of Voltaire: Perfection is attained by slow degrees; it requires the hand of time.” Access Order, 93 F.C.C.2d at 364 (statement of Commissioner Quello).
For the reasons stated, we affirm the Commission’s orders in principal part, and remand for further consideration consistent with this opinion the segments on party line service and small telephone companies’ election of “average schedule company” status.
It is so ordered.
. Until recently, 22 larger local telephone companies — the Bell Operating companies (BOCs)— were subsidiaries of AT & T. In 1982, the Justice Department and AT & T entered into a consent decree providing that AT & T would divest itself of the BOCs as of early 1984. See United States v. AT & T, 552 F.Supp. 131 (D.D.C. 1982), aff’d mem., 460 U.S. 1001, 103 S.Ct. 1240, 75 L.Ed.2d 472 (1983). In addition to the BOCs, almost 1500 independent local companies provide local service.
. In the early days of telephone service, pay telephone (coin box) revenues, in addition, were vitally important to local telephone companies. Cf. Illinois Bell Tel. Co. v. Moynihan, 38 F.2d 77 (N.D.Ill.) (striking down coinbox rates prescribed by Illinois Commerce Commission as confiscatory), rev'd, Smith v. Illinois Bell Tel. Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930).
. When the local company was a subsidiary of AT & T, the reimbursement took place through a division of revenues within the AT & T corporate family.
. Prior to the creation of the FCC, see Communications Act of 1934, 47 U.S.C. §§ 151-609 (1976 & Supp. V 1981), the Interstate Commerce Commission exercised regulatory authority over interstate wire communication.
. All interstate revenues were paid into an interstate “pool.” AT & T allocated a share of revenues out of the pool to each local telephone company in proportion to its share of the total *400investment allocated to the interstate jurisdiction. See Brief for FCC at 10.
. MCI contends that the FCC has no responsibility to promote universal service because “the Commission has no statutory responsibility for local communications," and therefore “it would be unlawful for it to promulgate a policy designed to ... benefit local communications.” Brief for Intervenor MCI Telecommunications Corp. at 19 n. 20. We reject MCI's argument. Congress directed that, “so far as possible, ... all people of the United States” are to have adequate telephone facilities at reasonable prices.
. This position is advanced by petitioners National Association of Regulatory Utility Commissioners and California, and by intervenors National Association of State Utility Consumer Advocates, et at, Florida Public Service Commission, New York State Department of Public Service, Independent Alliance, and Missouri Public Service Commission. It is addressed by the FCC and intervenors MCI Telecommunications Corp., AT & T, BOCs, United States Independent Telephone Association, IBM Corp., and Rochester Telephone Company.
. These arguments are made by petitioners National Association of Regulatory Utilities Commissioners and California, and by intervenors National Association of State Utility Consumer Advocates, Michigan, Rural Telephone Coalition, Florida Public Service Commission, and Independent Alliance. They are addressed by the FCC and intervenors AT & T, BOCs, United States Independent Telephone Association, and Rochester Telephone Company.
. Parties mounting these attacks are petitioners Public Service Commission of the District of Columbia and North American Telecommunications Association. The FCC, supported by intervenors BOCs and ROLM Corp., defends its actions.
. This attack is made by petitioner Rural Telephone Coalition.
. See supra note 10.
. Petitioner MCI Telecommunications Corp. and intervcnor Satelco, et at make this claim; the FCC’s position is supported by intervenors AT & T, BOCs, GTE Corp., Ad Hoc Telecommunications Users Committee, and Association of Data Processing Service Organizations.
. Petitioner Aeronautical Radio, Inc. and intervenor American Petroleum Institute, et al. make this claim; the FCC is supported by intervenors AT & T and BOCs.
. Intervenor American Petroleum Institute levels this attack.
. Petitioner Aeronautical Radio, Inc. here contests the FCC’s decision; intervenors AT & T and BOCs defend the FCC.
The validity of the premium carrier’s charge levied against AT & T above that assessed against the OCCs is under consideration in a separate review proceeding. AT & T v. FCC, No. 84-1087 (D.C.Cir. filed March 9, 1984).
."Petitioners,” as used in our analysis, describes both petitioners and intervenors appearing in opposition to the Commission's decision.
