Midtec Paper Corp. v. United States

Opinion for the Court filed by Circuit Judge D.H. GINSBURG. D.H. GINSBURG, Circuit Judge: In this case we review a decision and order of the Interstate Commerce Commission in which the agency applied its newly53 adopted rules governing “competitive access” in the railroad industry. Midtec Paper Corporation and the Soo Line Railroad filed a joint complaint alleging that the only railroad with rail trackage directly serving Midtec’s facilities, the Chicago and North Western Transportation Company (C & NW), enjoyed a monopoly in the market for Midtec’s rail transportation needs and exploited its market power at Midtec’s expense. As a “captive shipper,” Midtec, supported by the Soo, requested that the Commission order the C & NW to give the Soo direct access to Midtec’s facilities by providing it with “reciprocal switching” service and with “terminal trackage” rights over the C & NW’s tracks. These remedies were necessary, they argued, to promote the public interest and the competition policies adopted by Congress in the Staggers Rail Act, Pub.L. No. 96-448, 94 Stat. 1895 (1980). The Commission was of a different view, and dismissed the complaint. We conclude that the agency’s decision is consistent with its statutory mandate and supported by substantial evidence. We therefore deny Midtec’s petition for review. I. Background Kimberly, Wisconsin, is in the Fox River Valley, a major papermaking area. Midtec acquired a paper mill there and invested nearly $130 million in it over a period of years. Like the other paper mills in the valley, Midtec relies substantially on rail service to receive raw materials and to ship paper products. Two major carriers, the C & NW and the Soo, serve the paper manufacturers in the Fox River Valley. The main lines of these two railroads intersect at Appleton, Wisconsin; they exchange traffic there at “interchange” facilities located approximately eight miles from Midtec’s mill. The mill itself is on the C & NW’s Kaukauna branch line, which connects it to the Appleton interchange. Thus, in order to share in Midtec’s traffic, any rail carrier other than the C & NW must enter into some cooperative arrangement with the C & NW. There are three general types of arrangement by which a railroad might be able to participate with the C & NW in the carriage of Midtec’s rail shipments. We describe the arrangements and the legal standards governing their imposition by the ICC before going on with the facts of this case. First, the C & NW and another railroad (or railroads) could agree to establish “through routes,” whereby Midtec’s traffic would be carried between its mill and the traffic’s origin or destination in part by each of the participating carriers. Under this arrangement, railcars carrying Midtec’s traffic would be transferred from one carrier to another at an interchange facility like that maintained by the Soo and the C & NW at Appleton. The railroads would quote a “joint rate” to Midtec, the revenues from which would be apportioned between them pursuant to an agreed “divisions” formula. If the carriers could not agree between themselves, the Commission could, if “it considers it desirable in the public interest,” 49 U.S.C. § 10705(a)(1) (1982),1 prescribe through routes, joint rates, and division formulae. Second, another railroad could operate its railcars over the C & NW’s Kaukauna branch line, and thus provide “single line” rail service to Midtec’s mill at Kimberly, i.e., by publishing rates to or from the point it does not serve over its own tracks. The Commission is authorized by section 223 of the Staggers Act to require a carrier to provide such joint use of terminal area facilities,2 “including main-line tracks for a reasonable distance outside of a terminal [to] another rail carrier if the Commission finds that use to be practicable and in the public interest without substantially impair — 54 — ing the ability of the rail carrier owning the facilities or entitled to use the facilities to handle its own business.” 49 U.S.C. § 11103(a) (1982). Again, if the carriers cannot agree on terms for so-called terminal trackage rights, the Commission may prescribe them. Historically, the Commission has required a party requesting terminal trackage rights to satisfy the “practicable and in the public interest” criteria of this section by demonstrating “some actual necessity or compelling reason” why such an arrangement should be ordered; this requires a showing of “more than a mere desire on the part of shippers or other interested parties for something that would be convenient or desirable to them.” Jamestown Chamber of Commerce v. Jamestown, W. & N.R. Co., 195 I.C.C. 289, 291 (1933). Finally, a carrier could obtain “reciprocal switching” service, under which the C & NW would, for a fee, transport that carrier’s railcars between Midtec’s mill and an interchange facility, and the carrier would publish single line rates to and from the mill. Pursuant to section 223 of the Staggers Act, the Commission may require rail carriers to enter into such agreements, where it finds them to be “practicable and in the public interest, or where such agreements are necessary to provide competitive rail service.” 49 U.S.C. § 11103(c)(1) (1982). Again, if the carriers cannot agree on terms, the Commission may prescribe them. Initially, the Commission interpreted the “practicable and in the public interest” criteria of sections 11103(a) and 11103(c)(1), governing terminal trackage rights and reciprocal switching respectively, to be substantially identical. See Delaware & H. Ry. Co. v. Consolidated R. Co., 367 I.C.C. 718, 720-21 (1983) (Delaware & Hudson).3 With respect to reciprocal switching, the Commission initially interpreted the alternative criterion, that such an arrangement be “necessary to provide competitive rail service,” to require “a narrowf] (intramodal) focus ... [on] providing ‘competitive rail service’____” Id. at 728. It discounted evidence concerning the reasonableness of the rates charged by the carrier over which reciprocal switching was requested because, in the Commission’s view, such evidence “cannot determine the need for additional rail competition” to hold down rail rates. It explained that rail carriers have been given a great deal of flexibility to adjust their rates under the Staggers Act. We are convinced that Congress’s aim in creating section 11103(c) was to provide a competitive counterbalance to this broadened rate freedom. Id. at 729. We return now to the facts of the case before us. In 1981, when the market for paper products was in a slump, Midtec approached the C & NW, described its transportation needs and the market, and “asked for help” from the railroad. When it did not receive a satisfactory response from the C & NW, Midtec approached the Soo, and together they urged the C & NW to negotiate with the Soo and to provide it with “operating rights” over the C & NW's Kaukaüna branch line. The Soo warned the C & NW “that if the requests ... are rejected the Soo and Midtec will jointly petition the Interstate Commerce Commission for direct authority to serve Midtec pursuant to the provisions of § 11103(a) [governing terminal trackage rights] and, alternatively, [of] § 11103(c) [governing reciprocal switching]____” The C & NW, which found the proposal “most unusual” and “unacceptable,” refused to enter into negotiations. Midtec and the Soo then made good on their threat by filing a joint complaint with the ICC. — 55 — A. Midtec I Before the Commission, Midtec and the Soo argued that because it has access to rail service only via the C & NW’s Kaukauna branch line, Midtec is a “captive shipper,” i.e., that it is deprived of the benefits of competitive rail service; and therefore disadvantaged relative to competing paper mills in the region, which are directly served by more than one rail carrier. In particular, the complaint alleged that “Midtec’s ‘captive status’ on the [C & NW] subjects it to serious service and price disabilities.”4 It requested that the Commission order the C & NW to (1) make both reciprocal switching service and terminal trackage rights available to the Soo for Midtec traffic, (2) submit within 90 days an agreement setting forth the conditions and compensation governing these arrangements, and (3) in the event no agreement could be reached, that the Commission fix the terms upon which the reciprocal switching and terminal trackage rights could be exercised. After a hearing on the joint complaint, the Administrative Law Judge (ALJ) ordered the C & NW to enter into agreements with the Soo for both terminal track-age rights and reciprocal switching. Applying the standards established by the Commission in Delaware & Hudson, the AU found that such an order was practicable and in the public interest. The AU concluded that, in light of the pro-competitive policy of the Staggers Act, the “public interest” standard no longer requires “too great an emphasis on the harm to individual carriers.” Midtec Paper Corp. v. Chicago & N.W. Transp. Co., No. 39021, ALJ Decision at 10 (served Apr. 29, 1983). The AU further concluded that the C & NW’s service to Midtec had been inadequate and that a reciprocal switching arrangement is necessary to provide Midtec with competitive rail service. Specifically, he found that “[e]experience has shown that [the C & NW] is responsive to Midtec’s transportation problems (basically the need for lower rates to compete effectively) only when pushed and then somewhat reluctantly.” Further, in his view, “[a] finding that shippers only have direct access to a single rail carrier is evidence of a lack of competition.” Id. at 13. The AU thus determined that the evidence presented by the complainants satisfied the criterion of 11103(a) for terminal trackage rights and both of the disjunctive criteria of section 11103(c)(1) for reciprocal switching. The ICC Review Board affirmed the AU’s decision to require reciprocal switching, but in view of that relief, decided that “requiring the joint use of terminal facilities is not necessary.” Since it is less of an intrusion on a railroad’s property and operations, the Board noted, “[reciprocal switching will provide Midtec with the potential for competitive rail service with the least amount of government interference.” Midtec Paper Corp. v. Chicago & N.W. Transp. Co., 39021, Review Board Decision at 2-3 (served Sep. 9, 1983). The Commission, in turn, reversed the Review Board, concluding that neither reciprocal switching nor terminal trackage rights is appropriate. In the Commission's view, the record did not support the premise that the C & NW’s service to Midtec was inadequate: [T]he sole basis of [Midtec’s] allegations of inadequacy of service by the rail carrier which serves its facility is the alleged unreasonableness of the rates [it] must pay for rail transportation. In these circumstances, the Commission must give consideration to whether the serving carrier is market dominant,3 as market dominance is a prerequisite to a Commission finding of rate unreasonableness____Midtec has failed to make that showing in this proceeding. — 56 — Midtec Paper Corp. v. Chicago & N.W. Transp. Co., 1 I.C.C.2d 362, 364 (1985) (Midtec I). The Commission further concluded that, in order to satisfy the identical public interest standards applicable to requests for terminal trackage and for reciprocal switching, a complaining shipper must show that existing rail service is “so deficient that it seriously impairs the shipper's ability to compete.” Id. at 365. Finally, the Commission held that reciprocal switching was not “necessary to provide competitive rail service” to Midtec because “there is substantial evidence of intermodal ... and geographic competition sufficient to induce [the C & NW to provide the] rates and services that Midtec claims are needed____” Id. at 365-66. Midtec and the Soo sought review of the Commission’s decision in this court. While their petition for review was still pending, however, the Commission adopted rules governing the prescription of “competitive access” remedies, including reciprocal switching, through routes, and joint rates. Ex Parte 445, Intramodal Rail Competition, 1 I.C.C.2d 822 (1985) (codified at 49 C.F.R. Part 1144 (1986)). At the request of the Commission, we then remanded Midtec I to the agency for reconsideration in light of the newly-adopted competitive access rules (hereinafter “CARs”). B. Midtec II On remand, the Commission reopened the record in order to receive evidence and argument, especially regarding the “proper role of intramodal, intermodal, and geographic competition in light of [the] standards [of the CARs].” Although the CARs do not specifically cover terminal trackage rights, thejCommission did not think it could resolve a request for that type of relief apart from the considerations governing Midtec’s request for reciprocal switching. Midtec Paper Corp. v. Chicago and N. W. Transp. Co., No. 39021 Decision at 3 (served July 24, 1986). After the parties submitted additional evidence and briefs, the Commission again dismissed the joint complaint. The Commission concluded that, under the CARs, the