CELEBREZZE, Senior Circuit Judge.Page 736

Petitioners seek review of a series of decisions made by the Federal Communications Commission (FCC). In these decisions, the FCC has attempted to restructure the international telecommunications markets by heightening competition between the two principal modes of telecommunications: satellite systems and cable systems. Essentially, the decisions permit the Commercial Satellite Corporation (Comsat), which had functioned previously as a wholesaler of satellite services, to sell its satellite services directly to the public, a role which previously had been performed only by petitioners. By permitting Comsat to operate as a retailer of telecommunication services, the FCC expects to foster competition between satellite and cable systems and, thus, to promote greater efficiency in both systems. The issue in this case is simply stated: whether the FCC may permit Comsat to sell its satellite services to the general public and, if it may, whether it properly has exercised its power in this instance.

I. HISTORICAL BACKGROUND

A. The International Telecommunications Industry. Two means of transmitting international telecommunications messages are generally available: cable and satellite. The United States communications common carriers own and operate the international cable network. The communications satellites in dispute are owned and operated by the International Satellite Telecommunications Organization (Intelsat). The Communications Satellite Corporation (Comsat) is the only United States entity permitted direct access to Intelsat's satellite system. Comsat, in turn, is authorized by Congress to provide satellite transmission service only to "authorized entities." 47 U.S.C. Sec. 735(a)(2). In 1966, the FCC established a policy which generally limited the class of authorized users to common carriers. Authorized Entities and Users, 4 FCC 2d 421 (1966) ( Authorized User I). As a consequence of the FCC's 1966 policy, a two-tiered market developed; Comsat acted as a "wholesaler" by leasing satellite circuits to the common carriers, who in turn, acted as "retailers" leasing satellite circuits to the public for transmission of international telecommunications messages.

When Comsat first offered commercial communications satellite service in 1965, the international telecommunications markets were rigidly compartmentalized; two separate and distinct classes of common carriers operated at the retail level: voice and non-voice. American Telephone Telegraph Company (AT T) enjoyed a monopoly in the international voice communications market, while petitioners dominated the non-voice or record services market. Although technological developments in the international telecommunications industry have blurred the traditional distinction between "record" and "voice" systems, see, e.g., Overseas Communications Services, FCC 82-547 (1982), petitioners are still referred to as international record carriers (IRCs).

The IRCs provide two types of communications services: exchange and leased-channel. Exchange messages are sent through a common switched network which connects the users to other subscribers of the system. Examples of exchange-type services offered by the IRCs include telegraph, telex, and TWX. In contrast, a leased-channel is a private line between two or more specified points. Unlike exchange service, a leased-channel user leases all or part of a circuit at an agreed rate. Because leased-channel is more expensive than exchange service, and because the majority of users do not send a sufficient number of messages to require leased-channel service, the consumer market for leased-channel service is limited to a few large and sophisticated users.

Generally, exchange services are provided by the IRCs on an "end-to-end" basis. As its name implies, end-to-end service is a complete service. When exchange messages are sent by satellite, the IRCs make necessary arrangements with a U.S. domestic carrier for connecting circuits between operating centers and U.S. earth stations. The IRCs then lease satellite half-circuits from Comsat to transmit messages from earth stations to satellites. To complete the communications link, the IRCs lease satellite half-circuits from Comsat's foreign counterparts and arrange for connecting links between foreign earth stations and final destination points. When exchange messages are sent by cable, the IRCs make necessary arrangements with U.S. domestic carriers to sent exchange messages from hinterland operating centers to IRCs operating centers in designated "gateway" cities. The IRCs then transmit the messages through their submarine cable system and make arrangements with foreign communication corporations for connecting links.

Although the IRCs can provide leased-channel service on an end-to-end basis, most leased-channel customers are capable of providing their own link-up facilities. Consequently, leased-channel service generally is not provided by the IRCs on an end-to-end basis; instead, the IRCs simply provide service between designated service points. When the point to point service is between a satellite and a U.S. earth station, the service is called "basic" transmission service.

B. Authorized User I and Earth Station Ownership Policies. In 1966, the FCC concluded that it had authority, pursuant to the Satellite Act of 1962, to designate non-carriers as authorized users "upon a proper finding that non-carrier access to the space segment would serve the public interest and comport with the purposes and policies of the Satellite Act." Authorized User I, 4 FCC 2d 421, 428 (1966). The FCC, however, restricted the class of authorized users to carriers, absent "unique and exceptional circumstances," in order to protect the IRCs from potentially ruinous competition. The FCC feared that unless the IRCs were insulated from such direct competition, Comsat would capture the leased-channel market, which accounted for approximately 18.9 percent of the IRCs' total revenues, and that the IRCs would be forced to raise their rates for exchange services. Id. at 433. Because the cost benefits of satellite technology would inure to a few leased-channel customers at the expense of many exchange service customers, id. at 432-33, the FCC concluded that such a result would be inconsistent with the primary policy of the Satellite Act to extend the benefits of satellite transmission to all users. Id. See Authorized User II, 90 FCC 2d 1395, 1401 (1982).

The FCC recognized that its decision to limit Comsat's role to that of a "carriers' carrier" would have certain anticompetitive effects. Specifically, the FCC acknowledged that its policy would neutralize intermodal competition between satellite and cable, and that absent such competition, the carriers would favor cable over satellite even though satellite offered a less expensive means of transmission. Accordingly, to assure that all users would enjoy the lower costs of satellite technology, the FCC established the "prescribed use" formula which required carriers to use satellite and cable circuits in approximately equal proportions, and the "composite rate" policy which required carriers to charge tariffs which took into account any cost savings due to the increased use of the less expensive satellite system.

