In this action brought under the Sherman Act by an unregulated competitor against a natural gas distributor, we must explore the interplay between state public utility regulation and federal antitrust law.
I
Cost Management Services, Inc. (“CMS”) is a Seattle-based company that purchases natural gas on the open market and sells it to various commercial and industrial consumers. In the trade area in which CMS sells gas, Washington Natural Gas Company (“WNG”) owns the only delivery facilities capable of transporting gas from the interstate pipeline to end users. Accordingly, customers who buy gas from CMS must obtain delivery of that gas through WNG’s delivery facilities. WNG charges those customers, who are known as “transporters,” a “transport charge” for the use of its facilities. Customers who purchase gas from WNG are known as “system sales” customers.
WNG is subject to regulation by the Washington Utilities and Transportation Commission (‘WUTC”). The WUTC has approved two tariffs which' are relevant to this ease.1 The first, Tariff 57,- applies when customers purchase gas from a source other than WNG. According to CMS, Tariff 57 dictates the terms and conditions which WNG must follow when it charges CMS’s customers for the use of WNG’s delivery facilities. The second tariff, Tariff 87, applies when system sales customers purchase gas directly from WNG. CMS alleges that a customer’s actual or anticipated gas usage must meet a particular minimum volume threshold in order to be eligible for the pricing set forth in Tariff 87.
In September 1994, CMS filed a complaint against WNG in federal district court alleging that in the relevant market for sale of gas, WNG has a market share of over 90 percent. The complaint' also alleged that WNG has engaged in a variety of practices in an attempt to monopolize that market, and that WNG has indeed monopolized that market.2 In particular, the complaint sought recovery based on four alleged violations of section 2 of the Sherman Act, 15 U.S.C. § 2 (1994).
First, CMS alleged that WNG has violated Tariff 87 by offering gas at the apparently low rates specified in that tariff to customers who do not meet the mandatory minimum volume requirements in the tariff. CMS alleged that WNG’s “off-tariff pricing” has made it impossible for others to compete for the sale of gas to those particular customers, and that competition in the market has been injured as a result.
Second, CMS alleged that WNG has engaged in what CMS calls “monopoly leveraging.” More specifically, CMS alleged that WNG has used its monopoly over gas delivery facilities in an unlawful attempt to monopolize the market for gas sales.3 CMS claimed that WNG has enhanced its monopoly over gas sales by including anticompetitive provisions in Tariff 57 which are designed to make it economically inefficient for consumers to purchase gas from anyone other than WNG.' CMS alleges, for example, that WNG has forced customers who purchase gas from sellers other than WNG to install expensive telemetry equipment which monitors the customer’s gas usage. Customers who purchase gas from WNG are not required to install similar equipment. In addition, CMS alleges that customers who purchase gas from sellers other than WNG must pay a monthly customer charge and are subject to a minimum monthly bill, while customers who pur*941chase gas from WNG pay no- customer charge and a lower monthly minimum.- CMS alleges that WNG has “no valid business reason” for including these provisions in the tariff. CMS further argues that WNG was able to obtain WUTC approval for the tariff only by repeatedly refusing to provide the WUTC with critical documents which CMS alleges that the WUTC admitted it needed in order to make an intelligent decision on whether to approve the tariff.
Third, CMS alleged that WNG violated Tariff 57 in an attempt to encourage customers not to purchase gas from CMS. More specifically, CMS alleged that under the express provisions of Tariff 57, customers who desired to convert from “system sales” (purchasing gas from WNG) to “transport sales” (purchasing gas from CMS) were only allowed to do so on October 1, 1994, and were required to commit to purchasing transport services from WNG for a period of one year. According to CMS, customers wishing to convert were required to notify WNG in writing no later than July 1,1994. On June 28,1994, CMS provided WNG with a list of customers who wished to convert. CMS alleged that immediately thereafter, WNG extended the switching deadline from July 1, 1994, to August 1,1994, and thén attempted to persuade customers who had given notice of their intent to convert to remain with WNG.
Fourth and finally, CMS alleged that WNG has interfered with CMS’s relationship with its customers. In particular, CMS alleged that certain customers have agreed to purchase gas from CMS and have asked CMS to represent them in their dealings with WNG, but that WNG has unlawfully refused to permit the customers to be represented by CMS. CMS alleged that this conduct constituted an additional attempt by WNG to eliminate competition.4
WNG moved to dismiss CMS’s Sherman Act claims under Fed.R.Civ.P. 12(b)(6). WNG contended that (1) CMS had faded to allege the elements of an antitrust violation, (2) the state action immunity doctrine was a bar to the claims,' (3) the “filed tariff” or Keogh doctrine was a bar to the claims, and (4) the claims were barred under the doctrine of primary jurisdiction. In' May. 1995, the district court entered an order .which granted WNG’s motion. The court found that CMS’s antitrust claims were barred by the state action immunity doctrine, but did not address WNG’s remaining contentions. The court also dismissed CMS’s state law claims without prejudice. CMS timely appealed.
II
CMS first argues that the district court erred by finding that its claims were barred by the “state action immunity” doctrine. This doctrine, which originated in the Supreme Court’s opinion in Parker v. Brown, 317 U.S. 341, 63 S.Ct 307, 87 L.Ed. 315 (1943), “exempts qualifying state and local government regulation from federal antitrust;'even if the regulation at issue compels an.otherwise clear violation of the federal, antitrust laws.” H. Hovenkamp, Federal Antitrust Policy: The Law of Competition and its Practice § 20.2, at 673 (West 1994) (hereinafter “Federal Antitrust Policy ”). The doctrine is based on principles of federalism, and on judicial recognition of the fact that “[continued enforcement of the, national antitrust policy grants the States more freedom, not less, in deciding whether to subject discrete parts of the economy to additional regulations and controls.” F.T.C. v. Ticor Title Ins. Co., 504 U.S. 621, 632, 112 S.Ct. 2169, 2176, 119 L.Ed.2d 410 (1992). However, a broad interpretation of the doctrine may inadvertently extend immunity, to, anti-competitive activity which the states did not intend to sanction. Because of a concern that a broad application of the doctrine will thus impede states’ freedom by threatening to hold them accountable for private activity they, do not condone “whenever they enter the realm of economic regulation,” the doctrine is said to be “disfavored.” Id. at 635-36, 112 S.Ct. at 2177-78.5
*942In California Retail Liquor Dealers Ass’n v. Midcal Aluminum Inc., 445 U.S. 97, 100 S.Ct. 937, 63 L.Ed.2d 233 (1980), the Supreme Court outlined a two-part test which courts must apply when determining whether alleged anticompetitive conduct is immunized under the doctrine. As recently described by the Court in its decision in Ticor Title,' supra, under the Midcal test
[a] state law or regulatory scheme cannot be the basis for antitrust immunity unless, first, the State has articulated a clear and affirmative policy to allow the anticompeti-tive conduct, and second, the State provides active supervision of anticompetitive conduct undertaken by private actors.