key issue in this case is whether [the C & NW] has engaged in or is likely to engage in conduct that is contrary to the rail transportation policy or is otherwise anticompetitive. The essential questions here are: (1) whether the railroad has used its market power to extract unreasonable terms on through movements; or (2) whether because of its monopoly position it has shown a disregard for the shipper’s needs by rendering inadequate service. Midtec Paper Corp. v. Chicago & N.W. Transp. Co., No. 39021, Commission Decision at 10 (served Dec. 15, 1986) (hereinafter Midtec II). The Commission therefore “examined the record on all commodities to determine whether [the C & NW] engaged in abusive rates or practices.” Id. at 10 n. 21. The Commission noted that the C & NW had cooperated with the Soo in establishing through routes and joint rates for the movement of some commodities to Midtec’s Kimberly mill. It therefore construed the complaint of Midtec and the Soo as “not alleging] that C & NW has refused to grant access; rather they object to the terms under which it has or will be granted.” Accordingly, the Commission found it “appropriate to focus on the terms under which through service has been offered and the quality of service that [the C & NW] has provided.” Id. So focused, the Commission concluded that “the record is almost completely devoid of evidence concerning the terms under which through transportation was offered [by the C & NW] either before or after the institution of this complaint.” That deficiency was fatal to the cases both for terminal track-age rights and for reciprocal switching. Id. at 11-12. The Commission concluded by noting, however, that [t]here remains the possibility that complainants have been unable to document anticompetitive conduct as required by our rules due to the posture of this case. Since the original complaint was brought in December of 1982, the [C & NW] has — 57 — had an opportunity to “clean up its act.” Although it has not been shown that [the C & NW’s] rates and practices prior to that time were unlawful, it has made what all parties agree are important concessions since that time. We repeat that this Commission stands ready to grant relief on an expedited basis if necessary to remedy anticompetitive conduct by this railroad in the future. Id. at 12. Midtec alone has filed the petition now before us for review of this decision.5 II. Preliminary Issues The parties on both sides have raised procedural issues relating to the Commission’s decision in Midtec I and its application of the CARs. We will briefly discuss these issues before focusing our attention on the merits of the Commission’s decision. A. Is Any Portion of the Commission’s Decision in Midtec I Under Review in this Proceeding? Intervenors and amici in support of the petitioner argue that portions of the Commission’s decision in Midtec I are before the court in this proceeding. Specifically, they contend that the Commission did not in Midtec II apply the statutory criteria for reciprocal switching — “practicable and in the public interest, or ... necessary to provide competitive rail service” — but rather reconsidered Midtec I only under the criteria of the CARs. As a result, they argue that the Commission’s initial consideration of the joint complaint under the statutory standards in Midtec I is, so to speak, the “law of the case,” and therefore subject to review in this proceeding. Thus, for example, they would have us review the Commission’s initial position that competitive access is warranted only where existing service is “so deficient that [it] seriously impairs the shipper’s ability to compete.” Although the Commission did not reach a different result in Midtec II, it is clear that it applied a different standard. As a practical matter, therefore, the Commission is surely correct in arguing before this court that its later decision “superseded its previous decision and is the one before the Court for review.” On its face, Midtec II applied the criteria in the CARs, which were promulgated after the decision in Midtec I and formed the basis for the Commission’s sua sponte request for the remand that resulted in reconsideration and the decision in Midtec II. While the Commission’s interpretation and application of the CARs and the statutory criteria underlying them are open to challenge in this proceeding, no purpose would be served by our reviewing the decision in Midtec I, and we decline to do so. B. Is Midtec Barred From Challenging the Commission’s Application of the CARs By Reason of Collateral Estoppel or of Jurisdictional Time Limit? The Commission argues that Midtec's challenge to its decision is but a “thinly disguised ... collateral attack” on the CARs. Because Midtec participated in the rule-making proceedings in which the CARs were adopted, and because this court upheld the CARs in Baltimore Gas and Elec. v. United States, 817 F.2d 108 (D.C.Cir.1987), the Commission also contends that the company is now collaterally es-topped from challenging the CARs. The amici in support of the Commission go one step further; they suggest that this court lacks jurisdiction to consider Midtec’s challenge because the company failed to seek judicial review of the CARs within the 60-day statutory time period for doing so. See 28 U.S.C. §§ 2321, 2342, 2344 (1982). — 58 — The latter argument need not detain us. Midtec does not attack the CARs on their face, but only the manner in which the Commission interpreted and applied them in this case. In adopting the CARs, the Commission left open the many questions sure to arise when the CARs were applied to particular requests for relief such as Midtec sought. Thus, the Commission stated that “[t]o the extent that the[ ] criteria [of the CARs] are relevant to reciprocal switching requests, they will be applied, as will the statutory standards.” Intramodal Rail Competition, 1 I.C.C.2d 822, 830 (1985). Similarly, the CARs do not by their terms govern requests for terminal trackage rights. The Commission has now applied the CARs to particular facts, and extended their application to terminal trackage rights, and the petition for review questions whether it did either consistently with the statutory mandate the rules purport to implement. See RCA Global Communications, Inc. v. FCC, 758 F.2d 722, 731 (D.C.Cir.1985); MCI Telecommunications, Inc. v. FCC, 765 F.2d 1186, 1191 (D.C.Cir.1985). We therefore conclude that passage of the statutory time limitation on review of the CARs does not affect the court’s jurisdiction to hear Midtec’s challenge to the decision presently under review. Nor is review of the issues raised by petitioner precluded by either the Commission’s decision to adopt the CARs or this court’s approval of those rules in Baltimore Gas. Collateral estoppel applies to issues of law and fact “actually litigated and determined by a valid and final judgment, ... the determination [of which] is essential to [that] judgment.” Hastings v. Judicial Conference of the United States, 829 F.2d 91, 98 (D.C.Cir.1987) (quoting Restatement (Second) of Judgments § 27 (1982)); see Montana v. United States, 440 U.S. 147, 153, 99 S.Ct. 970, 973, 59 L.Ed.2d 210 (1979). In Baltimore Gas, we expressly noted that “it is not yet apparent how the ICC will determine whether [particular conduct] is ‘anticompetitive’ on the merits —i.e., what weight will be placed on the various factors.” 817 F.2d at 118. The CARs are certainly no clearer in defining the precise circumstances in which reciprocal switching or terminal trackage rights will be prescribed. Thus, the meaning of the CARs as applied to these forms of competitive access has never been “conclusively determined” by the Commission, and their application to Midtec’s situation has not been reviewed before by the courts. Accordingly, Midtec is not now barred from pressing its arguments on the merits. III. The Commission’s Decision on the Merits • Midtec makes the quite extraordinary claim that, in reviewing the Commission's decision on the merits, this court should extend no deference to the ICC as an expert body, solely because that decision was made in “an adjudicative, not a rulemaking proceeding.” “In such cases,” we are told, “the ICC must proceed with a high degree of precision and clearly address all legal and factual issues without making conclusory statements about competition or the legal standard governing the complaint.” We take this argument to entail two distinct issues: (1) the standard under which this court reviews the agency's interpretation of the CARs and the relevant statutory provisions, and (2) the degree to which we scrutinize (a) the evidence presented to the Commission and the agency's findings based on that evidence, and (b) the conclusions that the agency draws from those findings. With respect to |the standard of review applicable to an agency’s interpretation of the statutes it administers, the distinction that Midtec draws between adjudication and rulemaking is without significance. Our inquiry is the same in each procedural context. We first examine the text of the implicated statute and, where appropriate, its legislative history; using the traditional tools of statutory construction, we seek to determine whether and how Congress resolved the specific issues of law raised in the proceeding under review, and confine the agency to consistency with Congress’s intent. If we find that Congress did not clearly resolve those is — 59 — sues, however, we must accept the agency’s interpretation so long as it is reasonable — i.e., “rational and consistent with the statute.” NLRB v. United Food & Commercial Workers Union, Local 23, 484 U.S. 112, 108 S.Ct. 413, 421, 98 L.Ed.2d 429 (1987); see Chevron USA, Inc. v. Natural Resources Defense Council, 467 U.S. 837, 842-45, 104 S.Ct. 2778, 2781-82, 81 L.Ed.2d 694 (1984). Congress wisely entrusted the administration of the national rail transportation policy to the Commission, not to the courts; it relied upon the cumulative experience and expertise of that body, not upon the quite different strengths of the courts. In such circumstances, we must defer to the agency’s statutory interpretation in order to give effect to the will of the legislature. This “principle of deference to administrative interpretations,” as the Supreme Court explained in Chevron: “has been consistently followed by this Court whenever decision as to the meaning or reach of a statute has involved reconciling conflicting policies, and a full understanding of the force of the statutory policy in the given situation has depended upon more than ordinary knowledge respecting the matters subjected to agency regulations. See, e.g., National Broadcasting Co. v. United States, 319 U.S. 190 [65 S.Ct. 997, 87 L.Ed. 1344]; Labor Board v. Hearst Publications, Inc., 322 U.S. 111 [64 S.Ct. 851, 88 L.Ed. 1170]; Republic Aviation Corp. v. Labor Board, 324 U.S. 793 [65 S.Ct. 982, 89 L.Ed. 1372]; Securities & Exchange Comm’n v. Chenery Corp., 332 U.S. 194 [67 S.Ct. 1575, 91 L.Ed. 1995]; Labor Board v. Seven-Up Bottling Co., 344 U.S. 344 [73 S.Ct. 287, 97 L.Ed. 377]. “... If this choice represents a reasonable accommodation of conflicting policies that were committed to the agency’s care by the statute, we should not disturb it unless it appears from the statute or its legislative history that the accommodation is not one Congress would have sanctioned.” 467 U.S. at 844-45, 104 S.Ct. at 2782 (quoting United States v. Shimer, 367 U.S. 374, 382, 383, 81 S.Ct. 1554, 1560, 1561, 6 L.Ed.2d 908 (1961)) (footnote omitted). The cases cited by the Supreme Court in the quoted passage make clear that this approach applies equally to agency rule-making and to agency adjudication. This has been confirmed by the Court more recently when it explained that the meaning of ambiguous statutory terms — and, we may add, of ambiguous agency rules— can only be given concrete meaning through a process of case-by-case adjudication. In that process of filling “any gap left, implicitly or explicitly, by Congress,” the courts must respect the interpretation of the agency to which Congress has delegated the responsibility for administering the statutory program. INS v. Cardoza-Fonseca, 480 U.S. 421, 107 S.Ct. 1207, 1221-22, 94 L.Ed.2d 434 (1987) (quoting Chevron, 467 U.S. at 843, 104 S.Ct. at 2781). We will therefore uphold the Commission’s interpretation of its rules and of the statutes it administers if its interpretation is consistent with the intent of the Congress, or if the intent of the Congress cannot be discerned, then if the agency’s interpretation is “reasonable.” See Pittsburgh & L.E. R. Co., 796 F.2d at 1541 (applying “reasonableness” test of Chevron in reviewing ICC interpretation of the Staggers Act in an adjudication). There is a distinction between rule-making proceedings and formal “on the record” adjudications with respect to the standard of review to be applied to agency factfinding. Pursuant to the Administrative Procedure Act, courts are instructed always to “hold unlawful and set aside agency action, findings, and conclusions found to be ... arbitrary [or] capricious....” 