Initially, the FCC limited ownership of the first three earth stations to Comsat. Proposed Global Commercial Communication Satellite System, 38 FCC 1104 (1965). This policy was to be in effect for two years during which time the FCC would consider applications for earth station ownership filed by Comsat and other qualified carriers. However, at approximately the same time that the FCC established its Authorized User I policy, it revised its earth station ownership policy and required joint ownership of all earth stations by Comsat and authorized carriers; Comsat was permitted to own a 50 percent interest in each earth station while the other interested carriers divided the remaining 50 percent interest. Subsequently, the carriers and Comsat formed the Earth Station Ownership Committee (ESOC) to make policy and investment decisions. Comsat assumed the general managerial responsibilities of the ESOC. This joint ownership policy was to be in effect for only a few years. In 1969, the FCC issued a notice of inquiry to review its earth station ownership policy, Earth Station Ownership Operation, 20 FCC 2d 735 (1969), but did not modify its policy; rather, it decided to maintain its joint ownership policy until further notice. This policy remains in effect today.

C. The SIN Decisions. In Communications Satellite Corp., 38 FCC 1298, 1304 (1965), the FCC established a temporary "carrier of the week" policy, which required customers who desired Intelsat television service to obtain this service from whichever carrier was designated as "carrier of the week." Thus, AT T and the IRCs offered television transmission service on a rotating basis. The FCC, believing that its carrier of the week policy promoted quality service, continued the policy indefinitely. American Tel. Tel. Co., 18 FCC 2d 402 (1969). In 1978, the Spanish International Network (SIN), a network which purchases Spanish television programming from foreign countries and sells it in the United States, petitioned the FCC for authorized user status. Spanish International Network, 68 FCC 2d 1260 (1978). SIN argued that television transmission was a "unique and exceptional" service and, thus, that it was entitled to authorized user status under the FCC's Authorized User I policy. Further, SIN believed that elimination of the carrier of the week policy would best serve the public interest.

Although the FCC refused to characterize television transmission service as "unique and exceptional," it did initiate a proceeding to determine whether it should waive its Authorized User I policy and permit networks to obtain television transmission service directly from Comsat. After considering comments submitted by interested parties, the FCC waived its Authorized User I policy and allowed television networks to obtain television transmission service directly from Comsat. Spanish International Network, Inc., 70 FCC 2d 2127 (1978) ( SIN I). Pursuant to SIN I, the FCC permitted Comsat to provide television services directly to all authorized users at U.S. earth stations and to all international television carriers individually. Communications Satellite Corporation, 76 FCC 2d 5 (1980) ( SIN II). Additionally, Comsat was permitted to revise its tariff for this new service. Communications Satellite Corporation, 79 FCC 2d 562 (1980) ( SIN III).

D. Authorized User II. The FCC issued a notice of proposed rulemaking to review its Authorized User I policy in Aeronautical Radio, Inc., 77 FCC 2d 535 (1980). In its notice, the FCC proposed to increase intermodal competition between cable and satellite systems by permitting Comsat to compete with the carriers at the retail level. Further, the FCC proposed to review its "prescribed use" and "composite rate" policies to determine whether these policies continued to serve the public interest. After considering the comments submitted by interested parties, the FCC abandoned its Authorized User I policy and established a new authorized user policy which permits non-carrier entities to obtain authorized user status and, thus, to purchase basic transmission services directly from Comsat. Moreover, the FCC permitted Comsat to establish a separate subsidiary to compete with the carriers in the end-to-end market upon the condition that Comsat treat all authorized users in a fair and consistent fashion. Finally, the FCC decided to make its composite rate policy discretionary for the IRCs and to limit its role in prescribing the use of cable and satellite facilities.

The FCC justified its decision in Authorized User II by enumerating changes in circumstance. The primary justification for initially restricting Comsat's entry into the retail markets, was that its entry would result in a substantial decrease in the IRCs' leased-channel revenues. Because leased-channel revenues appeared to be the fastest growing segment of the IRCs' business, and because leased-channel business was one of the IRCs' most profitable ventures, the FCC feared that a loss in leased-channel revenue would cause an increase in the IRCs' exchange service rates. Thus, the FCC concluded that the benefits of satellite technology would inure to the few leased-channel users at the expense of the many exchange service users and that such a result was incompatible with the Act's primary policy of extending the economies of satellite technology to all users.

A crucial assumption underlying the FCC's Authorized User I decision was that the cost of satellite circuits was substantially less expensive than cable circuits. In Authorized User II, the FCC concluded that the price difference was no longer substantial and that intermodal competition between cable and satellite was feasible. Consequently, the FCC no longer predicts a substantial revenue diversion due to losses from leased-channel revenue. Moreover, the FCC predicts that even if such a diversion does occur, the losses will not affect the IRCs' "overall profitability to such an extent as to impair their ability to provide service." Authorized User II, 90 FCC 2d 1394, 1416 (1982). In short, the FCC found that protecting the carriers from competition with Comsat for leased-channel business was no longer in the public interest.