. In a later opinion, Chief Justice Hughes characterized Smith as "a case where the segregation of properties and business was essential in order to confine the exercise of state power to its own proper province.” Lone Star Gas Co. v. Texas, 304 U.S. 224, 241, 58 S.Ct. 883, 891, 82 L.Ed. 1304 (1938).
. Petitioners emphasize this statement. See, e.g., Brief of Petitioner National Association of Regulatory Utility Commissioners at 26; Brief of Intervenor National Association of State Utility Consumer Advocates, et al. at 14 (hereafter, NASUCA Brief); Brief of Intervenor New York State Department of Public Service at 21. They argue that the FCC’s plan is inconsistent with Smith because under the FCC’s plan "local exchange customers ... bear all the burden.” NASUCA Brief at 20.
We read less into the Court’s words. The state commission, the Court explained, must count some exchange plant as interstate to avoid retaining interstate property, but not interstate revenues, in its calculations. By including interstate property in its jurisdictional domain, the state commission would artificially inflate the intrastate rate base, and hence intrastate charges.
.Some petitioners argue that their view of Smith became statutory law when Congress enacted 47 U.S.C. § 410(c) (1976). See NASUCA Brief at 16-23. Section 410(c) provides for the establishment of a Federal-State Joint Board to make recommendations to the FCC with regard to the separations process. The provision was introduced in 1971 at the urging of the National Association of Regulatory Utilities Commissioners (NARUC). NARUC complained that the FCC *408was allocating too many costs to the intrastate jurisdiction, to be recovered through local charges, and not enough to the interstate jurisdiction, to be recovered through AT & T long distance charges. See 117 Cong.Rec. 15,979-81 (1971).
NASUCA contends that the FCC’s plan "reduces the Joint Board proceedings under § 410(c) to a mere sham.” NASUCA Brief at 19. We cannot agree; Joint Board proceedings, now as before, determine which costs shall be allocated to the jurisdiction of the state commissions and which to the jurisdiction of the FCC. As support for the position that Smith and the § 410(c) — endorsed separations process go beyond jurisdiction and indeed dictate how costs must be recovered, NASUCA cites material printed in the Congressional Record by Senator Magnuson when he introduced the Senate version of § 410(c). The material states that the separations process should avoid placing any "unreasonable burden ... on either the interstate or intrastate users of the telephone service." It emphasizes the importance of local service, and of "allocatfingj a fair amount of the cost of providing local telephone service to the users of the interstate service." 117 Cong.Rec. 15,980 (1971). NASUCA leans too heavily on these words. When § 410(c) was enacted, local telephone plant costs in the interstate jurisdiction were in fact recovered through usage-based charges. Senator Magnuson may have been concerned with protecting the legitimate interests of the state commissions under that system. Nothing in the legislative history of § 410(c), however, convinces us that the section was meant to freeze in place for all time the 1971 system for recovery of costs.
. Section 2(b) reads in part:
[Sjubject to the provisions of section 301 of this title, nothing in this chapter shall be construed to apply or to give the Commission jurisdiction with respect to (1) charges, classifications, practices, services, facilities or regulations for or in connection with intrastate communication service by wire or radio of any carrier, or (2) any carrier engaged in interstate or foreign communication solely through physical connection with the facilities of another carrier not directly or indirectly controlling or controlled by, or under direct or indirect common control with such carrier ... except that sections 201 to 205 of this title shall, except as otherwise provided therein, apply to carriers described in clause [ ] (2)
47 U.S.C. § 152(b) (Supp. V 1981).
. Petitioners also cite § 221(b) of the Act, which reads:
Subject to the provisions of section 301 of this title, nothing in this chapter shall be construed to apply, or to give the Commission jurisdiction, with respect to charges, classifications, practices, services, facilities, or regulations for or in connection with wire, mobile, or point-to-point radio telephone exchange service, or any combination thereof, even though a portion of such exchange service constitutes interstate or foreign communication, in any case where such matters are subject to regulation by a State commission or by local governmental authority.
47 U.S.C. § 221(b) (1976).
Section 221(b) is irrelevant to the problem before us; its limitation on Commission regulation of telephone exchange service "was merely intended to preserve state regulation of local exchanges that happen to overlap state lines.” Computer and Communications Indus. Ass’n v. FCC, 693 F.2d 198, 216 (D.C.Cir.1982), cert. denied, 461 U.S. 938, 103 S.Ct. 2109, 77 L.Ed.2d 313 (1983).