5 U.S.C. § 706(2)(A). In reviewing an agency adjudication, however, as we do in this case, we are additionally instructed to set aside agency findings that are “unsupported by substantial evidence.” 5 U.S.C. § 706(2)(E); Central States Ent. v. ICC, 780 F.2d 664, 674 (7th Cir.1985). There is less to this distinction than might first appear. As then-Judge Scalia observed: — 60 — When the arbitrary and capricious standard is performing that function of assuring factual support, there is no substantive difference between what it requires and what would be required by the substantial evidence test, since it is impossible to conceive of a “nonarbitrary” factual judgment supported only by evidence that is not substantial in the APA sense — ie., not “ ‘enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion sought to be drawn ... is one of fact for the jury.’ ” Association of Data Processing Serv. Org. v. Bd. of Governors of the Fed. Res. Sys., 745 F.2d 677, 683-84 (D.C.Cir.1984) (emphasis in original) (quoting Illinois Central R.R. v. Norfolk & W. Ry., 385 U.S. 57, 66, 87 S.Ct. 255, 260, 17 L.Ed.2d 162 (1966) (in turn quoting NLRB v. Columbian Enameling & Stamping Co., 306 U.S. 292, 300, 59 S.Ct. 501, 505, 83 L.Ed. 660 (1939))).6 The distinction is not entirely without significance, however, for even where the “substantial evidence” test applies to agency factfinding, the arbitrary and capricious standard, which always applies as well, performs an additional function. It authorizes and obliges the reviewing court to guard against an agency’s drawing inferences that are “arbitrary” in relation to the facts found, no matter how substantial may be the support for those facts. See Data Processing, 745 F.2d at 683. The court therefore must engage in a “thorough, probing, in-depth review,” of the agency’s asserted basis for decision, ensuring that “the agency ... [has] examine[d] the relevant data and [has] articulatefd] a satisfactory explanation for its action including a ‘rational connection between the facts found and the choices made.’ ” Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2867, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962)). Deference is due the agency in this inquiry, for the reviewing court’s function is necessarily “narrow and a court is not to substitute its judgment for that of the agency.” Id. Thus, “[w]e will ... ‘uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned.’ ” Id. (quoting Bowman Transp., Inc. v. Arkansas-Best Freight Sys., 419 U.S. 281, 286, 95 S.Ct. 438, 442, 42 L.Ed.2d 447 (1974)). With these fundamental principles in mind, we turn at last to the merits of this case. Midtec argues that in dismissing its complaint the Commission acted arbitrarily, capriciously, and contrary to law because: (a) the CARs, as interpreted in this case, are inconsistent with the provisions of the Staggers Act governing reciprocal switching, and the Commission failed to consider the complaint under those statutory provisions; (b) the Commission applied the reciprocal switching standards of the CARs to Midtec's request for terminal trackage rights, even though the latter form of competitive access is not specifically covered by the CARs; and (c) the requirement of the CARs that Midtec demonstrate that reciprocal switching is “necessary to remedy or prevent an act that is contrary to the competition policies of 49 U.S.C. [§ ] 10101a or is otherwise anticompetitive” was interpreted with reference to irrelevant factors, such as rate reasonableness, and geographic and intermodal competition. It further argues that (d) there was substantial evidence of the C & NW’s anticompetitive behavior before the Commission, so that the agency’s dismissal of the joint complaint, even under the CARs, was arbitrary and capricious. A. The Relationship Between the Competitive Access Rules and the Staggers Act Criteria for Reciprocal Switching As we have seen, the Commission did not expressly address Midtec’s complaint under the criteria of section 11103(c)(1): “practicable and in the public interest or ... nec — 61 — essary to provide competitive rail service.” The petitioner and the intervenors supporting it argue, each in its own way, that the agency’s failure to do so requires a remand. Midtec argues that, notwithstanding the permissive language of section 11103(c)(1) (“The Commission may require rail carriers to enter into reciprocal switching agreements”), the Commission is required to order reciprocal switching whenever it is either practicable and in the public interest or necessary to provide effective rail competition. If Midtec is correct in this, then it seems it need not demonstrate that the C & NW engaged in conduct that is contrary to the competition policy of the Staggers Act or that is otherwise anti-competitive, which, as we have seen, is a threshold requirement under the CARs. The intervenors, on the other hand, acknowledge that the statute merely authorizes and does not require the Commission to prescribe reciprocal switching when the criteria are met, but contend that the Commission’s failure to address the criteria was an error. In their view, even if a complaining shipper cannot make out a case for reciprocal switching under the CARs because it cannot meet the threshold requirement of showing that the respondent railroad had committed or was likely to commit an act contrary to the competition policy of the Staggers Act or otherwise anti-competitive, the Commission must go on to consider whether to exercise its statutory discretion to grant relief. We rejected a nearly identical claim in Baltimore Gas. There, we upheld the Commission’s decision to “prescribe through routes and joint rates, and establish switching arrangements, only to remedy or prevent ‘anticompetitive’ acts,” 817 F.2d at 114 (emphasis in original), as a policy consistent with the Staggers Act. Id. at 115. Here, to the extent that the arguments of Midtec and the intervenors are not foreclosed by Baltimore Gas, we find them unconvincing. Midtec’s argument, that although the statute is cast in discretionary terms it does not authorize unfettered discretion in granting or denying the relief contemplated by the statute, was not addressed in Baltimore Gas. There is surely merit in the general premise underlying it. See Thompson v. Clifford, 408 F.2d 154, 158 (D.C.Cir.1968). Application vel non of that premise to the statute before us “depends on the context of the statute, and on whether it is fairly to be presumed that it was the intention of the legislature to confer a discretionary power or to impose an imperative duty.” ' Id. at 158 (quoting United States ex rel. Siegel v. Thoman, 156 U.S. 353, 359, 15 S.Ct. 378, 380, 39 L.Ed. 450 (1895)); see United States v. Rodgers, 461 U.S. 677, 706, 103 S.Ct. 2132, 2149, 76 L.Ed.2d 236 (1983) (although congressional use of the word “may” “usually implies some degree of discretion,” “[t]his common-sense principle of statutory construction ... can be defeated by indications of legislative intent to the contrary or by obvious inferences from the structure and purpose of the statute”); Union of Concerned Scientists v. Atomic Energy Comm’n, 499 F.2d 1069, 1078-79 (D.C.Cir.1974). We therefore look to the intent of Congress and the structure and purposes of section 11103, as well as to the broader purposes of the Staggers Act. We do not here face a question of first impression, however. Even before the Commission adopted the CARs, the Seventh Circuit stated, with specific reference to section 11103(c)(1): Prior to the passage of the Staggers Act, railroads often refrained from entering into reciprocal switching agreements. See S.Rep. No. 470, 96th Cong., 1st Sess. 41 (1979). The purpose of the Staggers Act was to encourage, under the appropriate circumstances, but not require, the Commission to approve railroad switching agreements. Central States Ent., 780 F.2d at 679. Our own review of the legislative history confirms this interpretation: the Commission is under no mandatory duty to prescribe reciprocal switching where it bélieves that doing so would be unwise as a matter of policy. This is not to say that the Commission may act on the basis of whim or caprice, nor to suggest that the Commission’s — 62 — decisions under section 11103(c) are unreviewable for want of any law to apply. See Heckler v. Chaney, 470 U.S. 821, 834-35, 105 S.Ct. 1649, 1657, 84 L.Ed.2d 714 (1985). In order to support its exercise of discretion, the agency must provide a reasoned analysis that is not manifestly contrary to the purposes of the legislation it administers. The Commission did just that when it adopted the CARs. These rules narrow the agency’s discretion under section 11103 by describing, for example, the circumstances in which it would not grant discretionary relief — where there is no reasonable fear of anticompetitive behavior. We could not say in Baltimore Gas, and cannot say now, that the Commission’s narrowing of its own discretion is manifestly inconsistent with the terms or the purposes of section 11103, or with the broader purposes of the Staggers Act; it is not “unreasonable,” therefore, under the standard of review mandated by Chevron. We repeat: Our task ... is only to determine whether the ICC has arrived at a reasonable accommodation of the conflicting policies set out in its governing statute____ The regulations take into account shippers’ interests in reasonable rates, see 49 U.S.C. §§ 10101a(l) and (6) (1982), by, for example, requiring the ICC to set aside through route cancellations that are “anticompetitive.” They also reflect the Staggers Act’s strong emphasis on preserving and enhancing competition, see 49 U.S.C. § 10101a(l), (4), and (5) (1982), and at the same time restrict the circumstances under which the ICC will exercise its power to require through routes and joint rates, thereby limiting federal regulatory control over the industry, see 49 U.S.C. § 10101a(2) (1982). Baltimore Gas, 817 F.2d at 115. The quoted passage was concerned specifically with the Commission’s power to prescribe through routes and joint rates, rather than its authority to prescribe reciprocal switching. Midtec argues that the distinction is critical, because unlike reciprocal switching agreements, the “efficiency of [through routes] depends on the willingness and ability of the participants to agree on the material terms, particularly the prices to be charged ultimate customers and a formula for dividing revenues or profits." As a result, Midtec maintains, “Congress took pains to circumscribe the agency’s previously substantial powers over rates including mandatory joint rates ... [whereas] [t]here is no such hostility expressed toward reciprocal switching/trackage rights.” We noted in Baltimore Gas that, unlike the provisions the Staggers Act governing Commission scrutiny of a carrier’s decision to cancel through routes and joint rates, which limited the ICC’s powers, section 11103(c)(1), “increased the ICC's regulatory power — by authorizing the agency to require railroads to enter into agreements to ‘switch’ other railroads’ cars to and from shippers located along each other’s lines____” 817 F.2d at 113 (emphasis in original). It does not follow, however, that Congress intended the Commission to prescribe reciprocal switching rights more freely than through routes. The Commission has long had authority to prescribe through routes and joint rates; the Staggers Act only circumscribed the Commission’s authority to suspend through route and joint rate cancellations. Compare 49 U.S.C. § 10705(a)(1) (1982) with 49 U.S.C. § 15(3) (1970). In contrast, “[t]he Commission’s authority to order ... switching arrangements had ... been unclear,” prior to the passage of section 11103(c)(1). Baltimore Gas, 817 F.2d at 113 (citing H.R.Rep. No. 1035, 96th Cong., 2d Sess. 67 (1980), U.S.Code Cong. & Admin.News 1980, pp. 3978, 4012). That the Congress increased the options available to aggrieved shippers and carriers and clarified the Commission’s authority to order additional forms of competitive access does not, without more, suggest that one form of relief should be granted more liberally than another, much less that one is mandatory while the other is discretionary. Indeed, the House Report speaks of “joint service” agreements and reciprocal switching in parallel terms: “The Committee intends for the Commission to permit and — 63 — encourage reciprocal switching as a way to encourage greater competition. Likewise joint service agreements should be encouraged in order to improve shipper service and efficiency.” H.R.Rep. No. 1035, supra at 67, U.S.Code Cong. & Admin.News 1980 at 4012 (emphases added). If Congress had intended any disparity in the Commission’s discretion to deny these remedies, moreover, it would almost certainly have been by making reciprocal switching less rather than more available. Where the parties fail to reach agreement on competitive access arrangements, the prescription of through routes provides all participating carriers more flexibility to respond to competitive conditions than does requiring reciprocal switching. Once a carrier is ordered to establish a through route, it has a common interest with