GTE SOUTH, INCORPORATED, Plaintiff-Appellant, and United States of America, Intervenor-Plaintiff, v. Theodore V. MORRISON, Jr.; Hullihen W. Moore; I. Clinton Miller, in their official capacities as Commissioners of the Virginia State Corporation Commission; Cox Fibernet Commercial Services, Incorporated; AT & T Communications of Virginia, Incorporated; MCI Telecommunications Corporation; MCImetro Access Transmission Services of Virginia, Incorporated, Defendants-Appellees, and Attorney General of the Commonwealth of Virginia, Intervenor-Defendant, 199 F.3d 733

Affirmed by published opinion. Judge MICHAEL wrote the opinion, in which Judge HOWARD and Judge FRIEDMAN joined. OPINION MICHAEL, Circuit Judge: The Telecommunications Act of 1996, Pub.L. 104-104,110 Stat. 56, codified at 47 U.S.C. § 251 et seq. (sometimes, the Act), requires local telephone companies, heretofore monopolies, to make their facilities and services available to would-be competitors at negotiated prices or, if negotiations fail, at prices to be set in arbitration proceedings before state utility commissions. The Act gives federal district courts jurisdiction to review state commission determinations.

Here, the Virginia State Corporation Commission (the SCC) determined prices in arbitration proceedings brought by new entrants into Virginia’s local telephone markets, Cox Fiber-net Commercial Services, Inc. (Cox), MCI Telecommunications Corporation and MCImetro Access Transmission Services of Virginia, Inc. (collectively, MCI), and AT & T Communications of Virginia, Inc. (AT & T), against the incumbent company, GTE South Incorporated (GTE).

After the SCC arbitration GTE sued Cox, AT & T, MCI, and the SCC commissioners in district court alleging that the SCC’s pricing decisions failed to meet the requirements of the Act. The district court granted

summary judgment for the defendants (the new entrants and the commissioners), thus upholding the SCO’s arbitration decisions. GTE appeals, and we affirm. I. The breakup of AT & T in the early 1980s brought competition to the long distance telephone market. The local market, however, has been a different story. Until the passage of the 1996 Act, state utility commissions continued to regulate local telephone service as a natural monopoly. Commissions typically granted a single company, called a local exchange carrier (LEC), an exclusive franchise to provide telephone service in a designated area.

Under this protection the LEC built a local network — made up of elements such as loops (wires), switches, and transmission facilities — that connects telephones in the local calling area to each other and to long distance carriers. The 1996 Act brought sweeping changes. It ended the monopolies that incumbent LECs held over local telephone service by preempting state laws that had protected the LECs from competition. See 47 U.S.C. § 258. Congress recognized, however, that removing the legal barriers to entry would not be enough, given current technology, to make local telephone markets competitive.

In other words, it is economically impractical to duplicate the incumbent LEC’s local network infrastructure. To get around this problem, the Act allows potential competitors, called competing local exchange carriers (CLECs), to enter the local telephone market by using the incumbent LEC’s network or services in three ways. First, a CLEC may build its own network and “interconnect” with the network of an incumbent. See id. § 251(c)(2). Second, a CLEC may lease elements (loops, switches, etc.) of an incumbent LEC’s network “on an unbundled basis.” See id. § 251(c)(3).

Third, a CLEC may buy an incumbent LEC’s retail services “at wholesale rates” and then resell those services to customers under its (the CLEC’

s) brand. See id. § 251(c)(4). The Act details procedures for allowing a CLEC access to the incumbent LEC’s facilities and services. The CLEC first makes a request to the incumbent for interconnection or for access to its network or services. Thereafter, both parties must negotiate in good faith in an effort to reach agreement on terms and conditions (including price) of access. See id. §§ 251(c)(1), 252(a)(1). If negotiations fail' — it is hard to see how they would not — either party may petition the state utility commission to arbitrate open issues. See id. § 252(b). The terms imposed by the state commission in arbitration must “meet the requirements of

section 251 ... including the regulations prescribed by the [FCC] pursuant to

section 251.” Id. § 252(c)(1). The Act includes general standards for a state commission to use in arbitrating open price (or rate) issues. See id. §§ 251(c), 252(d). Finally, the Act authorizes any party aggrieved by the arbitration decision of a state commission to bring an action in federal district court to determine whether the arbitration decision “meets the requirements of’ §§ 251 and 252. See id. § 252(e)(6). We digress for a moment to discuss the issuance and status of the FCC rules.