Petitioners warned the FCC that its Authorized User II POLICY would permit Comsat to use its "monopoly" position to engage in predatory pricing and to cross-subsidize its subsidiary. Further, petitioners urged the FCC to consider the merits of direct access and independent earth station ownership, two related issues, prior to permitting Comsat to enter the industry's retail markets. The FCC found that the petitioners' concerns regarding Comsat's potentially illegal activities were premature: the FCC believed that the regulatory tools at its disposal were sufficient to prevent Comsat from engaging in predatory pricing. Further, the FCC determined that its requirement that Comsat establish a separate subsidiary prior to entering the retail exchange service market would permit the FCC to monitor effectively Comsat's activities. Accordingly, if Comsat engaged in a practice of cross-subsidizing its subsidiary, then the FCC would take appropriate action. Finally, the FCC concluded that it could properly implement its change in authorized user policy prior to resolving the direct access and independent earth station ownership issues. In any event, the FCC indicated that it required more information before it could resolve these complex issues. In an effort to collect necessary information regarding these issues, the FCC issued two notices of inquiry.

II. STANDARD OF REVIEW

When reviewing a "notice-and-comment" rulemaking proceeding by the FCC, the reviewing court must determine whether the agency's decision is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. Sec. 706(2)(A). Petitioners challenge the FCC's Authorized User II decision on two grounds. First, petitioners contend that the FCC improperly concluded that the Satellite Act of 1962, 47 U.S.C. Sec. 701 et seq., grants the FCC broad discretion to designate non-carriers as authorized users. Alternatively, petitioners argue that the FCC's decision was arbitrary and an abuse of discretion because the FCC failed to consider adequately a number of factors relevant to its decision in Authorized User II.

When an agency interprets a statute which it has been charged with administering, the agency's interpretation must be accepted, see, e.g., Ft. Pierce Utilities Authority of Ft. Pierce v. United States, 606 F.2d 986 (D.C. Cir.), cert. denied, 444 U.S. 842, 100 S.Ct. 83, 62 L.Ed.2d 54 (1976), unless its conclusion is inconsistent with obvious congressional intent. See, e.g., United States ex rel. Dancy v. Arnold, 572 F.2d 107 (3rd Cir. 1978). When reviewing an agency's policy decision, however, we must determine whether its decision is arbitrary, capricious or an abuse of discretion. In determining whether an agency's action is arbitrary or an abuse of discretion, a reviewing court must assume that the agency acted properly, e.g., Ethyl Corporation v. E.P.A., 541 F.2d 1 (D.C. Cir.) (en banc), cert. denied, 426 U.S. 941, 96 S.Ct. 2663, 49 L.Ed.2d 394 (1976), and it must refrain from substituting its judgment for that of the agency. E.g., Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 824, 28 L.Ed.2d 136 (1971). A court's inquiry, however, is not limited to determining whether the agency's action has a rational basis, e.g., Texaco, Inc. v. F.E.A., 531 F.2d 1071 (TECA), cert. denied, 426 U.S. 941, 96 S.Ct. 2662, 49 L.Ed.2d 394 (1976); a reviewing court must also determine whether the agency's decision was based on a consideration of all relevant factors. E.g., Motor Vehicle Mfrs. Ass'n v. State Farm Mut., ___ U.S. ___ (1983), 103 S.Ct. 2856, 2866-67, 77 L.Ed.2d 443 (1983); Bowman Transportation v. Arkansas — Best Freight System, Inc., 419 U.S. 281, 285, 95 S.Ct. 438, 441, 42 L.Ed.2d 447 (1974) ("A reviewing court `must consider whether the decision was based on a consideration of relevant factors.'"), quoting Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971); Ethyl Corp. v. E.P.A., 541 F.2d 1, 36 (D.C. Cir. 1976) (en banc) (Close scrutiny is necessary "to determine whether the agency decision was rational and based on consideration of relevant factors"); In re Dept. of Energy Stripper Well Exemption, 520 F.Supp. 1232, 1268 (D.Kan. 1981) ("The relevant factors test simply requires that the court determine the underlying issues or factors the agency must consider in order to reach a rational result."). With these principles in mind, we turn to petitioners' arguments.

III. THRESHOLD QUESTION

The threshold question is whether the Satellite Act of 1962, 47 U.S.C. Sec. 701 et seq., permits the FCC to designate non-carriers as authorized users which are able to lease satellite channels directly from Comsat. The plain language of the Act indicates that the class of "authorized users" contains entities other than carriers. Moreover, the plain language, the legislative history, and the statutory scheme of the Act, convince us that Congress intended to give the FCC broad discretion to designate non-carriers as authorized users and that the FCC is not required to maintain a two-tiered structure in the industry. Accordingly, we hold that the Act grants the FCC broad discretion to designate non-carriers as authorized users.

Although the Satellite Act does not define the class of permissible authorized users, the parties agree that the FCC is the proper licensing agency in this instance, and that the FCC may, under some circumstances, designate non-carriers as authorized users; the question is which non-carriers are eligible for authorized user status. Petitioners argue that the only non-carrier entity entitled to authorized user status is the United States government. According to petitioners, such a limitation on the FCC's discretion is consistent with Congress' predictions that Comsat would serve the industry as primarily a "carriers' carrier." In response, the FCC asserts that it may designate non-carriers as authorized users whenever it finds that such a designation will serve the public interest. We believe that the FCC's statutory interpretation is correct.