the other participating carrier(s) to set joint rates applicable to the routes at levels responsive to the demand for the movement of particular commodities, and to avoid further regulatory intervention. Even if they cannot agree on joint rates and the division of revenues among themselves, and the Commission steps in, the agency may take account of the differing costs and transportation needs presented by various kinds of traffic in setting the joint rates and divisions. By contrast, the carrier favored by mandated reciprocal switching may have every incentive to avoid agreement on switching rates, and to petition the Commission to fix the compensation. Once the switching charge is fixed, the benefited carrier retains the ability to engage in demand-based ratemaking, adjusting its charges according to the dictates of the market for particular transportation services. Because that carrier publishes “single line” rates, the switching carrier has no effective voice in determining those rates, and is not entitled to a division of the proceeds. As the Commission observed in this case, the switching carrier thus “lose[s] the ability to price its portion of the through service in response to the varying demands for different commodities or movements.” Midtec II, at 6. In this sense, therefore, prescribing reciprocal switching may invite more, rather than less, reliance on the Commission’s regulatory powers to establish reasonable rates than would prescribing through routes. This result would seem to be in tension, if not outright conflict, with at least one of the competition policies of the Staggers Act: “it is the policy of the United States Government ... to allow, to the maximum extent possible, competition and the demand for services to establish reasonable rates for transportation by rail.” 49 U.S.C. § 10101a(l) (1982). Congress contemplated that the Commission would prescribe through routes that the agency deems to be “in the public interest” just as it contemplated that reciprocal switching would be available where “practicable and in the public interest” or necessary to provide effective rail competition. Under the CARs, requests for either type of competitive access are met at the threshold with the same limitation, viz. that the access requested be “necessary to remedy or prevent an act that is contrary to the competition policies of 49 U.S.C. [§ ] 10101a or is otherwise anticompetitive.” This is a reasonable accommodation of what we called in Baltimore Gas the “fifteen different and not entirely consistent goals” of the national rail transportation policy set forth in section 101(a) of the Staggers Act. 817 F.2d at 112 (footnote omitted). We see no basis in the text of the statute or in its legislative history for concluding that the Commission acted unreasonably in treating these alternative means of “competitive access” under the same rules. We therefore reject the arguments to the effect that the CARs are either “optional” or “extra-statutory” guidelines, or that the Commission must exercise its statutory discretion anew in each particular case rather than rely on the generic approach in the CARs. See Fook Hong Mak v. INS, 435 F.2d 728, 730-32 (2d Cir.1970). B. The Standards Governing Terminal Trackage Rights In the decision under review, the Commission explained its reason for extending the requirement of anticompetitive — 64 — behavior (actual or feared) to requests for terminal trackage rights, as follows: Although the [competitive access] rules ... do not specifically cover terminal trackage rights, the underlying public interest test in 11103(a) and (c) is the same and, given the relationship between the issues and the remedies, we believe that the public interest analysis should be similar. We declined to promulgate rules to govern applications for terminal trackage rights in part because “[t]he joint rates, through routes, and reciprocal switching mechanisms should be sufficient to provide shippers and carriers with ample competitive access where necessary,” because few such cases had been brought, and because the agreement (discussed and adopted in Intramodal Rail Competition ] did not cover this subject. We stated that requests for terminal trackage rights will be considered “on an individual case basis”____ In so doing, we think that a focus on anticompetitive conduct (or the imminent threat of it) by the carrier possessing the essential rail line is appropriate, but not necessarily exclusive. Midtec II, at 7 (quoting Intramodal Rail Competition, 1 I.C.C.2d at 835). Petitioner and its supporting intervenors argue that this decision was unreasonable and contrary to the Act. Intervenors also suggest that in adopting this requirement, the Commission made a radical departure from prior precedent governing terminal trackage rights, and hence should be required to explain itself with special care. The Commission gave Midtec’s request for terminal trackage rights short shrift, but its analysis was neither arbitrary nor manifestly inconsistent with congressional intent or Commission precedent; “the agency’s path may reasonably be discerned.” Motor Vehicles Mfrs. Ass’n v. State Farm Mutual Ins. Co., 463 U.S. at 43, 103 S.Ct. at 2867. First, like the provision governing reciprocal switching, section 11103(a) provides that the Commission “may require” terminal trackage rights where “practicable and in the public interest.” The logical inference that the Commission’s discretion should be exercised and constrained in like manner under both provisions is strongly supported by the Conference Report on the Staggers Act, which states: “The standard ‘practicable and in the public interest’ [governing reciprocal switching rights] is the same standard the Commission has applied in considering whether to order the joint use of terminal facilities.” H.R.Conf.Rep. No. 1430, 96th Cong., 2d Sess. 116 (1980), U.S.Code Cong. & Admin.News 1980, pp. 3978, 4148. Indeed, one would expect that since the use of a carrier’s terminal facilities by another carrier is clearly a more significant intrusion into the operations of the servient carrier, see S.Rep. No. 470, 96th Cong., 1st Sess. 42 (1979) (describing reciprocal switching as “more limited action” than terminal trackage rights), the Commission's discretion to order terminal trackage rights would, if anything, be more, not less, contrained than its discretion to order reciprocal switching. Second, as we have seen, the Commission has long required a complainant seeking terminal trackage rights to demonstrate “some actual necessity or compelling reason” for such relief. Jamestown Chamber of Commerce, supra. To require some showing of (actual or potential) anticompetitive behavior on the part of the carrier over which access is sought does not manifestly depart from this standard. Even if the meaning of “some actual necessity or compelling reason” has been narrowed to some degree as a result of the more specific requirements of the CARs — and that is not a certainty — we cannot say that such revision was “arbitrary or capricious.” See Brae Corp. v. United States, 740 F.2d 1023, 1038 (D.C.Cir.1984) (agency’s change in policy is “not to be reviewed under a heightened standard of scrutiny”) (citing Motor Vehicles Mfrs. Ass’n v. State Farm Mutual Ins. Co., 463 U.S. at 40-44, 103 S.Ct. at 2865-67). Finally, neither Midtec nor the intervenors has offered any argument as to why, in this case, terminal trackage rights should be ordered even if reciprocal switch — 65 — ing is not warranted. Perhaps there are cases in which trackage rights are better tailored to the attendant circumstances, but Midtec has made no argument of this type. It claims only that if the Soo obtained both reciprocal switching and trackage rights, then Midtec’s hand would be strengthened in bargaining with the C & NW. No doubt, but at least without more, this hardly seems relevant; competition policy is not a matter of regulators handicapping would-be competitors in order to create an evenly matched contest. Midtec’s request for terminal trackage rights, like its request for reciprocal switching, is predicated upon the theory that its “captive status” “subjects it to serious service and rate disabilities.” Midtec I, Complaint at 3, para. 8. Under these circumstances, we cannot say it is unreasonable for the Commission to require Midtec and the Soo to demonstrate that terminal trackage rights are necessary to remedy or to prevent an act on the part of the C & NW that is contrary to the competition policy of the Staggers Act or that is otherwise anticompetitive. We now explore the nature of that requirement. C. The Meaning of the Threshold Requirement Under the CARs We repeat, for convenience, the Commission’s statement of the issues it identified for determining whether relief under the CARs is appropriate. The Commission said: The essential questions here are: (1) whether the railroad has used its market power to extract unreasonable terms on through movements; or (2) whether because of its monopoly position it has shown a disregard for the shipper’s needs by rendering inadequate service. Midtec II, at 10 (footnote omitted). The Commission also stated that it was attentive to the possibility of classical categories of competitive abuse: foreclosure; refusal to deal; price squeeze; or any other recognizable forms of monopolization or predation. [It] also considered whether there was any evidence of abuses under the competitive standards of the Rail Transportation Policy, including inadequate service or excessive prices. Id. at 3. Finally, the agency noted that evidence of intramodal, intermodal, and geographic competition “can be helpful in determining whether anticompetitive conduct has occurred or is likely to occur.” Id. at 9. Midtec challenges the Commission’s interpretation of the CARs’ requirement that Midtec demonstrate that the C & NW has acted or is likely to act anticompetitively or contrary to the competition policies of the Act on three grounds: (1) the Commission improperly required the complainants to adduce evidence regarding the “reasonableness” of the rates offered to Midtec by the C & NW; (2) the Commission’s focus on service inadequacies was irrelevant to the need for competitive access; and (3) under section 11103, the Commission was required to consider only intramodal competition — i.e., rail-to-rail competition — and to disregard intermodal and geographic competition. 1. Rate Reasonableness and “Market Dominance” In addressing the “essential questions” identified above, the Commission made the following observations and conclusions: [The CARs] emphasize several categories of specific information for evaluating the types of allegations made here as they relate to a possible abuse of market power. These categories include the revenues of the involved railroads, the comparative efficiency of routings, the comparative cost/revenue ratios for the carriers, and the rates sought to be charged. Despite the fact this proceeding was reopened specifically for consideration under the [CARs], complainants, on whom the burden of proof rests, have not submitted any of the specific evidence called for. There are no numbers in the record to give substance to complainants’ allegations. Midtec II, at 11. Because of the want of evidence regarding the reasonableness of the C & NW’s ratemaking practices, the — 66 — Commission did not credit Midtec’s “general allegations” that the C & NW refused to offer “competitively based rates.” Id. at 10-11. Midtec and its supporters argue vigorously that proceedings involving requests for reciprocal switching or terminal trackage rights under the CARs should not involve issues of “rate reasonableness” and that the Commission’s insistence on evidence relating to such issues was arbitrary. The gravamen of this argument is that in calling for such evidence in order to determine whether the C & NW has acted “contrary to the competition policies of 49 U.S. C. § 10101a or [in an] otherwise anticompetitive” manner the Commission transformed a simple complaint for reciprocal switching into a complex inquiry into the maximum reasonable rate that the C & NW could lawfully charge. This, in turn, they argue, effectively places on the complainant the onerous burden of proving that the respondent carrier is “market dominant” within the meaning of section 10709 of the Act — in essence, that the rates it charges the complainant exceed 180% of the carriers’ variable cost of providing the transportation — which triggers the Commission’s authority to prescribe reasonable rates. See 49 U.S.C. §§ 10709(aHc), 10701a(b)(l) (1982). And that, they argue, is inconsistent with the purpose of section 11103 specifically and the Staggers Act generally, viz. to emphasize enhanced competition rather than regulation as the guarantor shippers have against excessive rates. Midtec’s arguments fail to join issue with the Commission’s reasoning. The Commission clearly stated that “[t]o the extent that Midtec I ... stood for the proposition that market dominance is a jurisdictional prerequisite to obtaining relief in a competitive access controversy, that case is specifically overruled.” Midtec