Ticor Title, 504 U.S. at 631, 112 S.Ct. at 2175 (citing Midcal, 445 U.S. at 105, 100 S.Ct. at 943); see also Columbia Steel Casting Co., Inc. v. Portland Gen. Elec. Co., 60 F.3d 1390, 1395 (9th Cir.1995) (applying Midcal test).
The district court concluded that CMS’s claims were barred under this test. First, quoting Southern Motor Carriers Rate Conference, Inc. v. United States, 471 U.S. 48, 105 S.Ct. 1721, 85 L.Ed.2d 36 (1985), the court stated that in order to satisfy the first prong of the test, WNG must show only that Washington “clearly intends to displace competition in [the market for sale of natural gas] with a regulatory structure.” Id. at 64, 105 S.Ct. at 1730. The court concluded that this requirement was satisfied with respect to CMS’s Tariff 57 “monopoly leveraging” claim, because “[t]he Washington statutory scheme displaces pure competition in sale of • natural gas with a regulatory structure.” The court also concluded that Midcal’s “active supervision” requirement was satisfied with respect to this claim, because “the record amply demonstrates that the WUTC was actively involved in setting Tariff 57.” With respect to CMS’s off-tariff pricing claim, the court concluded that the first prong of the Midcal test was satisfied because “the WUTC is empowered to set and police pricing, including off-tariff violations.” (Emphasis added). The court did not explicitly discuss whether the “active supervision” requirement was satisfied with respect to CMS’s off-tariff pricing claim.
For a number of reasons, we conclude that the district court erred. First, although the court properly quoted Southern Motor Carriers, the court misapplied the first prong of the Midcal test. As noted above, the Supreme Court clarified in Ticor Title that this prong is satisfied if a.plaintiff establishes that “the State has articulated a clear and affirmative policy to. allow the anticompeti-tive conduct....” Ticor Title, 504 U.S. at 631, 112 S.Ct. at 2175 (emphasis added); see also id. at 627, 112 S.Ct. at 2174 (“state-action immunity ... permits anticompetitive conduct if authorized and supervised by state officials”) (emphasis added). Accordingly, the fact that Washington may have displaced competition in the market for sale of natural gas with a regulatory structure is not dispositive. Rather, the relevant question is whether the regulatory structure which has been adopted by the state has specifically authorized the conduct alleged to violate the Sherman Act. As applied to CMS’s off-tariff pricing claim, the first prong of the Midcal test will thus be satisfied only if WNG can establish either that (1) it did not participate in off-tariff pricing, or that (2) off-tariff pricing is legal under Washington’s regulatory scheme. Because this case arises on a motion to dismiss under Rule 12(b)(6), WNG cannot make the former showing here. In addition, CMS’s allegation that off-tariff pricing is illegal under Washington law is supported by a state statute which provides in relevant'part that'
[n]o gas company ... shall charge, demand, collect or receive a greater or less or different compensation for any service rendered or to be rendered than the rates and charges applicable to such service as specified in its schedule filed and in effect at the time
Wash. Rev.Code § 80.28.080 (1991). Moreover, the district court’s conclusion that the first prong of the Midcal test was satisfied because “the WUTC is empowered to set and police pricing, including off-tariff violations,” misses the' point. If, as the district court suggested, off-tariff pricing is a violation of Washington law, then WNG’s alleged conduct simply would not be protected by the state action immunity doctrine. Accordingly, the *943court erred in dismissing the off-tariff pricing claim on that ground.
The district court also erred in applying Midcal’s “active supervision” requirement to CMS’s Tariff 57 “monopoly leveraging” claim. In Ticor Title, the Supreme Court explained that
the purpose of the active supervision inquiry is not to determine whether the State has met some normative standard, such as efficiency, in its regulatory practices. Its purpose is to determine whether the State has exercised sufficient independent judgment and control so that the details of the rates or prices have been established as a product of deliberate state intervention, not simply by agreement among private parties. Much as in causation inquiries, the analysis asks whether the State has played a substantial role in determining the specifics of the economic policy. The question is not how well state regulation works but whether the anticompetitive scheme is the State’s own.
Ticor Title, 504 U.S. at 634-35, 112 S.Ct. at 2177. Furthermore, “[t]he mere potential for state supervision is not an adequate substitute for a decision by the State.” Id.
As should be clear from this discussion, the question of whether a state has “actively supervised” a state regulatory policy is a factual one which is inappropriately resolved in the context of a motion to dismiss. CMS has argued that the WUTC approved Tariff 57 only because WNG deliberately withheld key information from the WUTC which the WUTC repeatedly requested and which the WUTC admitted it needed in order to make an informed decision on whether to approve the tariff. The district court was required to accept CMS’s allegation as true. In addition, if CMS is able to prove its allegations, then the state action immunity doctrine almost certainly would not preclude its claim based on Tariff 57. See Ticor Title, 504 U.S. at 638, 112 S.Ct. at 2179 (finding that Montana had failed adequately to supervise its state regulatory policy because “a rate filing became effective despite the failure of the rating bureau to provide additional requested information”).