The Act directed the FCC to “establish regulations to implement the requirements” of § 251, that is, the requirements to advance local competition. Id. § 251(d)(1). On August 8, 1996, the FCC issued an order and rules implementing the local competition provisions of the Act. See In the Matter of Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, First Report and Order, 11 F.C.C.R. 15499 (1996) (First Report and Order).

Included were rules to be used by state commissions in determining the price new entrants would be charged for interconnection, access to unbundled network elements, and retail services bought for resale. A number of interested parties, mainly incumbent LECs and state utility commissions, filed petitions for review in several circuits challenging the FCC’s rules, especially those relating to pricing. The petitions were consolidated and assigned to the Eighth Circuit by the panel on multi-district litigation. See 28 U.S.C. § 2112(a).

On September 27, 1996, three days before the FCC’s rules were scheduled to go into effect, the Eighth Circuit entered a temporary stay of the rules, see Iowa Utils. Bd. v. FCC, 96 F.3d 1116, 1118 (8th Cir.1996), and later entered a stay of the pricing rules pending final decision, see Iowa Utils. Bd. v. FCC, 109 F.3d 418, 427 (8th Cir.1996). Ultimately, in July 1997 the Eighth Circuit vacated the pricing rules, holding that the FCC lacked jurisdiction to promulgate them. See Iowa Utils. Bd. v. FCC, 120 F.3d 753, 800 (8th Cir.1997), aff'd in part, rev’d in part sub nom. AT & T Corp. v. Iowa Utils.

Bd., 525 U.S. 366, 119 S.Ct. 721, 142 L.Ed.2d 835 (1999). The Eighth Circuit said that §§ 252(c)(2) and (

d) of the Act give state utility commissions exclusive authority to make pricing decisions under §§ 251 and 252, thereby depriving the FCC of any rulemaking power in that area. See Iowa Utils. Bd., 120 F.3d at 793-800. The Supreme Court disagreed. On January 25, 1999, the Court held that the FCC had jurisdiction to issue rules implementing the pricing provisions of the Act. See AT & T Corp. v. Iowa Utils. Bd., 119 S.Ct. at 733 (“[T]he Commission has jurisdiction to design a pricing methodology”). In other words, the FCC is empowered to issue “rules to guide the state-commission judgments.” Id.

Although the Court validated FCC rule-making, the case was remanded for the Eighth Circuit to consider substantive challenges to the rules. See id. 119 S.Ct. at 738. Nevertheless, the pricing rules took effect when the Supreme Court reversed the Eighth Circuit’s judgment of vacatur. We now turn to this case. After the passage of the 1996 Act, several CLECs, including Cox, MCI, and AT & T, sought to enter the local telephone market in Virginia. Accordingly, the CLECs made separate requests to GTE for interconnection to GTE’s network, access to GTE’s network elements, and wholesale rates for GTE’s retail services.

After negotiations failed, each of the CLECs filed a § 252(b)(1) petition with the SCC asking it to arbitrate the unresolved issues in the negotiations with GTE. On October 16, 1996, the SCC consolidated the arbitration cases. By the time the SCC heard the matter, the Eighth Circuit had issued its stay of the FCC’s pricing rules. Thus, the SCC relied mainly on the Act in making pricing determinations in the arbitration. The standards from the Act that guided the SCC on the open pricing issues are as follows.

Prices (or rates) for interconnection and unbundled network elements must be “just, reasonable, and nondiscriminatory.” 47 U.S.C. §§ 251(c)(2)(D), 251(c)(3). Specifically, these prices must be “based on the cost (determined without reference to a rate-of-return or other rate-based proceeding) of providing the interconnection or network element,” and they “may include a reason-able profit.” Id. § 252(d)(1).

The wholesale rates that an incumbent LEC may charge a CLEC for retail services must be determined “on the basis of retail rates charged to subscribers for the telecommunications service requested, excluding the portion thereof attributable to any marketing, billing, collection, and other costs that will be avoided by the local exchange carrier.” Id. § 252(d)(3). In the SCC proceeding the parties disagreed fundamentally about the methodology that should be used in making price determinations under the Act.