. In practice, such an adjustment would be unworkable. It would be prohibitively complex and inefficient to have the separations formula vary from subscriber to subscriber. Any equivalent adjustment would have to be based on the totality of subscriber plant investment and expenses.
. California additionally argues that the Commission’s power under § 205 of the Communications Act should be limited by § 203, which provides that all communications common carriers except "connecting carriers” — including those engaged in interstate communications "solely through physical connection with the facilities of another [unrelated] carrier,” 47 U.S.C. §§ 153(u), 152(b) (1976 & Supp. V 1981) —must file public schedules of their charges for interstate wire or radio communication. California reasons that just as connecting carriers are exempt from the tariff-filing requirement of § 203, they should be immune from Commission power under § 205. California Brief at 12. Sections 203 and 205, however, serve different goals. The § 203 filings of a non-connecting carrier in any case show "all charges for itself and its connecting carriers," 47 U.S.C. § 203, and nothing in § 203 prevents the Commission from thus evaluating the charges of connecting carriers, finding them unlawful, and prescribing just and reasonable charges under § 205. Further, California's argument that connecting carriers are wholly safe from the Commission’s § 205 power seems hard to reconcile with 47 U.S.C. § 152(b), which provides that "sections 201 to 205 of this title shall, except as otherwise provided therein, apply to [connecting carriers].” But see New York Brief at 11-12, 23-24 (quoting § 152(b) but omitting crucial language).
. But cf. California Brief at 16, 18 ("Today, ... as a result of the historical development of alternative means of communications, most large users have already bypassed the local exchange”; nonetheless, "the limited threat posed by 'uneconomic' bypass does not justify the extreme measures proposed by the FCC.”).
. NARUC and RTC invoked § 214 of the Act as empowering the Commission to act as these petitioners proposed. It appears, however, that the Commission's § 214 powers do not extend to entities other than common carriers. See 47 U.S.C. § 214 (1976).
California suggests that "the practical alternative to bypass lies not in prohibiting new servic*413es nationwide, but in [the states’] addressing the phenomenon as it actually arises in actual situations." California Brief at 17. To the extent California means the FCC should have remitted to the states the task of prohibiting construction of specific bypass facilities, the Commission appropriately refused to rely on local agencies to make technological judgments the Commission considered itself incompetent to undertake successfully. To the extent California means the state commissions should be left to address the bypass threat through “rate-making structures [that] provide for a reduced NTS payment by high-volume users” on a case-by-case basis, see Oral Argument Transcript at 65-68 (statement of Bruce Renard, counsel for Florida Public Service Commission), we again find the Commission’s decision reasonable. The FCC could well conclude that it was unsound to address distortions in the interstate rate structure by introducing complementary distortions in the intrastate rate structure.
. Michigan argues similarly that no evidence supports the Commission's concern that pricing long-distance service above cost artificially represses usage. Brief of Intervenors the State of Michigan and the Michigan Public Service Commission at 7. Michigan notes the Commission’s statement that it ”ha[d] some questions concerning the methodology used in [one] study” of long-distance calling, and that it found "[certain assumptions underlying the results of [another] study ... questionable or unclear.” See Access Order ¶ 112, 93 F.C.C.2d at 275 & n. 39.
We do not see the Commission’s recognition of flaws in studies it considered as tantamount to a rejection of those studies as worthless. The Commission’s reference to "artificialf ] restriction of] calling patterns,” id., we further note, did not figure as a major item in its analysis.
. One petitioner criticizes Commission reliance on economic theory in allocating the costs of access to the local exchange. California Brief at 15. California, however, has not persuaded us that "the generally accepted principle that recovery of fixed costs through usage charges impairs economic efficiency is not applicable to telecommunications,” Further Reconsideration Order ¶ 11, 49 Fed.Reg. at 7,811, nor has it convinced us that economic efficiency is not a legitimate Commission goal.
Certain petitioners concede that "the Commission's decision to impose flat rates for NTS cost recovery follows logically from the perceived problems of discriminatory and preferential rates and uneconomic bypass,” but argue that the agency has not justified "imposition of such flat rates on end users rather than on interexchange carriers.” Brief of Intervenors Roseville Telephone Company, et al. at 23. Imposition of flat rates on interexchange carriers, however, ultimately results in passing those costs on to end users on a usage-sensitive basis. A pass-on of that kind, if it became the final solution, would defeat the FCC’s goal.