Accordingly, we hold that the court erred in dismissing CMS’s claims on the ground of state action immunity.6
Ill
WNG next urges us to affirm the district court on the ground that the “filed tariff’ or Keogh doctrine precludes CMS’s claims. The Keogh doctrine stems from Keogh v. Chicago & Northwestern Ry. Co., 260 U.S. 156, 43 S.Ct. 47, 67 L.Ed. 183 (1922).7
In Keogh, a private shipper claimed that rates which' had been submitted to and approved by the Interstate Commerce Commission (“ICC”) under the authority of the Interstate Commerce Act (“ICA”) had been fixed pursuant to a conspiracy prohibited by the Sherman Act. The shipper alleged that because of the conspiracy, the rates were “higher than the rates would have been, if competition had not been thus eliminated.” Id. at 160, 43 S.Ct. at 48. The shipper sought damages based on the difference between the rates established pursuant to the agreement and the rates previously in effect. Id.
The defendants argued that approval of the rates by the ICC conclusively established *944that the rates were “reasonable and nondiscriminatory,” id. at 161, 43 S.Ct. at 49, and the Court agreed. The Court noted that the ICC approval “did not foreclose the possibility that slightly lower rates would also have been within the zone of reasonableness that the Commission would also have found lawful,” Square D Co. v. Niagara Frontier Tariff Bureau, Inc., 476 U.S. 409, 415, 106 S.Ct. 1922, 1926, 90 L.Ed.2d 413 (1986) (describing Keogh), nor did it “require rejection of Keogh’s contention that the combination among the railroads violated the Sherman Act.” Id. Nonetheless, the Court held that Keogh could not “recover damages under [the Sherman Act] because he lost the benefits of rates still lower, which, but for the conspiracy, he would have enjoyed.” Keogh, 260 U.S. at 162, 43 S.Ct. at 49.
The Court offered four rationales for its holding. First, the Court noted that under the ICA, shippers who had been injured by illegally filed rates had a right to recover their actual damages in proceedings before the ICC. Id. at 162-63, 43 S.Ct. at 49-50. Because damages were available in an ICC proceeding, the Court doubted that Congress intended to provide a duplicative remedy in the Sherman Act. Id. Second, the Court concluded that granting an antitrust damages remedy only to those shippers who chose to sue would result in an arbitrary and discriminatory rate structure, because the successful litigant would receive a lower effective rate than other shippers. The Court concluded that this would conflict with “the paramount purpose of Congress-prevention of unjust discrimination.” Id. at 163, 43 S.Ct. at 49-50. Third, the Court noted that a damage recovery would require a showing that a hypothetical lower rate should and would have been adopted by the ICC. Whether or not the agency would have actually approved a different rate was a question best left to the agency itself, rather than the courts. Id. at 163-64, 43 S.Ct. at 49-50. Fourth and finally, the Court explained that individual shippers would not be injured by the illegal rate, because if the rate applied to all of the shipper’s competitors, the increase in the rate would simply be passed on to the shippers’ customers. Id. at 164-65, 43 S.Ct. at 50.
The Keogh doctrine has been vigorously criticized by a number of leading observers. Professor Hovenkamp, for one, has argued that “[n]one of these arguments had much to be said for them at the time they were originally made, and they are even less sensible today.” Federal Antitrust Policy, supra § 19.6 at 660; see also P. Areeda & H. Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application ¶ 227.2a, at 267-69 (Supp.1994) (arguing that modem legal developments have undermined rationales for Keogh doctrine). Similar considerations dominated Judge Friendly’s opinion for the Second Circuit in Square D Co. v. Niagara Frontier Tariff Bureau, Inc., 760 F.2d 1347 (2d Cir.), cert. granted, 474 U.S. 815, 106 S.Ct. 57, 88 L.Ed.2d 47 (1985), aff'd, 476 U.S. 409, 106 S.Ct. 1922, 90 L.Ed.2d 413 (1986). In an opinion which the Supreme Court subsequently praised as “characteristically thoughtful and incisive,” Square D Co., supra, 476 U.S. at 423, 106 S.Ct. at 1930, Judge Friendly effectively invited the Supreme Court to overturn Keogh. He, noted that each of the rationales which may have justified the Keogh, rule in 1922 had been undermined by subsequent legal developments. More specifically, he noted that under current law, antitrust damage remedies are available in spite of the fact that similar remedies are also available under other federal statutes. Square D, 760 F.2d at 1353-54. Accordingly, the Court’s earlier conclusion that Congress did not intend to provide for duplicative remedies seemed mistaken. Second, the Court’s concerns about possible rate discrimination which might result if only certain shippers decided to sue was undermined by the emergence of modern class actions. Id. at 1352. In addition, Judge Friendly explained that “the fear that recovery of antitrust damages would operate ‘like a rebate’ to produce discrimination among shippers' ignores the difference between a carrier’s passing money under the table to a favored shipper and a court’s deciding that a shipper had suffered antitrust injury.” Id.
Third, Judge Friendly noted that concern that “an antitrust action would require the carrier to prove that a hypothetical lower rate conformed to the requirements of the *945ICA” was no longer compelling in light of the fact that “[t]he process in which the ICC would have to engage if a court requested it to determine whether lower rates procured by competition would be reasonable and nondiscriminatory does not differ in essence from that which it must undertake” in other antitrust actions. Id. Fourth and finally, Judge Friendly noted that the Keogh Court’s concern that “‘no court or jury could say that, if the rate had been lower, [the shipper] would have enjoyed the difference,’ since the benefit ‘might have gone to his customers,’ ” id. at 1353 (quoting Keogh, 260 U.S. at 165, 43 S.Ct. at 50), was no longer valid in light of subsequent decisions in which the Court had “rejected the argument that a plaintiff cannot recover damages it was able to pass on to its customers in the antitrust context.” Id. (citations omitted).
The Supreme Court ultimately declined Judge Friendly’s invitation to overrule Keogh. See Square D, 476 U.S. at 417-24, 106 S.Ct. at 1927-31.8 However, the Court did not directly challenge the validity of his observations. Instead, the Court *‘assum[ed] that petitioners are correct in arguing that the Keogh decision was unwise as a matter of policy,” id. at 420, 106 S.Ct. at 1928, but nonetheless affirmed on grounds of stare de-cisis. The Court explained that
the developments in the six decades since Keogh was decided are insufficient to overcome the strong presumption of continued validity that adheres in the judicial interpretation of a statute. As Justice Bran-déis himself observed, a decade after his Keogh decision, in commenting on the presumption of stability in statutory interpretation: “Stare decisis is usually the wise policy because in most matters, it is more important that the applicable rule of law be settled than that it be settled right._” We are especially reluctant to reject this presumption in an area that has seen careful, intense, and sustained congressional attention. If there is to be an overruling of the Keogh rule, it must come from Congress, rather than from this Court.