The most basic rule of statutory construction requires that courts attribute to the words of a statute their plain meaning. See, e.g., Banks v. Chicago Grain Trimmers, 390 U.S. 459, 465, 88 S.Ct. 1140, 1144, 20 L.Ed.2d 30 (1968). The Act indicates that Comsat may lease satellite channels to "United States communications common carriers and to other authorized entities, foreign and domestic." 47 U.S.C. Sec. 735(a)(2) (emphasis added), and that Comsat may "contract with authorized users, including the United States Government, for the services of the communications satellite system." 47 U.S.C. Sec. 735(b)(4) (emphasis added). Congress has indicated that Comsat may deal directly with "carriers", a well-defined group, as well as with "other authorized entities," a group whose membership is unclear. Petitioners would define the members of the latter class as including only carriers and the United States government. Such an interpretation, however, is inconsistent with the plain meaning of the statute because it renders the statutory language "other authorized entities" a redundancy. If Congress intended to restrict the class of authorized users to authorized carriers and the United States government, it knew how to do so. Although the intended scope of the class of "authorized entities" is not defined expressly by the Act, the plain language indicates that the words "other authorized entities" mean non-carriers, including the United States government, who receive a license from the appropriate licensing agency. Thus, the plain language of the Act does not limit the FCC's discretion to designate non-carriers as authorized users.

The legislative history indicates that Congress specifically considered language which would have limited Comsat's ability to offer retail services. Prior to the passage of the Act, the FCC objected strenuously to the proposed language "and to other authorized entities" because it believed that such language was broad enough to allow Comsat to compete with carriers on the retail level. In February of 1962, Newton Minow, Chairman of the FCC, urged the Senate Committee on Aeronautical and Space Sciences, which was considering Senate Bill S. 2814, to delete "other authorized entities" and to restrict expressly Comsat's ability to enter the retail market. Chairman Minow offered the Senate Committee a proposed revision of S. 2814 which would have limited Comsat's ability to serve the general public. To assure that Congress was aware of the FCC's opposition to S. 2814, Chairman Minow also appeared before the House Committee on Interstate and Foreign Commerce on March 14, 1962 and warned the Committee, which was considering the House version of S. 2814, that the proposed language "authorized entities" would permit Comsat to compete with the carriers at the retail level. Again, Chairman Minow expressed the FCC's desire that the language "authorized entities" be deleted and that Comsat's role in the industry be defined expressly "as a communications common carriers' carrier." Communications Satellites: Hearings on H.R. 10115 and 10138 before the House Comm. on Interstate and Foreign Commerce, 87th Cong., 2d Sess. 408, 416-17 (1962). On April 11, 1962, Chairman Minow testified before the Senate Commerce Committee reiterating the FCC's interpretation that the "authorized entities" language gave it unwanted discretion to permit Comsat to compete with the carriers at the retail level. Clearly, Congress was aware that the FCC, the agency charged with the responsibility of choosing which entities would be entitled to authorized user status, would not consider Congress' predictions concerning Comsat's role in the industry as a limitation on its discretion to designate non-carriers as authorized users. Yet, in the bills' final version, Congress retained the "authorized entities" language. Congress' action in this regard is a strong indication that it did not intend to limit the FCC's discretion to designate authorized users and thus to allow Comsat to enter the retail market, by requiring the FCC to promote a two-tiered structure.

Finally, the legislative scheme of the Act also supports the conclusion that Congress' predictions concerning Comsat's role as a carriers' carrier were not intended as a limitation on the FCC's discretion in designating non-carriers as authorized users. The Act specifically permits the FCC to "grant appropriate authorizations for the construction and operation of each satellite terminal station (earth station), either to the corporation (Comsat) or to one or more authorized carriers or to the corporation and to one or more such carriers, jointly, as will best serve the public interest, convenience, and necessity." 47 U.S.C. Sec. 721(c)(7). The provision permits both Comsat and the IRCs to own earth stations and, thus, to provide independently earth station service. Entities that are permitted to independently own and operate earth stations can provide retail services; thus, the Act grants both Comsat and the IRCs the ability to offer retail services. Also, the Act encourages the FCC to maximize competition "in the provision of equipment and services utilized by the system" and to assure that Comsat is operated in such a fashion as to "maintain and strengthen competition in the provision of communications services to the public." 47 U.S.C. Sec. 701(c). Clearly, if Congress intended to foreclose competition between Comsat and the IRCs at the retail level, it would not have permitted the FCC to allow both Comsat and the IRCs to provide retail services and encouraged the FCC to promote maximum competition in the provision of satellite services. Thus, the scheme of the Act is further evidence that Congress did not intend to foreclose retail competition between Comsat and the IRCs.

We find that the plain language, the legislative history and the scheme of the Act indicate clearly that the FCC may designate non-carriers as authorized users, and that the FCC is not required to maintain a two-tiered structure in the industry. Congress' predictions concerning Comsat's role in the industry are insufficient to support petitioners' inference that Congress intended to limit the FCC's discretion to designate non-carriers as authorized users. In light of the FCC's repeated requests that Congress adopt language limiting expressly Comsat's ability to compete with the carriers at the retail level and Congress' refusal to adopt such language, we believe the more reasonable inference is that Congress' predictions were not intended as limitations on the FCC's discretion. Moreover, a review of the Act reveals that both Comsat and the common carriers may offer a retail service and that one of the policies of the Act is to promote maximum competition in the industry when such competition furthers the public interest. We believe that the scheme of the Act indicates that Congress did not intend to restrict the FCC's ability to promote retail competition between Comsat and the common carriers. Accordingly, we hold that the Satellite Act permits the FCC to designate non-carriers as authorized users so long as such status furthers the public interest, and therefore, that the Act affords the FCC the discretion to allow Comsat to compete at the retail level.