For network element pricing, the CLECs proposed the Hatfield model, a forward-looking methodology for determining costs. It is a Total Element Long Run Incremental Cost (TELRIC) model. The Hatfield model starts from scratch and simulates a new telephone network for the relevant local service area. The model calculates (on a forward-looking basis) the direct and indirect costs of building and operating the elements of this simulated network, basically using the most economical and efficient configuration and technology available today.

For network element pricing, GTE proposed a much different model, the Market Determined-Efficient Component Pricing Rule (M-ECPR). M-ECPR is at bottom a methodology that allows recovery of historical costs. Although the M-ECPR model calls for estimating the costs of providing each network element on a forward-looking basis, the calculation does not end there. It adds components that allow the incumbent LEC to recover all of the actual costs it has incurred in constructing its network, including its historical costs reflected on its books.

At the SCC arbitration the parties also proposed different methods for determining wholesale prices for services to be resold. The dispute centered on how to calculate the costs that “will be avoided,” 47 U.S.C. § 252(d)(3), by an incumbent LEC when it provides retail services to a CLEC. The CLECs argued that the wholesale prices should be determined by assuming (for purposes of the calculation) that the incumbent LEC would leave the retail market altogether; thus, all marketing, billing, collection, and like costs would be deducted from the incumbent’s retail price.

GTE pointed out that it will not be leaving the retail market (its customer base will just be smaller), and it argued for the subtraction of only those costs that will be reduced by its loss of retail sales. On December 11, 1996, the SCC issued an order that in large measure adopted the CLECs’ proposals for pricing standards. The SCC generally approved the use of the Hatfield model for network element pricing in the interim before permanent ratemaking, but it adopted two modifications suggested by the SCC staff.

First, the input “fill factor” (the percentage of distribution plant that is actually in use) was decreased in response to concerns expressed by GTE. Second, the input for cost of capital was increased to 10.55 percent. On wholesale pricing, the SCC interpreted costs that “will be avoided” to mean “those costs that would be reasonably avoidable by a local exchange company furnishing only wholesale service.” These arbitration decisions were incorporated into “interconnection agreements” between GTE and the individual CLECs. The interconnection agreements were approved by the SCC in May 1997.

On June 30, 1997, GTE filed a complaint in federal court in the Eastern District of Virginia against the three CLECs (Cox, AT & T, and MCI) and the SCC commissioners, alleging (among other things) that the SCC violated the Act (1) by employing a forward-looking cost methodology to determine prices for network elements, (2) by adopting a pricing methodology that denied GTE its historic costs, and (3) by calculating wholesale prices on the assumption that GTE will get out of retailing entirely. Both sides cross-moved for

summary judgment, and the district court granted

summary judgment to the CLECs and the SCC commissioners, holding that the SCC’s pricing methodologies met the requirements of the Act. See GTE South, Inc. v. Morrison, 6 F.Supp.2d 517 (E.D.Va.1998). GTE appealed. After the Supreme Court in Iowa Utilities Board confirmed FCC rulemaking power with respect to the Act’s pricing provisions, we allowed the parties to re-brief this case.

GTE, the appellant, raises the following arguments in its revised brief: first, the FCC’s rules cannot be applied retroactively; second, if the rules apply, we cannot apply them in the first instance and a remand to the SCC would be necessary; third, we should exercise jurisdiction to invalidate the rules as contrary to the Act, but if we decide that we lack jurisdiction to review the rules, we should stay this appeal until the Eighth Circuit decides the substantive challenges to the rules; fourth, if we reach the merits and apply the rules, we should reverse the district court because the SCO’s pricing decisions do not comply with the rules.

We turn to these arguments. II. GTE contends that we may not use the FCC’s pricing rules to evaluate the SCO’s arbitration determinations because any application of the rules to this case would give them an impermissible retroactive effect. As we have mentioned, the Eighth Circuit stayed the pricing rules on September 27, 1996, shortly before they were slated to take effect on September 30, 1996.