Id. at 424, 106 S.Ct. at 1930-31 (footnotes omitted) (emphasis added). Finally, the Court offered a relatively narrow interpretation of the Keogh rule: “Keogh simply held that an award of treble damages is not an available remedy for a private shipper claim-' ing that the rate submitted to, and approved by, the ICC was the product of an antitrust violation.” Id. at 422, 106 S.Ct. at 1929.9
The fact that Square D’s endorsement of Keogh was only lukewarm is of particular importance here, because WNG is essentially asking us to extend Keogh to preclude rate-related suits brought by competitors, as opposed to customers, of regulated' entities. The weight of authority, and the better-reasoned authority, supports CMS’s argument that Keogh simply does not apply to such suits. See, e.g., Federal Antitrust Policy § 19.6, at 660 (“Most courts correctly hold that Keogh does not apply to competitor suits-for example, suits challenging filed rates as predatory pricing.”); Antitrust Law ¶ 227.2c, at 270 (“Many courts have allowed antitrust damage actions by actual or potential competitors against allegedly predatory *946filed rates.”)- As Professor Hovenkamp has explained:
First, 'of the -original rationales given for Keogh only the third, the primacy of agency rate determination, would apply to a competitor suit. Indeed, even that argument is weak because the agency generally scrutinizes rates to see if they are too high, making sure that cost figures are justified. It much less frequently scrutinizes tariff requests to see if the requested rates are too low. Second, a doctrine as indefensible as Keogh should be narrowly tailored.
Federal Antitrust Policy § 19.6 at 660.
The Second, Third, and Eighth Circuits have all held that Keogh does not preclude rate-related suits brought by competitors of regulated entities. See In re Lower Lake Erie Iron Ore Antitrust Litigation, 998 F.2d 1144 (3rd Cir.1993), cert. dismissed, 510 U.S. 1021, 114 S.Ct. 625, 126 L.Ed.2d 589 (1993); City of Kirkwood v. Union Elec. Co., 671 F.2d 1173, 1179 (8th Cir.1982), cert. denied, 459 U.S. 1170, 103 S.Ct. 814, 74 L.Ed.2d 1013 (1983); City of Groton v. Connecticut Light & Power Co., 662 F.2d 921, 929 (2d Cir.1981) (“an anticompetitive practice embodied in a tariff may violate the antitrust laws if it ... impacts upon competitors as opposed to customers”)’; Essential Communications Sys., Inc. v. American Tel. & Tel. Co., 610 F.2d 1114, 1121 (3d Cir.1979).
In City of Kirkwood, the Eighth Circuit explained that Keogh does not apply to competitor suits because “[a] rule formulated to ensure uniformity of rates as between customers should not give an unfair advantage to a utility in its dealings with competitors.” City of Kirkwood, 671 F.2d at 1179. Similarly, in Essential Communications, the Third Circuit explained that
In this case the plaintiff is not suing in the capacity of a customer for communication services. Essential seeks recovery for injury to its business or property from actions taken by the-defendants in formulating a tariff, and in rendering customer services. The [defendant] will not be asked to disgorge to any customers any revenues derived under the filed tariff. Indeed, it can continue to collect those revenues until a new tariff is filed. There is no policy conflict, actual or potential, therefore, between the Section 4 Clayton Act remedy and the antidiscrimination purposes of the filed tariff rule.
Essential Communications, 610 F.2d at 1122 (emphasis added). Finally, one district court has persuasively observed that
[although the Supreme Court’s ruling in Square D saved Keogh from extinction, it did not breathe a new expansive energy into the doctrine. The decision represents simply an unwillingness to deliver a coup de grace to a weak and forcefully criticized doctrine because that function might be more appropriately carried out by Congress. That holding does not justify expansion of the doctrine to nullify competitor suits.
Capital Freight Serv., Inc. v. Trailer Marine Transp. Corp., 704 F.Supp. 1190, 1195 (S.D.N.Y.1988) (Leval, J.); see also Frontier Enter. v. Amador Stage Lines, Inc., 624 F.Supp. 137, 143 (E.D.Cal.1985) (declining to apply Keogh in suit involving competitor).
The Sixth Circuit is the only circuit which has concluded that Keogh bars rate-related claims brought by competitors. See Pinney Dock & Transport Co. v. Penn Central Corp., 838 F.2d 1445 (6th Cir.), cert. denied, 488 U.S. 880, 109 S.Ct. 196, 102 L.Ed.2d 166 (1988).10 However, we are not persuaded by the Sixth Circuit’s analysis. In Pinney Dock, the court conceded that “the anti-discrimination arguments behind the Keogh doctrine lose their force in competitor lawsuits such as this.” Id. at 1457. The court nonetheless concluded that
[w]hen the ICC approves a rate ... it necessarily takes an anticompetitive action. This action is justified by the [Interstate Commerce Act] because the statute assumes that the pro-competition policies of the antitrust laws have been taken into account but also assumes that the [Inter*947state Commerce Commission] possesses and ought to have the power to override those policies in order to further national transportation policy as expressed in the Act. Rates must not only protect against overcharging captive customers but must also keep in mind the economic costs of delivery of the service. Regulation of one aspect inevitably begets regulation of the other. Thus, the ICC is the sole source of the rights not only of shippers, but of the entire public, including competitors.
Id. at 1457 (emphasis added).
As has been noted elsewhere, the Sixth Circuit’s reasoning is somewhat obscure. Indeed, Judge Leval cogently observed in Capital Freight,
the Sixth Circuit’s argument that ICC review “assumes that the pro-competition policies of the antitrust laws have been taken into account,” 838 F.2d at 1457, is somewhat inconsistent with the Supreme Court’s statement in Georgia v. Pennsylvania Railroad Co., 324 U.S. 439, 460, 65 S.Ct. 716, 727 [89 L.Ed. 1051 (1945)], that “the Interstate Commerce Act does not provide remedies for the correction of all the abuses of ratemaking which might constitute violations of the anti-trust laws,” and is at odds with the Second Circuit’s assertion that “an anti-competitive practice embodied in a tariff may violate the antitrust laws if it ... impacts upon competitors as opposed to customers.”