II, at 9. The Commission did say, however, that rate issues remain relevant in proceedings of this type, and as we have seen, it did express dismay at the lack of “numbers” in the evidence Midtec presented. But it also specifically stated that its need for rate evidence related to “rate issues ... other than maximum reasonableness,” viz. “whether access had been granted [by the C & NW] on reasonable terms.” Midtec II, at 9. It thus appears that the Commission contemplated that competitive access may be prescribed if the rate practices of a respondent carrier demonstrate that such relief is necessary to remedy or to prevent an act contrary to the competition policies of the Act or that is otherwise anticompetitive, even if the respondent carrier’s rates are not so high as to make it “market dominant” within the meaning of section 10701a(b)(l). Evidence concerning rate issues seems unavoidably relevant in the complaint proceedings now under review, if only because of the nature of Midtec’s claims. In its complaint, Midtec stated: Midtec’s principal competitors enjoy the benefits of vigorous rail competition. Most are served by at least two carriers. Thus, Midtec’s “captive status” on the [C & NW] subjects it to serious service and price disabilities. These disabilities have grown more acute because of Midtec's greatly expanded operations and the new pricing flexibilities available under the Staggers Rail Act of 1980. Midtec I, Complaint at 3, para. 8. Before the Commission, the complainants argued more specifically that: (1) the company was “severely disadvantaged in its market” visa-vis competing paper producers; (2) the C & NW’s “refusal to negotiate lower rail rates inevitably squeezes Midtec’s margins and permits the carrier to usurp for itself any efficiency gains Midtec might make”; and (3) the C & NW has “use[d] its monopoly position to squeeze the Soo into giving it totally undeserved revenues, even where it performs no work____” Even if the “market dominance” test applicable to complaints about rate “unreasonableness” is not applicable to complaints for Commission-prescribed competitive access, therefore, and some different or lesser case for introducing competing rail service is sufficient under the CARs, there would remain the need for some “numbers” or other evidentiary basis for determining whether Midtec’s allegations could be substantiated. Certainly, for example, the mere allegation that the rates — 67 — the C & NW charges Midtec are higher than the rates charged its competitors does not, without more, reveal whether the C & NW has acted anticompetitively or contrary to the statutory policies on competition. The CARs, moreover, specifically provide that the agency “shall take into account all relevant factors, including ... [t]he revenues of the involved railroads on the affected traffic, ... [t]he rates or compensation charged or sought to be charged by the railroad or railroads from which prescription or establishment is sought, ... [and] the revenues, following the prescription, of the involved railroads’ for the traffic in question____” 49 C.F.R. § 1144.5(a)(1) (1986). Indeed, it is hard to imagine how the Commission could reach an informed conclusion regarding the complainants’ case for competitive relief based on the competition policies of the Staggers Act without some evidence, some “numbers,” pertaining to the pre-complaint rates charged by the C & NW, the rates the Soo proposed to offer Midtec if reciprocal switching or terminal trackage rights were prescribed, and perhaps the rates Midtec's competitors are charged. We find Midtec’s argument that evidence concerning rates is irrelevant to its request for competitive access to be quite incredible. There remains the question, however, of what precisely the Commission’s call for “numbers” and its concern about the “reasonableness” of the terms on which the C & NW offered rail service to Midtec required the complainants to demonstrate. We note that the Commission’s answer to this question is not entirely clear. Although it stated that “there are rate issues in these cases other than maximum reasonableness,” it did not specify what particular rate practices it deems to be contrary to the competition policies of the Staggers Act or to be otherwise anticompetitive. Before the Commission, and before this court, Midtec relies on the general theory that the access provisions of section 11103 were intended by the Congress to increase interrail competition in order “to offset the very substantial rate advantages given the railroads” under other provisions of the Staggers Act. This understanding of the purposes of section 11103, Midtec contends, furthers the national rail policy of “allowftng], to the maximum extent possible, competition and the demand for services to establish reasonable rates for transportation by rail____” 49 U.S.C. § 10101a(l) (1982). The corollary to this theory, as we intimated above, is that evidence about specific rates and carrier costs is neither necessary in nor relevant to making out a case for competitive access. The assumption underlying Midtec’s theory is that even if the C & NW is not “market dominant,” it nonetheless enjoys some degree of market power, and that it has used that power to deprive Midtec of the ability to obtain “competitive rates.” Midtec argues that in situations of this type, where the Commission lacks authority to prescribe reasonable rates because the captive shipper cannot demonstrate that the carrier is market dominant, section 11103 was intended to take up the slack, so to speak. Otherwise captive shippers facing supra-competitive rates would be faced with a “heads-I win, tails-you lose conundrum.” In other words, Midtec contends that section 11103 was intended to be an alternative means of obtaining rate relief, requiring the Commission affirmatively to move the national rail system toward a regime more like perfect competition, with the attendant benefits of marginal cost ratemaking. Captive shippers that are unable to make the case for rate regulation by demonstrating market dominance, on this view, would be able to secure the benefits of effective rail competition through a regulatory order requiring a non-market dominant carrier with some unspecified degree of “market power” to provide competitive access to another carrier. It is implicit in this theory that a carrier’s ability to charge a captive shipper rates above the levels that would obtain if additional carrier service were introduced offends the competition policies of the Staggers Act or is otherwise anticompetitive. We believe that proposition is inconsistent with Congress’s intent to deregulate railroad ratemaking in the absence of a mar — 68 — ket dominant carrier. A brief examination of the statute confirms this conclusion. Under the Staggers Act, “market dominance” is defined as “an absence of effective competition from other carriers or modes of transportation for the transportation to which a rate applies.” 49 U.S.C. § 10709(a) (1982) (emphasis added). A carrier is presumed under the statute, moreover, not to enjoy market dominance, and hence, to be subject to “effective competition” with respect to a particular service, if the revenue it obtains from the rates therefor does not exceed by more than a specified percentage its variable costs in providing the service. 49 U.S.C. § 10709(d)(2) (1982); see Coal Exporters Ass’n v. United States, 745 F.2d 76, 97 (D.C.Cir.1984); Arkansas Power & Light Co. v. ICC, 725 F.2d 716, 719 (D.C.Cir.1984); Ford Motor Co. v. ICC, 714 F.2d 1157, 1159 (D.C.Cir.1983).7 The purpose of the Staggers Act in granting this rate-making freedom to non-dominant carriers was to “end for most rail service decades of ICC control over maximum rates and to permit carriers not having market dominance to set rates in response to their perception of market conditions.” Arkansas Power & Light, 725 F.2d at 719 (quoting Bessemer & L.E. R.R. v. ICC, 691 F.2d 1104, 1108 (3d Cir.1982)); see H.R.Conf.Rep. No. 1430, supra, at 80, U.S. Code Cong. & Admin.News 1980, p. 4111 (provisions of Staggers Act granting railroads “rate freedom” “continues the policy started under the Railroad Revitalization and Regulatory Reform Act of 1976 of substantially eliminating rate regulation of railroads where there is effective competition”). As the Third Circuit has succinctly stated, “[f]or railroads subject to effective competition from other rail carriers or modes of transportation Congress opted to deregulate rates.” Consolidated Rail Corp. v. United States, 812 F.2d 1444, 1449 (3d Cir.1987). More generally, “[t]he primary goal of the Act was to revitalize the railroad industry by reducing or eliminating regulatory burdens. Faced with railroad bankruptcies and the need to assure railroads of adequate revenues, Congress revamped the structure of railroad regulation in order to restore the industry to health.” Coal Exporters Ass’n, 745 F.2d at 80-81. At the same time, Congress recognized the need for regulatory intervention “where there is an absence of effective competition and where rail rates provide revenues which exceed the amount necessary to maintain the rail system and to attract capital.” 49 U.S.C. § 10101a(6); see Coal Exporters Ass’n, 745 F.2d at 96-98. The “market dominance” test serves both of these congressional policies by precluding the Commission from scrutinizing rates where “effective competition” exists, while ensuring that “rail rate flexibility would not result in [captive] shippers bearing a disproportionate share of responsibility for the needed improvements in the railroads’ financial position.” Arkansas Power & Light, 725 F.2d at 719 (quoting ICC decision under review) (brackets in original); Coal Exporters Ass’n, 745 F.2d at 97. In other words, in the absence of a market dominant carrier, it is to be presumed that there is “effective competition” and that there is no need for rate regulation. Midtec has not, of course, requested the Commission directly to fix the rates that the C & NW charges the company; it recognizes that the Commission has no authority to do so if the rates Midtec is now being charged do not exceed the level at which a carrier is deemed “market dominant.” It would seem just as inconsistent with the compromise between carrier and shipper interests inherent in the market dominance test, however, to conclude that a carrier whose rates do not bespeak market dominance is nonetheless subject to access regulation merely because those same rates are supra-competitive.8 The market — 69 — dominance test obviously does not ensure that captive shippers will pay only those rates that would obtain under “perfect competition,” i.e., rates equal to the marginal cost of providing the service; instead, it expresses a Congressional determination of the circumstances in which competition is deemed to be “effective” and in which regulation is not necessary. The alternative reading Midtec would give to the competition policies of the Act leads to the improbable conclusion that Congress contemplated that a market could be both “effectively competitive,” so as to preclude direct rate regulation, and yet insufficiently competitive to require indirect rate regulation through compelled competitive access. If the Commission were authorized, as Midtec’s argument entails, to prescribe reciprocal switching or terminal trackage whenever such an order could enhance competition between rail carriers, it could radically restructure the railroad industry. We have not found even the slightest indication that Congress intended the Commission in this way to conform the industry more closely to a model of perfect competition. See Baltimore Gas & Elec., 817 F.2d at 115. We therefore think that it is entirely consistent with the purposes of the Staggers Act that the C & NW may enjoy some degree of “market power,” enabling the railroad to price its services above marginal costs, without liability to compelled competitive access; and the Commission was not unreasonable in adopting the policy of reserving its authority to order competitive access for situations where it is “necessary to remedy or prevent an act that is contrary to the competition polices of 49 U.S.C. § 10101a or is otherwise anticompetitive.” In Pittsburgh & L. E. R.R. v. ICC, 796 F.2d 1534, 1540 (D.C.Cir.1986), we stated: [Mjarket dominance is a prerequisite for a finding of unreasonableness, not for prescription of joint rates. Prescription is stated as a power additional to the power that arises from a finding of market dominance. Nothing that we say today should be construed to cast doubt on this conclusion. There may be myriad carrier practices that can properly be considered anti-competitive or contrary to the competition policy of the Staggers Act, without any need to determine whether a carrier is market dominant. When the basis of a request for competitive access is solely that a carrier’s rates are too high, however, the statute makes clear that such rates can be deemed anti-competitive or contrary to the competition policy of the Staggers Act only upon a showing of market dominance. Setting aside Midtec’s broad attack on the