The CLECs in this case began arbitration proceedings before the SCC in October 1996; in those proceedings, which concluded in May 1997 (while the stay was still in effect), the SCC relied mainly on the Act, and not the FCC rules. The Eighth Circuit’s stay expired on July 18, 1997, when that court vacated the pricing rules. See Iowa Utils. Bd., 120 F.3d at 820 (“Upon the filing of this opinion and order, the provisions of our stay order are deemed expired.”).

However, on January 25,1999, the Supreme Court in Iowa Utilities Board reversed the Eighth Circuit’s judgment vacating these rules, allowing them to take effect. While GTE concedes that Supreme Court decisions are usually retroactive and apply to cases pending on direct review, it urges us to except Iowa Utilities Board from the general rule. An exception is warranted, GTE argues, because in Iowa Utilities Board the Court breathed life into “substantive rules delineating new legal obligations that were stayed before they ever went into effect ... (and were then vacated).” Appellant’s Br. at 22.

Any present application of the pricing rules to the SCO’s arbitration decisions would therefore be impermissibly retroactive, according to GTE. Generally, “judicial construction of a statute is an authoritative statement of what the statute meant before as well as after the decision of the case giving rise to that construction.” Rivers v. Roadway Express, Inc., 511 U.S. 298, 312-13, 114 S.Ct. 1510, 128 L.Ed.2d 274 (1994). See also Harper v. Virginia Dep’t of Taxation, 509 U.S. 86, 97, 113 S.Ct. 2510, 125 L.Ed.2d 74 (1993) (holding that Supreme Court’s

interpretation of a law is controlling and “must be given full retroactive effect in all cases still open on direct review and as to all events, regardless of whether such events predate or postdate [the Court’s] announcement of the rule”). Thus, the Supreme Court’s determination that the FCC has jurisdiction to issue pricing rules would appear to compel the conclusion that the FCC always had such jurisdiction and that the rules apply as of the effective date originally scheduled. See, e.g., United States v.

Gonzalez-Sandoval, 894 F.2d 1043, 1052-53 (9th Cir.1990)(where underlying offense was committed during five months between Ninth Circuit’s invalidation of Sentencing Reform Act and Supreme Court’s decision overruling Ninth Circuit, criminal sentence had to conform to Sentencing Guidelines despite law of circuit at time crime was committed). GTE argues that this straightforward analysis is insufficient here only because the Eighth Circuit stayed the effective date of the pricing rules while it was deciding to vacate them.

Because the stay was a valid, if erroneous, order, the FCC rules did not go into effect until the Supreme Court reversed the Eighth Circuit. GTE’s retroactivity argument thus hinges on the point that the stay prevented the rules from taking effect both before and during the SCC arbitration. GTE’s emphasis on effective dates, however, overlooks the central concern of retroactivity analysis. A regulation “does not operate ‘retrospectively’ merely because it is applied in a case arising from conduct antedating the [regulation’s] enactment.” Landgraf v.

USI Film Prods., 511 U.S. 244, 269, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). Rather, a “court must ask whether the new provision attaches new legal consequences to events completed before its enactment.” Id. at 269-70, 114 S.Ct. 1483. The law with true retroactive effect is one that “takes away or impairs vested rights acquired under existing laws, or creates a new obligation, imposes a new duty, or attaches a new disability” to past transactions. Society for Propagation of the Gospel v. Wheeler, 22 F. Cas. 756, 767 (No. 13,156)(CC NH 1814)(Story, J.).

See also Landgraf, 511 U.S. at 269 n. 23, 114 S.Ct. 1483 (citing similar formulations). When we consider GTE’s argument in light of these principles, it becomes clear that application of the FCC’s pricing rules in this appeal would have no genuine retroactive effect. First, GTE cannot identify any vested right that it acquired during the Eighth Circuit’s stay. For example, that court’s stay and invalidation of the FCC’s rules on jurisdictional grounds did not give GTE a vested right to historical costs under its

interpretation of the Act, an

interpretation that was vigorously contested by the CLECs and the SCC staff. Nor could GTE reasonably rely on an

interpretation of the Act that the FCC, as we shall see, had already rejected in its (stayed) rules. Second, the SCC made price determinations that will govern future business relations among GTE and its competitors in the local telephone market. Thus, applying the FCC rules in these proceedings will not impose new obligations or duties with respect to past transactions. See Landgraf, 511 U.S. at 273, 114 S.Ct. 1483 (“When the intervening statute authorizes or affects the propriety of prospective relief, application of the new provision is not retroactive.”).