704 F.Supp. at 1195-96 (quoting City of Groton, 662 F.2d at 929). We agree with this conclusion. In addition, even if we were to agree with Pinney Dock’s assumption that the ICC has necessarily taken “the pro-competition policies of the antitrust laws” into account when calculating its rates, we fail to see how that justifies a conclusion that a competitor may not sue under the Sherman Act based on an allegation that the rates which were adopted were adopted in part because of an antitrust violation on the part of the defendant. Cf. Otter Tail Power Co. v. United States, 410 U.S. 366, 372, 93 S.Ct. 1022, 1027, 35 L.Ed.2d 359 (1973) (“Activities which come under the jurisdiction of a regulatory agency nevertheless may be subject to scrutiny under the antitrust laws.”); City of Long Beach v. Standard Oil Co. of California, 872 F.2d 1401, 1408 (9th Cir.1989) (holding that existence of federal price controls did not absolve companies of antitrust liability, because “if the price ceilings were based on prices set artificially low as a result of an unlawful conspiracy, liability should still exist”), as amended on denial of reh’g, 886 F.2d 246 (9th Cir.1989), cert. denied, 493 U.S. 1076, 110 S.Ct. 1126, 107 L.Ed.2d 1032 (1990). In addition, even if such an assumption is appropriate with respect to the Interstate Commerce Commission, we question whether it would be appropriate to extend a similar assumption to cases involving the plethora of state agencies which approve commercial tariffs of a variety of regulated enterprises. .
Nor are we persuaded by the Sixth Circuit’s reliance on Georgia v. Pennsylvania Railway Co., 324 U.S. 439, 65 S.Ct. 716, 89 L.Ed. 1051 (1945), in support of its conclusion that Keogh precludes competitor suits. It is true that in Georgia, the Supreme Court applied Keogh in a case in which the State of Georgia was suing in its dual capacity as a customer and a competitor of the defendant. See Pinney Dock, 838 F.2d at 1457 n. 12 (citing Georgia); see also Lifschultz Fast Freight, Inc. v. Consolidated Freightways Corp. of Delaware, 805 F.Supp. 1277, 1295 (D.S.C.1992) (citing Georgia in support of its conclusion that Keogh applies in competitor suits), aff'd, 998 F.2d 1009 (4th Cir. July 6, 1993) (unpublished disposition), cert. denied, 510 U.S. 993, 114 S.Ct. 553, 126 L.Ed.2d 454 (1993). However, the Third Circuit has persuasively explained that in Georgia, the state of Georgia
sued both in its parens patriae capacity and as owner of a railroad and other public institutions. Although Georgia’s ownership of a railroad placed it in competition with the defendants, the [Court] clearly focused on Georgia’s parens patriae claim and treated the injury to the state as proprietor merely as “make weight.” ... [I]t is obvious that in rendering its decision the Court’s focus was on Georgia’s claims as a customer. We, therefore, do not construe Georgia as Supreme Court precedent to apply Keogh to bar competitor suits.
*948 In re Lower Lake Erie Iron Ore, 998 F.2d at 1161-62 n. 6 (emphasis added).. We agree. In addition, although not specifically discussed by the Third Circuit, we also note that the court’s conclusion is supported by the fact that in Square D, supra, Judge Friendly distinguished certain cases on the grounds that “[a]ll of these cases, moreover, were suits by competitors rather than users of the service, so that many of the considerations relied upon in Keogh would not apply.” Square D, 760 F.2d at 1356 (citing Essential Communications, 610 F.2d at 1122) (emphasis added). Significantly, the Supreme Court did not take issue with this observation when it affirmed Judge Friendly’s opinion. , .
, Our conclusion that Keogh does not apply in competitor suits also finds some support in our decision in Barnes v. Arden Mayfair, Inc., 759 F.2d 676 (9th Cir.1985). In Barnes, two companies which had developed new methods of pasteurizing and packing milk sued two Alaskan dairies and a shipping company under the Sherman Act. The plaintiffs alleged that the dairies had undertaken various anticompetitive tactics to keep the plaintiffs out of the Alaskan dairy market. In particular, the plaintiffs alleged that the dairies had convinced the shipping company to help them keep the plaintiffs out of Alaska by illegally manipulating the tariff rates which the carrier charged for shipping milk to Alaska. The plaintiffs also alleged that an employee of the shipping company had discouraged a potential purchaser of one of the plaintiffs distributorships from entering into an agreement with the plaintiffs.
The defendants argued that the plaintiffs’ complaint was barred under Keogh. We rejected this argument by noting that
[the plaintiffs] do not seek damages solely for an illegally-set price, as did the plaintiffs in Keogh. Rather, they claim a conspiracy to destroy their businesses through a broad range of anticompetitive behavior. The district court observed that [the plaintiffs] did not allege merely that illegal rates were set, but in addition that [the defendants] undertook further alleged anticompetitive acts by attempting to prevent a prospective distributor from entering into an agreement with [one of the plaintiffs]. Because such activities would be beyond the scope of the ICC’s jurisdiction, the antitrust action was not subject to dismissal on this ground.
Id. at 679 (emphasis added).
It is clear from this passage that we did not actually hold that a competitor may bring an action for damages based on “rate-related” activity; instead, we emphasized that the plaintiffs in that case had alleged that the defendants also participated in other, “non-rate related” conduct not covered by Keogh. However, our decision to allow CMS’s claims to proceed is consistent with our holding today that competitor suits are not barred by Keogh.
Finally, we note that our conclusion that Keogh should not be extended to preclude antitrust damage actions brought by competitors is also consistent with the Supreme Court’s observation that “exemptions from the antitrust laws are strictly construed and strongly disfavored,” Square D, 476 U.S. at 421, 106 S.Ct. at 1929, and with our prior observation that “[i]mplied repeals of the antitrust statutes are disfavored, and require a convincing showing that the antitrust laws and the regulatory program cannot coexist.” City of Long Beach, 872 F.2d at 1409.
Accordingly, because the rationales offered for the Keogh doctrine do not justify extending that doctrine to preclude rate-based damages actions brought by competitors of regulated entities, we decline to apply the doctrine here.