IV. RELEVANT FACTORS

Petitioners argue that the FCC's decision in Authorized User II is arbitrary because the FCC permits Comsat to enter the retail markets without considering adequately a number of relevant factors. First, petitioners claim that the FCC has failed to consider adequately a number of anticompetitive effects which will result from Comsat's entrance into the retail markets. In response, the FCC asserts that it has considered fully the potential anticompetitive effects that might result from its change in policy.

Petitioners' second argument concerns two issues which the FCC has deferred: direct access and independent earth station ownership. Because resolution of these issues may exacerbate or minimize the predicted anticompetitive effects, petitioners believe that these issues are germane to the FCC's Authorized User II policy. Consequently, the petitioners conclude that the FCC acted arbitrarily by failing to consider the direct access and independent earth station ownership issues prior to issuing its decision. The FCC, however, believes that while the direct access and independent earth station ownership issues may have merit, these issues are not "inextricably tied" to its Authorized User II policy. Authorized User II, 90 FCC 2d 1394, 1433-34 (1982). According to the FCC, the complexity of the direct access and independent earth station ownership issues and the lack of adequate information justifies its decision to defer these issues.

A. Anticompetitive effects. Petitioners claim that the FCC failed to analyze fully the anticompetitive effects of the FCC's change in authorized user policy. Specifically, they argue that the FCC's new policy will result in a substantial diversion of leased-channel business from the IRCs to Comsat; the diversion of revenues will result from the IRCs' inability to compete with Comsat for leased-channel business. The IRCs assert that the leased-channel and exchange service markets are interrelated, and that a substantial loss in leased-channel business will affect adversely exchange service prices and, thus, the public interest. The FCC does not contest the IRCs' reasoning that a substantial diversion of leased-channel business will affect adversely the public interest; rather, it challenges the IRCs' premise that a substantial diversion of leased-channel business is likely to result from its change in policy. Thus, the preliminary dispute is whether the FCC's Authorized User II policy will result in a substantial diversion of leased-channel business from the IRCs to Comsat because of the IRCs' inability to compete effectively on the basis of price.

Under the FCC's Authorized User II policy, competitors for leased-channel business will compete primarily on the basis of price. Leased-channel customers are sophisticated and able to arrange for their own link-up facilities. Leased-channel consumers do not depend on the IRCs for link-up service. We believe that the IRCs' ability to provide link-up service efficiently will not affect substantially the IRCs' competitive position in the leased-channel market. Moreover, leased-channel customers generally do not perceive any real distinction between cable and satellite service. Indeed, the FCC's use of prescribed use formulas presumes that satellite and cable circuits are, for the most part, interchangeable. Moreover, Comsat's basic transmission service and the satellite leased-channel service offered by the IRCs are virtually identical. In general, competition for leased-channel business does not depend upon a competitor's ability to provide link-up service, upon a competitor's ability to provide different services, or upon a competitor's ability to operate earth stations efficiently. We believe that the leased-channel customers are likely to distinguish between leased-channel competitors based on the competitor's price per circuit. Consequently, even small variations in price between competitors' satellite circuits or between competitors' cable and satellite circuits are likely to result in a diversion of leased-channel business.

In light of the leased-channel market's high sensitivity to price, intramodal competition for leased-channel business does not appear to be feasible under the FCC's Authorized User II policy. The FCC's Authorized User II policy would permit non-carriers to deal directly with Comsat for basic transmission service and would require Comsat to charge both carriers and non-carriers the same bundled rate for basic transmission service; thus, non-carriers and carriers can purchase satellite service at the same price. Authorized User II, 90 FCC 2d 1394, 1429 (1982). Non-carriers are unlikely to lease basic transmission from the IRCs if basic transmission service is available directly from Comsat at the same rate. Unless the IRCs are willing to offer basic transmission service to non-carriers at a rate less than cost, leased-channel customers will be encouraged to bypass the IRCs and lease basic transmission service directly from Comsat. In short, we find that the FCC's Authorized User II policy is likely to frustrate meaningful intramodal competition for leased-channel business between Comsat and the IRCs.

Because intramodal competition is unlikely, vigorous competition between the IRCs and Comsat for leased-channel business depends upon the feasibility of intermodal competition. In 1966, intermodal competition was not feasible because satellite circuits were substantially less expensive than cable circuits. In its Authorized User II decision, the FCC found that "technological advances" have made cable "much more cost competitive than it believed possible." Authorized User II, 90 FCC 2d 1394, 1414 (1982). To support this proposition, the FCC cites one example which indicates that the per circuit cost of cable voice-grade circuits is approximately $46,000 and notes that initial fiber optics projections indicate that the future cost for cable will be approximately $15,000 to $19,000 per circuit. The present cost of a satellite voice-grade circuit is $13,500. Based on this information, the FCC found that cable leased-channel will be competitive with satellite leased-channel at some time in the future. Id. at 1414 n. 21. Whether intermodal competition is presently feasible, however, is a separate question.