. Additionally, the court viewed the Commission’s action as involving promulgation of prospective rules to implement agency policy, not "adjudicat[ion of] disputed facts in particular cases” for which a trial-type hearing might be appropriate. 572 F.2d at 22 (quoting United States v. Florida East Coast Ry., 410 U.S. 224, 245, 93 S.Ct. 810, 821, 35 L.Ed.2d 223 (1973)).
. We express no opinion on whether the Act or longstanding Commission policy calls for an evidentiary hearing where the Commission determines, for example, whether a carrier-filed rate is just and reasonable; that kind of case is not before us. But see Investigation of Access and Divestiture Related Tariffs, FCC 84-201, at ¶ 113 (released May 15, 1984) (resolving cost issues and determining the legality of rates through notice and comment proceedings; finding that parties received "the full opportunity for hearing required by Section 205(b) of the Act”).
. Intervenors Roseville Telephone Company, et al. claim that the Commission gave insufficient notice of its intention to impose a plan involving flat-rate end user access charges. Brief of Intervenors Roseville Telephone Company, et al. at 8-15. The Commission’s request for comments, however, made it clear that the Commission was considering an approach under which ‘‘[ejvery customer ... would pay a flat (per line) access charge that did not vary with use, plus usage based interstate charges that reflected only usage sensitive facilities ... plus local charges.” MTS and WATS Market Structure: Fourth Supplemental Notice of Inquiry and Proposed Rulemaking, 90 F.C.C.2d 135, 140 (1982). The Commission described this approach as "extremely attractive” from the point of view of "equity between services and economic efficiency," but expressed concern about its effect on universal service. Id. at 140 & n. 10. Moreover, the Commission detailed a second approach also involving "flat-rate access charge[s],’’ id. at 142-44, and requested comment on the bypass issue. Id. at 145-47.
. The Commission stated:
We are sympathetic to the Rural Telephone Coalition’s concern that party-line service is an inferior grade of service that may deny many customers the benefits of more advanced communications services. Nevertheless, we believe that these customers will be able to decide whether the benefits to be gained from single-line service are worth the added costs. Interstate access charges that reflect costs will permit their choice of single or party-line service to be a rational one.
Reconsideration Order ¶ 41, 48 Fed.Reg. at 42,-990.
. The Commission noted, ”[i]t may be difficult for such affiliates [of cost companies] ... that are presently compensated as average schedule companies to develop cost data.” Reconsideration Order ¶ 195, 48 Fed.Reg. at 43,014.
. We note that this analysis can apparently be completed without jeopardy to the efficient, smooth implementation of the Commission’s orders, since the Commission has waived the requirement that companies elect between cost and average schedule status pending the modernization of average cost schedules. MTS and WATS Market Structure, 49 Fed.Reg. 10,549 (1984).
. Under the ENFIA agreement, OCCs paid progressively higher access rates. That transitional scheme was based partially on the increasing quality of interconnection, but also reflected a compromise rate level between the access rates paid by AT & T under separations regulations and that paid by other customers for local access. See ENFIA, 71 F.C.C.2d 440, 446-47, 455 (1979); ENFIA, 90 F.C.C.2d 6, 16-17 (1982). The OCC rate effectively increased as the volume of OCC business also increased, even • though the quality of access was not correspondingly higher. See Brief for FCC at 78.
. The Commission stated that while the physical facilities used to offer exchange access to OCCs may provide a lower quality interconnection than that offered AT & T, "[i]t is not clear, however, that this inferior level of interconnection is any cheaper to provide. Cost-based pricing would appear to require that all carriers pay their full costs regardless of any quality differences.” Access Order ¶ 151, 93 F.C.C.2d at 286. By determining that OCCs should not be required to pay the higher effective access rates levied against AT & T the Commission expressly acted to ameliorate the adverse impact on the OCCs’ customer base which would have accompanied fully equivalent rates for varying quality access costing roughly the same to provide. Those aspects of the Commission’s orders charging different access rates to AT & T and MCI are under appeal in a separate proceeding, AT & T v. FCC, No. 84-1087 (D.C.Cir. filed March 9, 1984). We express no opinion on the issues raised in that proceeding, holding only that transitional considerations based on preserving universal subscription and avoiding disruptions in service may lawfully be considered in structuring a rate scheme.