IV
WNG next argues that we should affirm the district court on the basis of the primary jurisdiction doctrine.11 Specifically, WNG argues we should decline to exercise jurisdiction over this case and should instead transfer the case to the WUTC for resolution *949of such “threshold” questions as whether WNG did in fact violate Tariff 87.
In United States v. General Dynamics Corp., 828 F.2d 1356 (9th Cir.1987), we. explained that:
The doctrine of primary jurisdiction operates as follows: When there is a basis for judicial action, independent of agency proceedings, courts may route the threshold decision as to certain issues to the agency, charged with primary responsibility for governmental supervision or control of the particular industry or activity involved. The doctrine applies when protection of the integrity of a regulatory scheme dictates preliminary resort to the agency which administers the scheme.
Id. at 1362 (internal quotation marks and citations omitted) (emphasis added); see also Mt. Hood Stages, Inc. v. Greyhound Corp., 616 F.2d 394, 403 (9th Cir.), cert. denied, 449 U.S. 831, 101 S.Ct. 99, 66 L.Ed.2d 36 (1980) (applying doctrine). We further explained that
[tjhere are four factors uniformly present in eases where the doctrine properly is invoked: (1) the need to resolve an issue that (2) has been placed by Congress within the jurisdiction of an administrative body having regulatory authority (3) pursuant to a statute that subjects an industry or activity to a comprehensive regulatory scheme that (4) requires expertise or uniformity in administration.
General Dynamics, 828 F.2d at 1362 (citations omitted) (emphasis added).
WNG’s argument that we should apply this doctrine in this appeal is misplaced. Because this case arises in the context of a motion to dismiss under Rule 12(b)(6), we must accept as true CMS’s allegation that WNG violated Tariff 87. For that reason, the “threshold decision” which WNG would have us refer to the WUTC must necessarily be resolved in favor of CMS. Accordingly, the primary jurisdiction doctrine is simply inapplicable here.12
V
Finally, WNG argues that we should affirm the district court on the alternate ground that CMS has failed to allege the elements of a Sherman Act claim.
Section 2 of the Sherman Act provides: “Every person who shall monopolize, or attempt to monopolize, or combine or conspire with any other persons, to monopolize any part of the trade or commerce ... [commits a felony].” 15 U.S.C. § 2 (1994). In order to state a claim for monopolization under this provision, a plaintiff must prove that: (1) the defendant possesses monopoly power in the relevant market; (2) the defendant has willfully acquired or maintained that power; and (3) the defendant’s conduct has caused antitrust injury. SmileCare Dental Group v. Delta Dental Plan of California, Inc., 88 F.3d 780, 783 (9th Cir.1996) (citations omitted).13 In order to state a claim for *950attempted monopolization, a plaintiff must prove: (1) specific intent to control prices or destroy competition; (2) predatory or anti-competitive conduct to accomplish the monopolization; (3) dangerous probability of success; and (4) causal antitrust injury. Id. (citations omitted). In order to survive a motion to dismiss under Rule 12(b)(6), an antitrust complaint “need only allege sufficient facts from which the court can discern the elements of an injury resulting from an act forbidden by the antitrust laws.” Newman v. Universal Pictures, 813 F.2d 1519, 1522 (9th Cir.1987), cert. denied, 486 U.S. 1059, 108 S.Ct. 2831, 100 L.Ed.2d 931 (1988). CMS’s complaint should not be dismissed unless it appears beyond doubt that CMS can prove no set of facts in support of its claim which would entitle it to relief. SmileCare Dental Group, 88 F.3d at 783.
WNG raises three specific challenges to CMS’s complaint. We address each in turn.
A
WNG first challenges the sufficiency of CMS’s allegation that WNG possesses monopoly power.14 WNG notes that the Supreme Court has defined monopoly power as the power to “control prices or exclude competition,” United States v. Grinnell Corp., 384 U.S. 563, 571, 86 S.Ct. 1698, 1704, 16 L.Ed.2d 778 (1966) (citation omitted), and asserts that because it cannot set its own rates it meets neither of these' standards. WNG also notes that market share standing alone does not establish monopoly power in a regulated industry, and argues that CMS’s complaint is defective because the only factual allegation which CMS offers on the monopoly power issue is its allegation that WNG has a 90 percent market share.
WNG’s first argument effectively boils down to an assertion that it cannot possess monopoly power because it is a regulated entity. However, the mere fact that WNG is a regulated utility does not preclude a finding that it retains monopoly power. See, e.g., Federal Antitrust Policy, supra § 19.5, at 658 (discussing “market power” offenses in regulated markets).15 Moreover, CMS has alleged, in addition to its allegation that WNG has a 90 percent market share, that (1) WNG has charged rates other than those mandated by its tariffs, of, in other words, that WNG retains the ability to control its prices, and (2) that WNG’s off-tariff pricing has “made it impossible for others to compete for the sale of Gas,” or, in other words, that WNG’s practices have excluded competition. These allegations are sufficient to survive .a motion to dismiss.
With respect to WNG’s second argument, it is true that market share standing alone does not automatically equate to monopoly power. See, e.g., SmileCare Dental Group, 88 F.3d at 783; Rebel Oil Co., Inc. v. Atlantic Richfield Co., 51 F.3d 1421, 1439 (9th Cir.1995) (offering comprehensive analysis of market power and noting that “[a] mere showing of substantial or even dominant market share alone cannot establish market power sufficient to carry out a predatory scheme. The plaintiff must show that new rivals are barred from entering the market and show that existing competitors lack the capacity to expand their output to challenge the predator’s high price.”), cert. denied, — U.S. -, 116 S.Ct. 515, 133 L.Ed.2d 424 (1995). However, as noted above, CMS’s allegation that WNG possesses monopoly power does not rest entirely on its market share allegation. Moreover, in Hunt-Wesson Foods, Inc. v. Ragu Foods, Inc., 627 F.2d 919 (9th Cir.1980), cert. denied, 450 U.S. 921, 101 S.Ct. 1369, 67 L.Ed.2d 348 (1981) we held that
[w]hile market share is just the starting point for assessing market power, we think that market share, at least above some level, could support a finding of market power in the absence of contrary evidence. Where such an inference is not implausi *951 ble on its face, an allegation of a specific market share is sufficient, as a matter of pleading, to withstand a motion for dismissal. With nothing but Hunt’s complaint before us, we cannot say that allegations that Ragu had a 65 per cent market share, and that the share was increasing, could not under any market conditions provide a basis for inferring the requisite market power.