Finally, and perhaps most importantly, the core principle disfavoring retroactive application of laws — that “settled expectations should not be lightly disrupted,” Landgraf, 511 U.S. at 265, 114 S.Ct. 1483—is inapplicable here. GTE had ample notice of the First Report and Order (containing the rules) because it was published prior to the SCC arbitration, and GTE surely knew that the FCC’s authority to issue pricing rules might ultimately be upheld by the Supreme Court.

See id. at 270, 114 S.Ct. 1483 (observing that “familiar considerations of fair notice, reasonable rebanee, and settled expectations” should guide retroactivity analysis); cf. Gonzalez-Sandoval, 894 F.2d at 1053 (criminal defendant was on notice that Supreme Court might vacate Ninth Circuit decision invalidating Sentencing Guidelines). In sum, there are no retroactivity principles that prevent us from applying the FCC’s pricing rules in this appeal. III.

GTE next invokes the Chenery doctrine to argue that we “cannot affirm the SCC’s decisions on the theory that they comply with the FCC’s rules for the simple reason that the SCC did not apply those rules.” Appellant’s Br. at 47. If the rules are to be applied, GTE argues, we must remand for the SCC to apply them in the first instance. We disagree. In SEC v. Chenery Corp., 318 U.S. 80, 63 S.Ct. 454, 87 L.Ed. 626 (1943), the SEC issued an enforcement order that was based solely upon an inapplicable principle of equity.

The question before the Supreme Court was whether the order could be upheld on ' an alternative ground not relied upon by the agency, that is, the facts would have allowed the SEC to rely on its discretionary authority and “special administrative competence” to issue the order. The Court refused to affirm the order, holding that “an administrative order cannot be upheld unless the grounds upon which the agency acted in exercising its powers were those upon which its action can be sustained.” Id. at 95, 63 S.Ct. 454.

The Chenery Court qualified this broad statement by explaining that “a judicial judgment cannot be made to do service for an administrative judgment” when “an [administrative] order is valid only as a determination of policy or judgment which the agency alone is authorized to make and which it has not made.” Id. at 88, 63 S.Ct. 454 (emphasis added). Thus, when an agency bases its order on an unsupportable rationale, we cannot uphold the order on an alternative ground that would require us to exercise any discretionary judgment entrusted to the agency. See ICC v.

Brotherhood of Locomotive Eng’rs, 482 U.S. 270, 283, 107 S.Ct. 2360, 96 L.Ed.2d 222 (1987). We may, however, affirm an order on an alternative ground that is “within the power of [an] appellate court to formulate.” Chenery, 318 U.S. at 88, 63 S.Ct. 454. Specifically, if the administrative order reaches the correct result and can be sustained as a matter of law, we may affirm on the legal ground even though the agency relied on a different rationale. See Koyo Seiko Co. v. United States, 95 F.3d 1094, 1101 (Fed.Cir.1996).

These principles arising from Chenery must be applied in light of the reviewing authority Congress assigned to federal courts under the Telecommunications Act of 1996. We determine whether a state utility commission’s arbitration decision “meets the requirements” of §§ 251 and 252 of the Act. See 47 U.S.C. § 252(e)(6). Congress did not intend for the statutory language to be the only guide because it authorized the FCC to issue rules “to implement the requirements” of § 251, the

section relating to duties and terms of interconnection, unbundled access, wholesaling, etc. See id. § 251(d)(1). In the normal course, state commissions will be applying the FCC rules in the first instance. The SCC did not do that here because of the Eighth Circuit’s stay, and we must therefore decide whether we may rely on the rules in this unusual circumstance. In addition to implicating Chen-ery, this question also involves the standard of review we follow. As we indicate in

part V, infra, we review the SCC’s

interpretation of the Act de novo, and we do not accord any deference to the SCC’s

interpretation of the Act. Nor would we accord any deference to the SCC’s

interpretation of the FCC’s rules, if the SCC had applied them. Because the SCC’s

interpretation of the rules could not circumscribe our power of review, we may rely on the rules to the extent they assist us in making the legal determination whether the SCC’s decision meets the requirements of the Act. IV. GTE presses further to avoid application of the FCC’s pricing rules by arguing that their methodology is contrary to the Act. Subsumed within this challenge to the rules is GTE’s argument that the FCC’s construction of the Act would effect an unconstitutional taking.