application of the market dominance concept to competitive access complaints, we return to the more specific allegations it has raised. Those specific allegations apparently, but only apparently, go beyond the mere claim that the C & NW’s rates are supra-competitive in the sense that they exceed the variable costs of providing service to Midtec. Recall that Midtec alleges that the C & NW’s rates put the company at a disadvantage vis-a-vis its competitors, that the C & NW has appropriated an unreasonable portion of the company's revenues, and that the C & NW has abused its market power by “squeezing” undeserved revenues from the Soo. The evidence before the Commission does not substantiate Midtec’s claim that it was at a competitive disadvantage in the downstream market for paper products. The evidence showed that Midtec more than doubled its production since it invested in the Kimberly mill and that in 1985 it served 157 cities, as compared to only 90 in 1982. Midtec II, Appendix at 4. Nor is there any evidence that Midtec’s competitors were able to serve markets foreclosed to Midtec because they were able to obtain more favorable rail rates, or indeed even that they did obtain more favorable rates.9 — 70 — There is, moreover, nothing inherently anticompetitive or contrary to the competition policies of the Staggers Act about transportation rates that preclude distant shippers from competing in downstream markets. Even if there were perfect competition among rail carriers, some shippers would be able to underprice their competitors because their transportation costs are lower, reflecting closer proximity or more efficient carriers serving them. For the disadvantageous rates to be material to the competition policies of the Staggers Act or otherwise anticompetitive, something more is needed. For example, an allegation that the rates charged are the result of a purposeful effort to constrain downstream competition by an agreement between the C & NW and Midtec’s competitors would surely suffice. Similarly, an allegation that the disadvantageous rates make the carrier, by definition, “market dominant,” should be sufficient. Midtec makes neither type of allegation, however. Midtec’s second allegation, that the C & NW's “refusal to negotiate lower rail rates inevitably squeezes Midtec’s margins and permits the carrier to usurp for itself any efficiency gains Midtec might make,” also suggests a need to demonstrate “market dominance.” There is no claim, for example, that this alleged “squeeze” is anticompetitive in any specific way. Instead, Midtec’s claim is really nothing more than a grievance that the C & NW's rates are too high. As we have seen, however, a rate cannot be considered “unreasonable” short of market dominance, and rates below the market dominant threshold cannot be regarded as anticompetitive or contrary to the competition policies of the Staggers Act.10 The final allegation relating to the C & NW’s ratemaking practices concerns the complaint that the C & NW has abused its market power by squeezing undeserved revenues from the Soo. The complainants argued that the C & NW offers Midtec an “attractive” “proportional” rate on inbound shipments of woodpulp moving between warehouse facilities at Neenah, Wisconsin and the Kimberly mill, but only on the condition that the C & NW share in the line-haul rate for the Soo’s movement of the woodpulp from its Canadian origin to Neenah, a portion of the service in which the C & NW allegedly does not participate. This, the complainants’ argued, “is a perfect example of [the C & NW’s] ability to use its monopoly position to squeeze the Soo into giving it totally undeserved revenues, even where it performs no work, just so [the Soo] can handle the [wood]pulp.” The Commission disposed of this claim by concluding that [i]t does not really matter whether [the C & NW] takes all its compensation in the proportional rate or whether it obtains additional compensation by some other — 71 — means. The real question is whether the overall compensation it receives is reasonable in light of the service it has performed. On the record here, we find no basis for finding that it is not. Midtec II, at 12. The contention that this rate arrangement shows that the C & NW has abused its market power does not bear close scrutiny. While the evidence is somewhat ambiguous, the arrangement appears to be the result of the C & NW’s efforts to attract traffic that was previously moving at higher cost by rail from Canada to the Neenah warehouse, where it was transshipped by truck to the Kimberly mill. The C & NW simply offered a competitive price, in the form of a joint rate on the long haul in which the C & NW provided the final segment of track. There is no evidence indicating that the arrangement was unprofitable to the Soo or that the overall compensation the C & NW received was unreasonable. For its part, Midtec has described the arrangement as “very nice” and “helpful” because “this favorable rate permits [Midtec] to follow the rail move into the Neenah warehouse with a rail switch in to [its] mill.” We cannot say that such an arrangement, without more is conduct that is contrary to the competition policies of the Staggers Act or that is otherwise anti-competitive. In summary, Midtec’s allegations that the C & NW’s ratemaking practices were anticompetitive rested on theories that required supporting evidence regarding rates, costs, and revenues. We find nothing arbitrary about the Commission’s conclusion that without such evidence the complaint must fail. Alternatively, Midtec argues that although its evidence did not contain “numbers,” it nonetheless satisfied the requirements of the CARs. The evidence presented, Midtec argues, clearly demonstrated that the C & NW had abused its market power because Midtec was able to obtain improvements in rates when it “or some third party pushed very hard to get [them] or because [the C & NW] decided to behave while this case is pending.” More particularly, Midtec claimed that the C & NW improved its rates because: (1) this litigation threatened the C & NW's monopoly position; (2) connecting carriers put pressure on the C & NW; or (3) “truck or water competition ... plainly threatened to capture [Midtec’s] business.” These facts, it maintains, show that the C & NW is not subject to effective competition and that regulatory intervention is necessary under the competition policy of the Staggers Act. The Commission rejected these grievances, saying: “That Midtec has had to bargain hard to obtain what it seeks is no reason to complain. It is only when hard bargaining results in an abuse of market power and an insistence on terms that are unreasonable that we should intervene.” Midtec II, at 11. Midtec’s evidence does not establish that it was unable to negotiate favorable rates, but only that the C & NW did not volunteer them. Midtec’s ability to get rate reductions at all surely suggests that the C & NW’s market power was constrained in some substantial way. It is no surprise, and no inherent offense to the Staggers Act, for the C & NW to try to maximize its profits at Midtec’s expense. The picture that Midtec has painted is simply not one of unrestrained market power being exercised against it. While it may be true that the rate concessions to which the C & NW agreed were the result of this lawsuit or the threat of legal action, rather than competitive market pressures, there is substantial evidence to the contrary. For example, the C & NW maintains that it began to offer reduced rates before the threat of legal action, and that even the concessions it made to Midtec after the filing of the joint complaint were not attributable to the threat of regulatory intervention; instead, these rate concessions were alleged by the C & NW to be responsive to the new, deregulated environment under the Staggers Act and to the existence of competition for Midtec’s traffic. We cannot say that the evidence establishes that the C & NW’s actions were — 72 — merely an opportunistic attempt to avoid the Commission’s regulatory intervention, and are likely to be short-lived. It was therefore not inappropriate for the agency to deny present relief, especially in light of its announcement that, since the evidence is ambiguous, it “stands ready to grant relief on an expedited basis is necessary to remedy anticompetitive conduct by this railroad in the future.” Midtec II, at 12. Indeed, if Midtec’s concerns are better founded than the Commission thought, the Commission’s expressed readiness to intervene “on an expedited basis” should continue into the future the restraining influence Midtec attributed in the past to the pendency of this case, thus benefiting Midtec without burdening the C & NW more than is necessary. We are left, then, with Midtec’s claim that in calling for specific evidence concerning rates, revenues, and costs, the Commission imposed an impossible burden on it. Midtec relies essentially on three points. First, it claims that obtaining evidence of this type was “logistically impossible” because it would require “an elaborate revenue/cost/efficiency analysis for each of its nine inbound and outbound products to and from each of the myriad origins and destinations.” In this regard, Midtec points to the evidentiary burdens shippers have confronted in documenting complaints of market dominance. Such proceedings are inherently fact-intensive, as are cases of the present sort. We think that Midtec overstates the requirements of the rules and of the Commission’s decision, however. Even in proceedings involving allegations that rates exceed the reasonable maximum, the Commission has ruled that “the revenue-to-variable cost threshold and other market dominance criteria can be adequately determined by using representative shipments.” Westinghouse Elec. Corp. v. Alton & S. Ry., No. 38188S, Decision at 2-3 (served Jan. 25, 1988). Nothing in the Commission’s decision here under review indicates that something more is required in complaint proceedings under section 11103, and, as we discussed above, it is not at all clear why anything less should suffice. Second, Midtec argues that evidence of this type has never before been required in a section 11103 complaint proceeding, and therefore it had no reasonable notice that it would be required to produce such evidence in this proceeding. Here Midtec cites decisions in which the agency did not require rate evidence, but they all antedate the CARs. Although it is true that when it reopened the record in this case the Commission did not specifically call for rate evidence, we cannot agree that Midtec was not reasonably placed on notice that such evidence was relevant to its complaint. As we have seen, the text of the CARs itself clearly indicates that evidence concerning existing and proposed rates and costs is to be considered in competitive access complaint proceedings. Moreover, as an argument for reopening the record— which Midtec opposed—the C & NW noted that the evidence presented in Midtec I “was not prepared with a view toward satisfying the standards set forth [in the rules].” In reopening the record thereafter, the Commission specifically called for “evidence and argument developed in light of [the] standards [of the CARs].” 11 In view also of the extensive discussion of rate evidence in Midtec I, and of the Commission’s conclusion that the lack of such evidence was fatal to the joint complaint in that proceeding, Midtec’s claim that it had no notice that rate evidence was relevant in Midtec II is far too strained to be accepted. Finally, Midtec notes that the Commission's discovery rules are very restrictive in some areas, and claims that it would have been denied discovery for the specific type of evidence demanded by the CARs. The fact is that Midtec did not seek discovery until after the Commission rendered its decision, Midtec had filed its petition for — 73 — review, and the matter was therefore in this court rather than before the Commission. As we have stated before, a litigant before an agency “cannot properly complain because it did not receive information it did not attempt to obtain through the procedures available to it.” Potomac Elec. Power Co., 744 F.2d at 192. Even if discovery would have been futile, moreover, some of the relevant rate evidence was readily available to Midtec. Midtec surely knew what it was paying the C & NW. Likewise, the Soo could have offered evidence of the rates it proposed to charge Midtec for the services then provided by the C & NW. Whether this evidence would have been sufficient to determine whether the C & NW should be ordered to provide competitive access, we do not know because the agency was not given the opportunity to consider that question. But surely such evidence was relevant to the complainants’ claim that granting the Soo direct access to the mill was required by the competition policies of the Staggers Act. They simply failed to avail themselves of the opportunity to conform their evidentiary submissions to the requirements of the Commission’s rules; Midtec cannot now complain that the evidentiary burden placed upon them was unduly prejudicial. 