Id. at 925 (emphasis added).
In sum, while CMS’s complaint perhaps could have provided more detailed allegations on this issue, we are not prepared to hold, based on CMS’s allegations that (1) WNG has a 90 percent market share in the relevant market, (2) WNG has been able to charge off-tariff prices, and (3) WNG’s alleged off-tariff pricing has excluded competition from the market, that CMS can prove no set of facts establishing that WNG has monopoly power. CMS’s allegations are sufficient to withstand a motion to dismiss. Cf. Rebel Oil, 51 F.3d at 1438 (“ARCO’s market share of 44 percent is sufficient as a matter of law to support a finding of market power, if entry barriers are high and competitors are unable to expand their output in response to supra-competitive pricing.”).
B
Second, WNG argues that CMS’s “monopoly leveraging” theoiy was rejected by this court in Alaska Airlines, Inc. v. United Airlines, Inc., 948 F.2d 536, 549 (9th Cir.1991), cert. denied, 503 U.S. 977, 112 S.Ct. 1603, 118 L.Ed.2d 316 (1992), and that CMS therefore cannot proceed on its theory that WNG has used its monopoly over gas delivery services in an attempt to monopolize the market for gas sales.
The dispute between the parties on this issue turns primarily on the definition of “monopoly leveraging.” In Alaska Airlines, we did in fact reject the Second Circuit’s holding in Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir.1979), cert. denied, 444 U.S. 1093, 100 S.Ct. 1061, 62 L.Ed.2d 783 (1980) that “a single firm may be liable for monopoly leveraging, even in the absence of a threat that the ‘leveraged’ market will be monopolized.....” Alaska Airlines, 948 F.2d at 546 (internal quotation marks and citations omitted) (emphasis added). We noted that
[i]n contrast to the traditional actions for monopoly and attempted monopoly, Berkey Photo’s “monopoly leveraging” doctrine has only two rather loose elements: 1) there must be monopoly power in some market, and 2) such power must be “exercise[d] ... to the detriment of competition” in another market. The Berkey Photo court slighted the functions performed by the elements of the traditional Section 2 offenses, stating: “There is no reason to allow the exercise of [monopoly] power to the detriment of competition, in either the controlled market or any other. That the competition in the leveraged market may not be destroyed but merely distorted does not make it more palatable.”
Id. at 547 (quoting Berkey Photo, 603 F.2d at 275). Accordingly, we rejected Berkey Photo’s definition of “monopoly leveraging” as an independent theory of liability under Section 2.16
It is clear from our analysis, however, that to the extent that “monopoly leveraging” is defined as an attempt to use monopoly power in one market to monopolize another market, this theory remains a viable theory under Section 2. For example, we explained that “[e'jven in the two-market situation, a plaintiff cannot establish a violation of Section 2 without proving that the defendant used its monopoly power in one market to obtain, or attempt to attain, a monopoly in the downstream, or leveraged, market.” Id. (emphasis added). We also explained that
the elements of the established actions, for “monopolization” and “attempted monopolization” are vital to differentiate between efficient and natural monopolies on the one hand, and unlawful monopolies on the other. Berkey Photo’s monopoly leveraging doctrine fails to differentiate properly *952among monopolies. The anticompetitive dangers that implicate the Sherman Act are not present when a monopolist has a lawful monopoly in one market and uses its power to gain a competitive advantage in the second market. By definition, the monopolist has failed to gain, or attempt to gain, a monopoly in the second market. Thus, such activity fails to meet the second element necessary to establish a violation of Section 2. Unless the monopolist uses its power in the first market to acquire and maintain a monopoly in the second market, or to attempt to do so, there is no Section 2 violation.
Id. at 548 (emphasis added). Finally, we concluded that “[i]f there is a dangerous probability that a monopoly will be created by leveraging conduct, then the conduct will be reached under the doctrine of attempted monopoly.” Id. at 549 (emphasis added).
It is clear from our opinion in Alaska Airlines that to the extent that CMS seeks to proceed on the basis of the “monopoly leveraging” theory adopted in Berkey Photo, it may not proceed. However, CMS has alleged that WNG has used its monopoly power in the gas delivery market in an attempt to monopolize the market for gas sales. Accordingly, CMS has alleged conduct which may be reached “under the doctrine of attempted monopoly;” Alaska Airlines, 948 F.2d at 549, and it may proceed on this theory. We emphasize, however, that under this theory, CMS must establish each of the elements normally required to prove an attempted monopolization claim under section 2 of the Sherman Act. See, e.g., SmileCare Dental Group, supra, 88 F.3d at 783.
C
Third, WNG challenges CMS’s attempted monopolization claim based on WNG’s alleged off-tariff pricing. WNG argues that CMS’s off-tariff pricing claim is in effect a claim for predatory pricing, and argues that because CMS has failed to allege that WNG’s prices were below its average total costs of production, CMS’s claim necessarily fails; CMS responds by arguing that in Farley Transp. Co. Inc. v. Santa Fe Trail Transp. Co., 786 F.2d 1342 (9th Cir.1985), and Western Concrete Structures Co., Inc. v. Mitsui & Co. (U.S.A.), Inc., 760 F.2d 1013 (9th Cir.), cert. denied, 474 U.S. 903, 106 S.Ct. 230, 88 L.Ed.2d 229 (1985), this court held that off-tariff pricing alone may constitute., a violation of the Sherman Act, independent of any allegation of predatory pricing.17 The parties’ arguments require us to examine Farley and Western Concrete.