Although cable is more competitive than it was in 1966, the FCC implicitly recognizes that cable remains a more expensive means of transmitting telecommunications messages. In its Authorized User II decision, the FCC states that "the rapidly diminishing gap between cable and satellite facilities cost should significantly aid the IRCs in pricing their leased-channel services competitively", id. at 1416, and that "cable should be competitive" when fiber optics is implemented by the carriers. Id. at 1414 n. 21 (emphasis added). These statements indicate that a gap between cable and satellite costs still exists, and that cable is not yet competitive with satellite. In light of the leased-channel market's high sensitivity to price, we believe that the IRCs' prediction that a substantial diversion of leased-channel business to Comsat will result from Comsat's entrance into the retail leased-channel market is based on more than speculation. Because we believe that a risk exists that Comsat's entrance into the retail leased-channel market will result in a substantial diversion of leased-channel business from the IRCs to Comsat, the critical question is what effect such a diversion will have on the public interest.

In Authorized User I, 4 FCC 2d 421, 432 (1966), the FCC's primary concern was that a substantial loss of leased-channel business by the IRCs would result in an increase in the IRCs' rates for exchange services. The FCC reasoned that a substantial loss of leased-channel revenues would substantially decrease the IRCs' operating revenues and, thus, force the IRCs to increase their exchange service rates. Id. Because an increase in exchange service rates would affect adversely many exchange service customers, the FCC concluded that permitting Comsat to compete with the IRCs at the retail level would be inconsistent with the Satellite Act's primary policy of extending the economies of satellite technology to all users and, thus, assumed that an increase in exchange service rates was not in the public interest. In Authorized User II, the FCC did not abandon its assumption; instead, it based its change in policy on a number of changed circumstances. Authorized User II, 90 FCC 2d 1394, 1415 (1982). First, the FCC explained that "the IRC's operations are, at least for the most part, profitable and that the IRCs are experiencing rapid growth." Id. Second, the FCC indicated that the IRCs' rates of return in 1976 for leased-channel business ranged from between -3.3 percent to 5.6 percent. In contrast, the IRCs' rate of return in 1976 for telex, which accounted for 62 percent of the IRCs' total revenues, ranged from between 31.7 percent to 58.3 percent. Id. at 1413. Based on the IRCs' overall profitability and the relative unimportance of leased-channel business to the IRCs' overall revenues, the FCC found that any diversion of leased-channel revenue from the IRCs to Comsat would not "lessen the IRCs' ability to provide service to the public or . . . otherwise impair the quality or availability of public service." Id. at 1419 (emphasis added). Although we accept the FCC's conclusion that a loss of leased-channel revenues by the IRCs will not threaten the IRCs' continued existence or impair their ability to provide quality exchange services, the question remains whether the FCC's Authorized User II policy will result in an increase in exchange service rates to the detriment of the many exchange service customers.

The FCC's Authorized User II policy could affect adversely exchange service rates and, thus, the public interest in a number of ways. For example, to the extent that a loss of leased-channel business lessens the volume of messages sent by cable, the marginal cost for all cable services will increase. Moreover, if leased-channel business is lost, the IRCs' exchange customers must pay a relatively higher proportion of the IRCs' administrative overhead. Further, the FCC's decision to eliminate composite rate formulas and, at least in part, prescribed use formulas, might also affect adversely the IRCs' exchange service prices. Presumably, the FCC's composite rate and prescribed use requirements have kept exchange service prices artificially low; in effect, cable customers have been subsidized by satellite customers. The elimination of composite rate formulas will reflect the true costs of satellite and cable. Cable remains a more expensive medium than satellite; accordingly, the elimination of composite rate formulas will reflect the generally higher cost of cable. In summary, we believe that the cost of cable exchange services in the short term is likely to increase as a result of the FCC's Authorized User II policy.

An increase in the IRCs' cable exchange service rates is unimportant unless the increase affects adversely the public interest. See Hawaiian Tel. Co. v. FCC, 498 F.2d 771 (D.C. Cir. 1974). In Authorized User I, 4 FCC 2d 421, 432-33 (1966), the FCC concluded that an increase in the IRCs' cable exchange service rates would affect adversely the public interest. In 1966, however, the common carriers were the only entities permitted to offer retail exchange services. Further, the carriers had a clear incentive to prefer cable over satellite transmission. Accordingly, any increase in the carriers' cable exchange service prices necessarily would have been borne by the exchange service customers. In Authorized User II, 90 FCC 2d 1394, 1435 (1982), the FCC authorized Comsat to provide retail satellite exchange service through a separate subsidiary. Clearly, the availability of retail satellite exchange service could minimize any adverse effects which may result from an increase in the IRCs' cable exchange service rates. Therefore, the question whether exchange service customers will be forced to pay higher exchange service prices as a consequence of the FCC's Authorized User II policy, depends upon the IRCs' willingness to offer satellite exchange services or upon Comsat's successful entrance into the retail exchange service market.

If the FCC's Authorized User II policy results in an increase in cable exchange service rates, the IRCs will have no incentive to offer satellite exchange services. Carriers are only entitled to earn a rate of return on capital invested reasonably. The cable network represents the IRCs' primary capital investment; thus, the IRCs have a clear incentive to prefer cable over satellite transmission. Absent FCC intervention, we doubt that the IRCs would sacrifice profits by offering satellite exchange services as an alternative to their higher cable exchange service rates. Thus, the availability of retail satellite exchange service under the FCC's Authorized User II policy depends primarily on Comsat's successful entrance into the retail exchange service market.