2. The Adequacy of the C & NW’s Service to Midtec Midtec and its supporting intervenors do not specifically argue that issues concerning the service the C & NW provided to Midtec are not relevant under the CARs. Instead, they argue that while this factor may be relevant under the “public interest” standards of section 11103, it has nothing to do with the alternative statutory criterion for ordering reciprocal switching, viz. whether it is “necessary to promote competitive rail service.” Our earlier discussion of the relationship between the CARs and the statutory standards disposes of this claim, for we have concluded that the requirement of the CARs that complainants demonstrate anticompetitive behavior or the probability thereof as a prerequisite to obtaining relief under section 11103 is not arbitrary or inconsistent with the purposes of the Staggers Act. Evidence of the carrier’s actual (or threatened) conduct — such as the adequacy of the service it provides to a captive shipper — is the most direct and probative evidence by which to assay whether the carrier has acted (or is likely to act) anticompetitively. We note again that Midtec initiated this proceeding with an allegation of “serious service and price disabilities.” The Commission’s call for evidence to gird this allegation is, therefore, hardly arbitrary. Such evidence is also patently relevant in determining whether the conduct of the C & NW is contrary to the competition policies of the Staggers Act. It is elementary that a monopolist may extract monopoly rents either by increasing its prices or by decreasing the quality of its product or service. An abuse of market power by a railroad thus may take the form of excessive rates for otherwise adequate services, or alternatively, otherwise “reasonable” rates for inadequate services. A railroad cannot be allowed to avoid regulatory intervention by charging rates below the threshold for price regulation under the Act while obtaining monopoly rents by lowering the quality of its services. The Commission is not authorized to look at the adequacy of service in determining whether a particular carrier is “market dominant,” however. Nor would it be administratively practical for it to assign quantitative values to the quality of a carrier’s services, in order to determine whether it is exercising market dominance over shippers through qualitative degradation. A qualitative analysis is required, therefore, to determine whether the carrier may nonetheless be exercising market power in contravention of the competition policies of the Act or in a manner that is otherwise anticompetitive. In particular, deficiencies in services may offend the national rail transportation policies of ensuring “a sound rail transportation system ... to meet the needs of the public,” ensuring “effective ... coordination between rail carriers and other modes,” and “... prohibiting] unlawful discrimination.” 49 U.S.C. § 10101a(13) (1982). — 74 — Before this court, Midtec argues only that the general quality of the C & NW's service is inadequate, and alludes to specific examples only to illustrate this point. To the extent that Midtec relies on the proposition that the improvements in the services it receives, like the improvements in the rates it is charged, occurred only under the threat of litigation, our discussion in the preceding section applies equally here. Of Midtec’s continuing service complaints, we discuss only three representative examples. First, Midtec claims that the C & NW is unresponsive to its need for special 100-ton capacity boxcars. These cars are needed because many of Midtec’s customers purchase paper products in 100-ton lots, and “by being restricted to [the 90-ton capacity boxcars provided by the C & NW], Midtec must split these loads into two cars with the accompanying increase in price.” The C & NW stated that the boxcars provided were indeed 100-ton capacity cars, which it acquired and upgraded specifically to meet Medtec's needs. The Commission concluded that “[w]hile the capacity of the cars is in dispute, Midtec did not establish that [the C & NW] was indifferent to its needs for quality service by failing to supply car types which it requested.” Midtec II, at 12. The record supports this conclusion; there is no evidence that Midtec ever requested larger capacity boxcars from the C & NW or otherwise indicated to the carrier that the boxcars provided were inadequate to its needs. The second alleged service inadequacy relates to Midtec’s need for inbound shipments of clay. The company currently receives all of its clay from the southeast via a through route involving an interchange at East St. Louis between either the Southern Railway System or the Southern Coast Line Railroad and the C & NW. Midtec argued before the Commission that if it were given competitive access to the Kimberly mill, the Soo could provide a more direct and efficient routing for this traffic via a through route involving an interchange at Louisville, Kentucky. This routing, Midtec argued, would “provide the competitive option which is now absent and which accounts for the fact that we have been unable to secure rate relief for this traffic from our present carriers.” The C & NW countered by stating that it has offered Midtec a more efficient routing for this traffic through an interchange at Chicago; it expressed its confidence that the C & NW could “more than match any joint service the Soo could provide on movements to and from the Southeast.” Midtec conceded that it chose to route its inbound shipments of clay through “the somewhat more circuitous interchange of East St. Louis, rather than Chicago,” not because the C & NW insisted on such a routing, but instead “to give the [C & NW] the benefit of a longer haul, which is the only pressure [Midtec] can put on the origin carrier.” Midtec nonetheless argues for granting the Soo competitive access rights so Midtec could “exert far more pressure on the origin carriers to reduce [its] rail costs.” We agree with the Commission that the C & NW “cannot be blamed for a circuitous routing which is not of its choosing and which it will remedy if asked.” Midtec II, at 12. This alleged service inadequacy cannot fairly be regarded as carrier conduct that is anticompetitive or contrary to the competition policy of the Staggers Act. The final allegation of inadequate service we consider involves inbound shipments of coal, which Midtec presently receives via barge/truck routing. The traffic originates in Eastern Kentucky and West Virginia; it is shipped to Green Bay, Wisconsin by water carrier, and then transshipped to the Kimberly mill by truck. Midtec argues that the Soo can provide single line rail service for this traffic, a routing that Midtec contends “could very well provide an attractive alternative source for [its] coal needs____” The C & NW responds that it is able to provide through route service on coal shipments via a Chicago interchange, but that “the reason why inbound coal moves by truck rather than rail has nothing to [do] with the quality of service [the C & NW] is able to provide; rail is not used because of the limited capacity of Midtec’s coal stor — 75 — age and track facilities at Kimberly.” Midtec conceded that the C & NW offered reduced rates in an effort to obtain the coal traffic, and that Midtec rejected that offer because its “track to the coal storage area had become derelict” and “[cjonsidering the costs of repairing the track, the rail rate was not competitive____” Again, there is nothing in this allegation that indicates that the C & NW has abused its market power. To the contrary, it appears that intermodal competition has effectively kept the C & NW from participating in Midtec’s coal traffic. In sum, viewing the evidence as a whole, we find no support for the claim that the C & NW has exercised market power by providing inadequate service to the detriment of Midtec. 3. Geographic and Intermodal Competition Relying on the Commission’s prior decision in Delaware & Hudson, Midtec argues that the Commission erred by considering evidence of intermodal and geographic competition in determining whether reciprocal switching or terminal trackage rights were warranted under the CARs. According to Midtec, when the Commission adjudicates a complaint pursuant to section 11103, it is required to look exclusively at intramodal (i.e., rail vs. rail) competition. That would be a pretty silly exercise from an economic or a common sense perspective. Not surprisingly, Congress required no such foolishness. The CARs commit the agency to considering “all relevant factors,” 49 C.F.R. § 1144.5(a)(1), specifically including inter-modal competition, Intermodal Rail Competition, 1 I.C.C.2d at 827; and it could hardly be doubted as a matter of economics that intermodal and geographic competition are relevant in determining whether there is the potential for anticompetitive conduct in a particular market. Market forces are impersonal; they do not distinguish between modes unless there are functional differences that limit one mode’s effectiveness in competing for traffic. Congress was specifically apprised of this elementary economic principle when the Senate Committee on Interstate and Foreign Commerce reported on the Staggers Act: The rate provisions ... are designed to protect shippers who have no transportation alternative, and who are being charged rates that are too high. The test of a transportation alternative is a sound one. If a shipper can rely on a transportation alternative, which would include another railroad, a barge, or a truck, at a transportation cost which is not substantially greater than the rail transportation cost, then competition is present. Competition will serve to hold down rates, and the railroad involved would have no market power. H.Rep. No. 1035, 96th Cong., 2d Sess. 39 (1980), U.S.Code Cong. & Admin.News 1980, p. 3984. And, as we stated in Central Vermont Ry. v. ICC, 711 F.2d 331, 336 (1983): A major impetus behind the deregulation of railroads in the Staggers Rail Act ... was Congress’ recognition that railroads generally had to compete with other modes of transportation. See, e.g., H.R. Rep. No. 1430 (Conf.Rep.), 96th Cong., 2d Sess. 79 (1980), reprinted in 1980 U.S. Code Cong. & Ad.News 4110: [Historically the enactment of the Interstate Commerce Act was essential to prevent an abuse of monopoly power by railroads____ However, today, most transportation is competitive and many of the Government regulations affecting railroads have become unnecessary and inefficient. Nearly two-thirds of intercity freight is transported by modes of transportation other than railroads. If Congress can rely on truck-rail competition to obviate the need for regulation, surely the ICC can do so as well. This is especially so in light of the congressional policy “to minimize the need for Federal regulatory control over the rail transportation system.” 49 U.S.C. § 10,101a(2) [sic]. See also Brae Corp. v. United States, 740 F.2d 1023, 1039 (D.C.Cir.1984); Western — 76 — Coal Traffic League v. United States, 719 F.2d 772, 777-80 (5th Cir.1983) (en banc). There is nothing in either the text or the legislative history of section 11103 that remotely suggests that the Commission should focus narrowly on intramodal competition and blind itself to all other sources of competition. The Senate Report on the original bill did state that the relevant section “deals with intramodal competition among railroads.” S.Rep. No. 470, 96th Cong., 2d Sess. 41 (1979). That is just to say, however, that an order prescribing reciprocal switching by its nature increases the absolute number of rail carriers providing service on the trackage segment over which switching is ordered, and thus the provision “deals with intramodal competition.” 12 It says nothing about whether competition from other modes should be excluded from the Commission’s analysis of the need for competitive rail access. In fact, the same report went on to state that “[i]n areas where reciprocal switching is feasible, it provides an avenue of relief for shippers served by only one railroad where service is inadequate.” Id. at 42 (emphasis added). This observation clearly recognizes that the absence of rail competition does not necessarily mean that a “captive shipper” is being denied adequate service, such that regulatory intervention is necessary. Otherwise, any shipper served by only one railroad would automatically be entitled to an order requiring reciprocal switching or terminal trackage rights. As we have seen above, there is no warrant for the view that Congress intended such a radical restructuring of the railroad industry. To the extent that the Commission previously interpreted the intent of section 11103(c)(1) to focus exclusively on intramodal competition, it is clear to us that it erred. In Midtec I, the Commission, too, noted the error, and to that extent it expressly overruled Delaware & Hudson. The Commission said: Our earlier view was based only on the Senate Report ... accompanying section 203(b) of Senate Bill No. S. 1946, stating that section 203 (the forerunner of what ultimately added section 11103(c)) deals with intramodal competition among railroads. At that time we read this language as requiring us to consider only existing rail competition in reciprocal switching determinations. A better interpretation of the Senate Report, and one we adopt here, is that section 11103(c)(1) allows consideration of all existing competition and the addition of another rail carrier, an intramodal competitor, only when there is inadequate competition. This interpretation is in keeping with the rail transportation policy of ensuring effective competition among rail carriers and other modes. 