In Western Concrete, the plaintiff and one of the defendants (“VSL”) were competitors in the business of post-tensioning. The court explained that post-tensioning “is a construction process ... that involves the stretching of steel strand tendons within concrete slabs or girders.” Western Concrete, 760 F.2d at 1015. Because the cost of steel strand represented more than half of the companies’ direct costs, the price of steel strand significantly affected the companies’ respective competitive positions. Id. Under regulations promulgated by the United States Treasury Department, importation of steel strand below certain “trigger prices” was illegal. The plaintiff alleged that the defendants had conspired to import steel strand at prices below the trigger prices, and had done so in order to allow VSL to monopolize the industry. Id. Importantly, the plaintiff “expressly disclaimed” that it was proceeding on a predatory pricing theory. Id. Instead, the plaintiff alleged that the defendants had imported steel at prices below the legal levels, knowing that the plaintiff could not legally match those rates', and had done so in order to obtain an illegal competitive advantage and monopolize the market.
The district court dismissed the plaintiffs claim for failure to state a claim, concluding that “price cutting does not violate section 1 [of the Sherman Act], unless it is ‘predatory’ in the sense that it is below cost.” Id. at *9531016. We reversed. In doing so, we effectively treated the plaintiffs claim as one alleging predatory pricing. However, relying in part on our decision in Transamerica Computer Co., Inc. v. International Bus. Mach. Corp., 698 F.2d 1377 (9th Cir.), cert. denied, 464 U.S. 955, 104 S.Ct. 370, 78 L.Ed.2d 329 (1983), we concluded that it was not necessary for the plaintiff to establish that the defendant’s prices were below its costs. We noted that in Tmnsamerica Computer, we had stated that “this court has ... recognized that prices exceeding average total cost might nevertheless be predatory in some circumstances.” Tmnsamerica Computer, 698 F.2d at 1387. Relying in part on this exception, we held that the plaintiff had stated a valid claim. Judge Sneed succinctly explained the legal theory which we found to be adequate in that case:
The improper conduct alleged in this case is a conspiracy to sell at prices below those that other competitors, acting within the law, could offer. This is no more and no less than a form of predatory pricing. The only difference between the conduct alleged in this case and ordinary predatory pricing is that here it is the law, rather than costs, that prevents the defendant’s competitors from matching the defendant’s price.
Western Concrete, 760 F.2d at 1020 (Sneed, J., concurring in part and dissenting in part) (emphasis added).18
We applied the majority’s and Judge Sneed’s reasoning in Western Concrete to Farley. In Farley, the plaintiff sued two of its competitors, a trucking company and a railroad company, under Sections 1 and 2 of the Sherman Act. Farley, 786 F.2d at 1344. The defendants and plaintiff were each subject to tariff schedules approved by the Interstate Commerce Commission through a regional tariff bureau. Id. Under the Interstate Commerce Act, it was illegal for the parties to charge any rates not authorized by the tariff schedules. The plaintiff argued that the defendants had conspired to violate the tariffs by providing false information concerning their cargos to the tariff bureau, and had thereby obtained lower rates than those which the plaintiff could legally offer to its customers. The plaintiff alleged that the violation of the tariff illegally deprived it of business, and distorted the structure of the market. Id. at 1347.
As Judge Sneed had done with respect to the claim in Western Concrete, we noted that the plaintiffs theory in Farley “closely resembles predatory pricing.” Id. We also observed that “[u]nder ordinary circumstances, Farley’s theory would be difficult to sustain in light of the fact that the tariff for the Santa Fe Plan V was above average total cost.” Id. However, relying on the legal theory outlined in Judge Sneed’s concurring opinion in Western Concrete, and noting that the facts in Farley were “virtually identical” to those in Western Concrete, we held that the plaintiff had stated a Sherman Act claim. Id. at 1348.
CMS argues that Western Concrete and Farley should be read broadly for the proposition that a plaintiff seeking recovery for attempted monopolization may satisfy its burden of showing “predatory or anticompet-itive conduct,” SmileCare, 88 F.3d at 783, merely by showing that the defendant has offered prices below those outlined in a legally binding tariff. We disagree.
Our decisions in Western Concrete and Farley proceeded from the premise that the plaintiffs in those cases were seeking recovery under a legal theory which amounted to a variation of predatory pricing. In each case, we noted that a predatory pricing claim normally may not proceed unless the plaintiff establishes that the defendant’s prices are below its costs. However, as Judge Sneed recognized, the rationale for the “below cost” rule is simply that a defendant who is setting its prices below its costs may legitimately be presumed to be doing so in an attempt to drive its competitors out of the market by charging economically inefficient prices which its competitors cannot match. Be*954cause the defendant’s prices are below its costs, the pricing may legitimately be deemed “predatory.” The rule which we adopted in Western Concrete and Farley simply acknowledged that in cases where both the plaintiff and defendant are regulated entities subject to tariff restrictions, a defendant may be able to achieve the same result by illegally charging prices below its tariff rate, knowing that its competitor cannot match those rates without similarly violating the law. Cf. Farley, 786 F.2d at 1350 (“[T]he purpose of the [defendant’s] scheme to falsify the weight and composition of cargo was to arrive at a rate lower than that possible under [the tariff which the plaintiff was obligated to follow].”). By charging prices below this rate, the defendant can force its competitors into a Hobson’s choice: either match the defendant’s prices and risk incurring legal sanctions, or fail to match those prices and run the risk of being underbid and, eventually, driven from the market. Accordingly, we held that in these particular, narrow set of circumstances, “below-tariff pricing may serve as a surrogate for below-cost pric-ing...." Coastal Transfer Co. v. Toyota Motor Sales, U.S.A., 833 F.2d 208, 212 (9th Cir.1987).
Accordingly; CMS’s broader reading of Western Concrete and Farley is erroneous. In order for CMS to establish, on the basis of the theory outlined in these cases, that WNG participated in “predatory or anticompetitive conduct to accomplish the monopolization,” SmileCare, 88 F.3d at 783, CMS would have to establish that (1) CMS and WNG are regulated entities in the same relevant market; (2) each is precluded by law from setting prices below those outlined in controlling tariffs; (3) WNG has illegally offered prices below the tariff prices; and (4) CMS cannot match WNG’s prices without similarly violating the law. Because CMS is not a regulated entity, it cannot establish a violation of section 2 based on Western Concrete and Farley. We therefore dismiss CMS’s off-tariff pricing claim outright.
VI
For the foregoing reasons, the district court’s order is REVERSED, and this ease is REMANDED for further proceedings not inconsistent with this opinion.
REVERSED and REMANDED.