Throughout these proceedings, Comsat has maintained steadfastly that it has no intention of establishing a subsidiary to compete for exchange service business. If Comsat does not enter the retail exchange service market, exchange service customers must depend necessarily on the IRCs for exchange service. Further, even if Comsat does enter the exchange service market, there are no guarantees that it will be able to compete effectively with the carriers for exchange service business. Unlike the leased-channel market, competitors in the exchange service market must be efficient in arranging foreign communication links; thus, competition is based on service as well as price. Over the years, the IRCs have established working relationships with a number of foreign communications companies. These relationships give the IRCs a distinct advantage over new entrants in the market. See ITT World Communications, Inc. v. FCC, 595 F.2d 897 (2d Cir. 1979). Although we do not question Comsat's ability, the end-to-end market presents more obstacles to new entrants than the leased-channel market. In summary, Comsat has indicated its clear intent not to enter the retail exchange service market. Moreover, the retail exchange service market is more difficult to enter than the leased-channel market. In light of these factors, the prospect of Comsat successfully entering the retail exchange service market in the near future is highly unlikely.

In conclusion, we believe that the FCC's Authorized User II policy is likely to result in an increase in cable exchange service rates. Whether the increase in cable exchange services rates will affect adversely the public interest depends upon either the IRCs' willingness to offer satellite exchange services or upon Comsat's successful entrance in the exchange service market. The IRCs have a monetary incentive to prefer cable over satellite and Comsat has maintained steadfastly that it has no intention of entering the exchange service market. Accordingly, we believe that any increase in cable exchange service rates will be borne by the many exchange service customers and, thus, that the FCC's Authorized User II policy will have some adverse effects on the public interest.

The fact that the FCC's Authorized User II policy may have some adverse effects on the public interest does not mean that the FCC's new policy of heightening intermodal competition by deregulating the industry is necessarily inconsistent with the overall public interest. In making public interest determinations which will affect an entire industry, the FCC is responsible for weighing the potential benefits against the detriments of a proposed policy. In this instance, the FCC has enumerated a number of potential benefits which may result from its change in policy. Authorized User II, 90 FCC 2d 1394, 1419-1422 (1982). Further, the FCC has considered the likelihood that its new policy will have some adverse effects on the public interest. Although we disagree with the FCC's assessment of the potential detrimental effects which may result as a consequence of the FCC's Authorized User II policy, we do not believe that the FCC's failure to afford sufficient weight to the likely detrimental effects of its new policy is, in itself, a sufficient basis from which this Court can conclude the FCC's Authorized User II policy is necessarily inconsistent with the public interest. In this case, however, the petitioners not only claim that the FCC failed to consider adequately the detrimental effects of its change in policy, but also, that the FCC failed to consider issues which may exacerbate or minimize these expected adverse effects. Specifically, petitioners argue that the FCC has abused its discretion by failing to consider the direct access and independent earth station ownership issues prior to issuing its Authorized User II decision.

B. Direct Access and Independent Earth Station Ownership. The FCC has determined that no direct connection exists between its policy decision to foster intermodal competition in the leased-channel and exchange markets and the direct access and independent earth station ownership issues. Consequently, the FCC announced its change in authorized user policy and initiated separate proceedings to consider the direct access and independent earth station ownership issues. The petitioners argue, however, that the direct access and independent earth station ownership issues are germane to the FCC's Authorized User II policy and, thus, that the FCC abused its discretion by failing to consider these relevant factors. In our view, direct access and independent earth station ownership by the IRCs would affect directly the FCC's policy of promoting intermodal competition. Because of the potential impact that these issues could have on the FCC's policy of promoting intermodal competition, we believe the FCC abused its discretion by restructuring the entire telecommunications industry prior to considering the merits of direct access and independent earth station ownership.

All of the parties agree that direct access and independent earth station ownership by the IRCs will increase the likelihood of intramodal competition for leased-channel business. Presumably, if the IRCs are permitted direct access, they will be able to lease space segment capacity from Intelsat at a rate less than Comsat's present tariff. Clearly, a decrease in the IRCs' cost for the space segment would enhance the IRCs' leased-channel market position and, thus, would increase the potential for intramodal competition for leased-channel business between Comsat and the IRCs. Further, if the IRCs are permitted to independently own and operate earth stations, a new basis for competition will exist: Comsat and the IRCs would compete on their respective abilities to provide efficient earth station service. Thus, the expected consequence of permitting the IRCs direct access and independent ownership and operation of earth stations is an increase in intramodal competition between Comsat and the IRCs for leased-channel business.

An increase in intramodal competition in the leased-channel market may be consistent in some respects, and inconsistent in other respects, with the FCC's primary policy of encouraging intermodal competition in both markets. Theoretically, maximum intermodal competition in both the exchange service and the leased-channel markets could be furthered by restricting the IRCs' future capital investments to cable and the development of fiber optics. A decision to deny the IRCs direct access and independent earth station ownership would have such an effect: more of the IRCs' investment capital would be available to promote the cable medium if the IRCs were foreclosed from investing in satellites. In contrast, if the FCC decides to permit the IRCs' to invest in the satellite system either by allowing them direct access or independent earth station ownership, the IRCs would have an incentive to earn a rate of return on their investments by channeling some of their business via satellite; the IRCs willingness to offer satellite exchange services may minimize the adverse consequences of an increase in cable exchange service rates. Clearly, the direct access and independent earth station ownership issues affect directly the IRCs' investment decisions and, thus, their ability to promote either cable or both cable and satellite.