. E.g., Access Order ¶ 33, 93 F.C.C.2d at 252: "[T]his decision [to address the problem of uneconomic bypass] does not, in any way, constitute a judgment that subsidizing the costs of basic telephone service for certain customers or for all customers is improper.” See also AT & T, 78 F.C.C.2d 1296, 1297 (1980). Of course, the Commission’s finding of an unjustified subsidy may precede a holding that rates are unreasonably discriminatory. See, e.g., Western Union Int'l, Inc. v. FCC, 568 F.2d 1012, 1019 (2d Cir. 1977).
. American Trucking Ass'ns, Inc. v. FCC, 377 F.2d 121, 129-31 (D.C.Cir.1966), cert. denied, 386 U.S. 943, 87 S.Ct. 973, 17 L.Ed.2d 874 (1967), establishes only that once the Commission has identified a rate as unlawfully discriminatory, the carrier is not entitled to perpetuate that rate as serving other important policy interests under the Act. Cf. Aeronautical Radio, Inc. v. FCC, 642 F.2d 1221, 1237 (D.C.Cir.1980), cert. denied, 451 U.S. 920, 101 S.Ct. 1998, 68 L.Ed.2d 311 (1981) (Wilkey, J., dissenting); Nader v. FCC, 520 F.2d 182, 211 (D.C.Cir.1975) (Fahy, S.J., dissenting) (both cases upholding FCC action in the absence of an initial finding of unlawful discrimination).
. MTS & WATS Market Structure, Report and Second Supplemental Notice of Inquiry, 77 F.C. C.2d 224 (1980) ("Second Supplemental Notice").
. "[W]e shall use our best judgment to develop an initial surcharge level, pending development of [other] charges by the carriers. First, we note that private lines attached to a PBX are capable of leaking' into local exchange. Because most private lines are connected to PBXs, most private lines are capable of leaking. Although one might assume that all private lines would leak if capable of doing so, we are aware of some private lines connected to PBXs that actually may not be used in connection with local exchange services to make interstate calls. We believe a fair estimate of the number of such lines would be 20 percent of all private lines. Thus, we estimate that 80 percent of all private lines do leak through a PBX or other patching or switching device. We shall assume that 8 percent of all communications made over such lines are interstate, based on the latest data available to us on average subscribed line usage for interstate MTS and WATS services. Eight percent of 80 percent is 6.4 percent, which represents the proportion of all private line usage that "leaks” into the local exchange. We further assume, based on estimates submitted in this proceeding, that nonpremium carriers would pay approximately $400-$500 in monthly carrier usage charge under the access charge plan. Taking 6.4 percent of these figures, we arrive at a range of approximately $25-$32 per month per line. We will select the lower end of this range, $25, as a conservative estimate of what the interim surcharge should be." Reconsideration Order, ¶ 88, 48 Fed.Reg. at 42,999.
. API’s counsel stated at argument that his clients would be forced to deal with multiple state commission filings as a result of the Commission's statement. We do not view such efforts as the kind of "burden" envisioned by the Abbott test. Interested parties will of course have to maintain surveillance of the regulatory environment in which they function. But this is an ordinary cost of doing business, and is not an inescapable result of the Commission’s announcement.
. Recent years have witnessed a renewal of interest in the traditional role of nondelegation doctrine. See American Textile Mfgs. Inst., Inc. v. Donovan, 452 U.S. 490, 543-48, 101 S.Ct. 2478, 2508-10 (Rehnquist, J., dissenting). See also, Note, "Rethinking the Nondelegation Doctrine”, 62 B.U.L.Rev. 257 (1982), Lane, "Schechter and the FTC: A Roving Commission", 39 Bus.Lawyer 153 (1983). As attention to this area of our law grows, it refocuses thought on one of the rationales against excessive delegation: the harm done thereby to principles of political accountability. Such harm is doubled in degree in the context of a transfer of authority from Congress to an agency and then from agency to private individuals. The vitality of challenges to the former type of transfer is suspect, but to the latter, unquestionable.
. “Foreign” indicates an exchange in another state, and not one in a foreign country.
. Greater Boston Television Corporation v. FCC, 444 F.2d 841, 852 (D.C.Cir.1970), cert. denied, 403 U.S. 923, 91 S.Ct. 2233, 29 L.Ed.2d 701 (1971).