49 U.S.C. [§ ]10101a(4) and (5). Midtec I, I.C.C.2d at 367 (emphasis in original). We agree. The practice of considering only intramodal competition, to the exclusion of other market forces that effectively constrain a railroad’s market power, is inherently illogical and was congressionally disapproved. The CARs also contemplate that the Commission will consider “geographic competition” in evaluating a competitive access complaint. In this regard, the rules place on the respondent carrier the burden of producing “clear and convincing” evidence, 49 C.F.R. § 1144.5(b)(2) (1986), that there “is a restraint on rail pricing stemming from a shipper’s or receiver’s ability to get the product to which the rate applies from another source____ Because the shippers and receivers can do this, the railroad must compete with the railroad serving the alternate source or destination.” Market Dominance Determinations and Consideration of Product Competition, 365 I.C.C. 118, 128 (1981) (defining geographic competition).13 — 77 — Midtec argues only that section 11103(c) limits the ICC to considering intramodal transportation, making no separate argument against consideration of geographic, as opposed to intermodal, competition. We see no apparent reason for excluding evidence of the effect of this competitive constraint in determining whether reciprocal switching or terminal trackage rights are necessary either to remedy or to prevent anticompetitive behavior. We therefore cannot say that the Commission’s decision to consider this factor was arbitrary or contrary to the purposes of the Staggers Act.14 In any event, the Commission did not rely on structural evidence, such as the existence of geographic competition, in dismissing the joint complaint. Midtec II, at 10 (“the existence of competition, or the lack of it in any given area, is not dispositive of this case”) Rather, as we have seen, the Commission found that the specific behavioral evidence presented by the complainants failed to show that the C & NW had acted or was likely to act anticompetitively. The Commission concluded that the structural evidence (“including intermodal, intramodal and geographic but not product competition which has been excluded by rule”) demonstrated only that “different commodities face varying degrees of competition.” Midtec II, at 10. It did not show “whether anticompetitive conduct has occurred or is likely to occur,” and therefore merely “useful background information.” Midtec II, at 9. This conclusion, which is supported by substantial evidence, is a sound one. The CARs focus on behavior that is contrary to the competition policies of the Staggers Act or that is otherwise anticompetitive. Structural evidence at best can indicate a potential for anticompetitive behavior in a particular market; the actual conduct of the carrier can provide a sound basis for determining the need for reciprocal switching or terminal trackage rights. In this case, however, such limited resort as the Commission had to structural evidence merely tended to confirm, not to refute, the conclusion, firmly grounded in the record, that the carrier had not been shown to engage in anticompetitive conduct.15 In sum, we find that evidence concerning rates, intermodal and geographic competition, and service inadequacies are relevant and not arbitrary factors in determining whether reciprocal switching or terminal trackage rights are necessary either to remedy or to prevent anticompetitive behavior in the railroad industry. IV. Conclusion Like the Commission, we note that the posture of this case may have had an effect on Midtec’s ability sufficiently to document the adversities it claims to have faced as a self-styled “captive shipper.” Our review of the evidence submitted by the complainants indicates to us that in large measure they proceeded on the theory that that status, and theinferences they would draw — 78 — therefrom, standing alone, were sufficient to entitle Midtec to additional carrier access. The Commission’s rules, adopted while this case was pending, decisively rejected that theory, and we have concluded that the rules are not contrary to the language or the policies of the Staggers Act. We recognize that the litigating burden on Midtec may have been increased by this midstream change of course. The Commission’s expressed willingness to grant relief on an expedited basis if necessary, however, convinces us that Midtec’s predictions of financial disaster are not well-founded. We have not the slightest reason to doubt that the Commission means what it says. On the record before us, we cannot say that the Commission acted arbitrarily in dismissing the joint complaint. The petition for review is therefore DENIED. Footnotes [1] In another form of the through route arrangement, Midtec could pay a "combination rate," which is the sum of the rates independently set by each individual railroad participating in the through route. See Pittsburgh and Lake Erie R. Co. v. ICC, 796 F.2d 1534, 1536-37 (D.C.Cir.1986). [2] The Commission determined that the C & NW’s Kaukauna branch line is a terminal facility within the meaning of 49 U.S.C. § 11103 (1982), which conclusion is not at issue in this review proceeding. [3] In the cited case, the Commission explained that the "actual necessity or compelling reason" test informing the "practicable and in the public interest" standard, see Jamestown Chamber of Commerce, supra, involves a four-step inquiry; before it will order reciprocal switching; (1) interchange and switching must be feasible; (2) the terminal facilities must be able to accommodate the traffic of both competing carriers; (3) the presence of reciprocal switching must not unduly hamper the ability of either carrier to serve its shippers; and (4) the benefits to shippers from improved service or reduced rates must outweigh detriments, if any, to either carrier. Delaware & Hudson, 367 I.C.C. at 720-21. [4] Regardless of how far the product must be transported or of the costs of transporting it, delivered prices for comparable paper products tend to be similar in any one area. Paper producers facing higher transportation costs wotild have to, therefore, manufacture their products more efficiently in order to be able to compete in a market with producers that incur lower transportation costs. [3] In determining whether a rail carrier is market dominant under 49 U.S.C. [§] 10709, the Commission considers whether a rail carri — 56 — er faces effective competition, Le., intermodal, intramodal, product and geographic. Market Dominance Determinations, 365 I.C.C. 118 (1981). [5] The Soo has not joined in Midtec’s petition for review, nor has it sought otherwise to participate in this proceeding. Intervening in support of the petitioner are the Chemical Manufacturers Association, National Industrial Transportation League, Western Coal Traffic League, Florida Pho.sphate Council, Growmark, Inc., Allied-Signal Corp., The Andersons, Farmland Industries, Inc., IMC Fertilizer, Inc., North Dakota Public Service Commission, and Pennzoil Sulphur Co. Precompetitive Rail Steering Committee, amicus curiae, has joined the brief of these intervenors. The C & NW hits filed a brief as intervening respondent. The Association of American Railroads and Consolidated Rail Corp. have filed a brief as amici curiae in support of the respondents. [6] The critical distinction between the "substantial evidence test" and the "arbitrary and capricious" test as applied to judicial scrutiny of agency factfinding lies not in the standard by which the courts review those findings, but rather in the scope of the judicial inquiry. See Data Processing, 745 F.2d at 684. [7] Even if a carrier’s rates are found to exceed the "market dominance” threshold, "there are congressionally provided for zones of rate flexibility based ... on revenue adequacy. Within them rates can be increased if a carrier’s revenues are found to be inadequate.” Coal Exporters Ass’n, 745 F.2d at 97 (citing 49 U.S.C. § 10707a). [8] An order granting terminal trackage rights or reciprocal switching can also lead to direct "rate regulation” because the Commission is empow — 69 — ered to fix the compensation governing reciprocal switching or terminal trackage rights if the carriers cannot reach agreement. See 49 U.S.C. § 11103(a) and (c)(1). [9] Midtec’s principal witness candidly conceded that they "do not know the terms of [their competitors’ rate] contracts,” but nonetheless asserted that “they have been able to negotiate favorable arrangements because they have competition." We agree with the Commission that — 70 — general allegations of this type would not support a conclusion that the C & NW’s conduct contravenes the competition policies of the Act or is otherwise anticompetitive. See Midtec II, at 11. Indeed, the Commission concluded that the available evidence supports the conclusion that complainants' contentions are incorrect. [The C & NW] denies that its rates are higher to Midtec than to other shippers, and complainants have submitted no evidence to the contrary. The C & NW argued before the Commission that it "offered similar rates and service to all of [the] mills [in Wisconsin and Upper Michigan,]’’ and that the C & NW’s "effective rail rates — the rates that actually move the traffic — are comparable for all of these mills, including the Midtec mill.” It supported this contention by noting that it would have been foolhardy for [the] C & NW to fail to accord Midtec the same rate relief and service it accorded to these other paper mills. The likely result of such a failure would have been that Midtec could not compete with these mills, its traffic would have shifted to the other mills and possibly to other rail carriers, and [the] C & NW would have been deprived of the opportunity to enjoy any revenues from the Kimberly mill notwithstanding the fact that it alone serves that mill. As we shall see, infra, this contention is supported by evidence of geographic competition. [10] Where there is no “effective competition," “the distribution of economic rents between carriers with market power and shippers [is not] a matter of regulatory indifference.” Coal Exporters Ass’n, 745 F.2d at 95. Where effective competition does exist, however, — Le., where the carrier is not market dominant — the carrier may set its rates at will. [11] The Commission went on to note that *‘[m]atters of concern include the proper role of intramodal, intermodal, and geographic competition in determinations under the new regulations." (Emphasis added.) Contrary to Midtec, we do not view this statement as reasonably indicating that only evidence of such competition was called for. [12] So, too, with the statement in a House colloquy that the "provisions will introduce additional competition between railroads." 126 Cong. Rec. H5906 (June 30, 1980). [13] Under Commission precedent, evidence concerning the following factors must be submitted in order to establish the potential for geographic competition: (1) the number of alternative geographical sources of supply or alternative destinations — 77 — available to the shipper or receiver for the product in question; (2) the number of these alternative sources or destinations served by different carriers; and (3) that the product available from each source or required by each destination is the same. Market Dominance Determinations, 365 I.C.C. at 134. [14] In other circumstances we have expressed some doubt about the weight to be given geographic competition in considering whether captive shippers would be subject to an abuse of market power "absent evidence about cross-elasticities of particular commodities with substitute products or commodities from other geographic locales____” Brae Corp. v. ICC, 740 F.2d 1023, 1039 (D.C.Cir.1984); see also Coal Exporters, 745 F.2d at 98-99 (absent statutorily required finding of lack of market power, claims of geographic competition unavailing to support exemption from regulation). We have never held that the Commission should disregard evidence of geographic competition for which a proper foundation has been laid. The Commission’s own requirement that evidence of geographic competition be "clear and convincing,” however, limits its ability to rely on such evidence in any case governed by the CARs. [15] The Commission found, for example, that "the evidence of geographic competition for outbound movements of coated paper is quite convincing." Midtec II, Appendix at 5. Notably, the Commission found, and we agree, that Midtec’s own evidence supports this conclusion. See id.