The FCC's decision concerning direct access and independent earth station ownership is also likely to affect Comsat's decision whether to enter the retail exchange service market. The FCC has promised that it will continue to monitor traffic to assure that both the satellite and cable systems are used reasonably. Authorized User II, 90 FCC 2d 1394, 1426 (1982). Presumably, the FCC plans to limit any substantial diversions from one mode to another by using its prescribed use formula. Such an assurance is inconsistent with the FCC's policy of encouraging intermodal competition because it discourages Comsat from entering the retail exchange service market and because it forces the IRCs to buy from their proposed competitor. If the IRCs, however, are permitted direct access and independent earth station ownership, the IRCs would not be forced to buy from their competitor and Comsat may find it necessary to enter the exchange service market if it hopes to get any exchange service business.

Finally, independent earth station ownership may also affect the cost of satellite services. If the IRCs are permitted independent earth station ownership and operation and if competition results in the more efficient operation of earth stations, presumably the price for earth station access will decrease. The FCC has identified the potential decrease in consumer cost that may result from an increase in competition for the provision of earth station service as a justification for reviewing its present joint ownership policy. Notice of Inquiry, Modification of Policy on Ownership and Operation of U.S. Earth Stations that Operate with the Intelsat Global Communications Satellite System, 90 FCC 2d 1452 (1982). A dramatic decrease in price would certainly affect the feasibility of intermodal competition for leased-channel business and, to a lesser extent, exchange service business. Conversely, if increased competition leads to a short term increase in the cost of earth station operations and, thus, an increase in the price of satellite services, intermodal competition may be more feasible. In any event, independent earth station ownership may affect indirectly the likelihood of intermodal competition in both markets.

We have enumerated a few of the potential effects that direct access and independent earth station ownership may have on the FCC's policy decision to promote intermodal competition in both markets. The FCC itself has recognized that the resolution of these issues will have a widespread effect on the structure of the industry. In the past, we have permitted the FCC to resolve some issues and to defer the resolution of other issues when the issues decided were not inextricably related to the issues deferred. See Western Union Int., Inc. v. FCC, 673 F.2d 539, 543-44 (D.C. Cir. 1982). In this case, however, the FCC has attempted to restructure the entire industry on a piecemeal basis. It has firmly established a policy of promoting intermodal competition without considering issues which concern directly the likelihood of intermodal competition. Moreover, the FCC has chosen to promote intermodal competition without considering whether intramodal competition is consistent with its new policy.

In our view, an agency does not act rationally when it chooses and implements one policy and decides to consider the merits of a potentially inconsistent policy in the very near future. Of course, the FCC is the agency with the necessary expertise to analyze fully and determine the extent to which direct access and independent earth station ownership will affect intermodal competition in both markets. Perhaps intermodal competition in both markets is consistent with direct access and independent earth station ownership. Perhaps Comsat will change its mind and enter the retail exchange service market without further incentive. On the other hand, intermodal competition may be frustrated if the IRCs are permitted to invest a substantial amount of capital in direct access from Intelsat or in the construction and operation of earth stations. Clearly, the FCC is in a better position than this Court to resolve these important questions. See, e.g., Springfield Television of Utah, Inc. v. FCC, 710 F.2d 620, 626 (10th Cir. 1983); Stone v. FCC, 466 F.2d 316, 322 (D.C. Cir. 1972). Until the FCC has considered and determined the widespread effects that the direct access and independent earth station ownership issues will have on the industry structure, however, neither the FCC nor this court can determine rationally which policy will best serve the public interest. Consequently, we hold that the FCC has abused its discretion by implementing its Authorized User II policy prior to considering the direct access and independent earth station ownership issues.

V. REMEDY

Petitioners contend that the FCC has violated its duty to decide issues presented within a reasonable time, 5 U.S.C. Sec. 555(B) (1970), by unduly delaying the resolution of the direct access and independent earth station ownership issues. Consequently, petitioners have requested this court to compel the FCC to initiate rulemaking proceedings for the purpose of resolving these issues. The FCC argues that the direct access and independent earth station ownership issues are serious and complicated and that it has only recently collected information necessary to resolve these issues. In light of the FCC's broad discretion to manage its docket and, thus, establish its priorities, see, e.g., Nader v. FCC, 520 F.2d 182, 195-97 (D.C. Cir. 1975), it urges this court to decline petitioners' invitation to interfere with the FCC's docket.

Although this court has indicated "that nine years should be enough time for an agency to decide almost any issue," id. at 206, we believe that we may not properly establish and enforce a strict timetable for the FCC in this instance. Because of the extremely complicated nature of the direct access and earth station ownership issues, we believe that any prediction concerning the length of time necessary to resolve these issues would be speculative. Moreover, the FCC has indicated that it requires further information from petitioners and others before it can decide the merits of these issues. Thus, the FCC's ability to resolve these issues depends, in part, on the petitioners' prompt action. Finally, we have every reason to believe that the FCC will not unduly delay its decisions. Indeed, the FCC's change in policy manifests its desire to promote the public interest by deregulating the international telecommunications industry. We believe that our decision requiring the FCC to address the direct access and earth station ownership issues prior to implementing a change in its Authorized User I policy is sufficient to insure that the FCC will act expeditiously.

We have considered all of the arguments presented by the various parties. We hold that the Satellite Act of 1962, 47 U.S.C. Sec. 701 et seq., gives the FCC broad discretion to designate non-carriers as authorized users. In this instance, however, the FCC has abused its discretion by failing to consider adequately a number of relevant factors prior to implementing its Authorized User II policy. Accordingly, the FCC's Authorized User II decision is vacated and remanded to the FCC for proceedings not inconsistent with this opinion.