In the event that we lack jurisdiction in this case to declare the rules invalid, GTE argues that we should stay this appeal until the Eighth Circuit issues its decision on their validity. Although we conclude that we lack jurisdiction to review the rules, we decline to issue a stay for reasons we discuss below. This case was commenced in district court under § 252(e)(6) of the Telecommunications Act and is before us on appeal from the judgment of the district court.

Under the Hobbs Act courts of appeals have “exclusive jurisdiction ... to determine the validity of ... all final orders” (including those relating to rulemaking) of the FCC. 28 U.S.C. § 2342(1). See also 47 U.S.C. § 402(a). That jurisdiction is invoked by filing a petition for direct review of an agency order in a court of appeals within sixty days of its issuance. See 28 U.S.C. §§ 2342, 2344. GTE attempts to avoid the Hobbs Act’s exclusive review mechanism by invoking § 252(e)(6). Because that

section gives district courts (and us on appeal) the authority to determine whether a state arbitration decision meets the requirements of the Act, GTE argues that we must necessarily have the authority to determine the validity of FCC rules if we apply them in evaluating an SCC decision. Thus, GTE argues that § 252(e)(6) implicitly grants to district courts (and to us on appeal) the jurisdiction that the Hobbs Act explicitly reserves to courts of appeals on petitions for review.

We disagree. “Generally, when jurisdiction to review administrative determinations is vested in the courts of appeals these specific, exclusive jurisdiction provisions preempt district court jurisdiction over related issues under other statutes.” Palumbo v. Waste Technologies Indus., 989 F.2d 156, 160-61 (4th Cir.1993)(quoting Connors v. Amax Coal Co., 858 F.2d 1226, 1231 (7th Cir.1988)). And a party cannot use an appeal from a district court to circumvent the Hobbs Act’s requirement that a challenge to an FCC order is subject only to direct review in a court of appeals. See, e.g., Missouri v.

United States, 109 F.3d 440, 442 (8th Cir.1997) (applying judicial review provisions of Clean Air Act). There is a good reason for this. The FCC is not a party to a § 252(e)(6) action, and surely Congress would not give a court the power to determine the validity of an agency’s rules when the agency itself is not a party. See City of Peoria v. General Elec. Cablevision Corp., 690 F.2d 116, 119 (7th Cir.1982)(“[T]he proper party defendant in the judicial review proceeding is the FCC, not a private company which ... may be indifferent to the validity of the rule.”).

Thus, the district court could not hear a challenge to the rules, see U.S. West Communications v. MFS Intelenet, Inc., 193 F.3d 1112, 1122-23 (9th Cir.1999), and neither can we in this appeal, id., 193 F.3d at 1120. We are also precluded from reviewing the validity of the FCC’s pricing rules under the statute governing consolidation of agency review proceedings, 28 U.S.C. § 2112(a). Various parties filed petitions for review of the FCC’s First Report and Order (issuing the pricing rules) in several courts of appeals. Pursuant to § 2112(

a) the multidistrict litigation panel consolidated the petitions and assigned the matter by lottery to the Eighth Circuit. See 28 U.S.C. § 2112(a)(3). That circuit- is now the sole forum for addressing challenges to the validity of the FCC’s rules. See id. § 2112(a)(5). This consolidation procedure for review of agency orders is in place “to avoid confusion and duplication by the courts” and “to prevent unseemly conflicts that could result should sister circuits take the initiative and issue conflicting decisions.” Westinghouse Elec. Corp. v.

NRC, 598 F.2d 759, 766-67 (3d Cir.1979)(internal quotation marks omitted). GTE is one of the parties attacking the substance of the rules in the Eighth Circuit proceeding. It cannot take a second bite at the apple and also challenge the rules in this circuit. Because we will not be addressing whether the FCC’s pricing rules are contrary to the Act, we proceed to GTE’s suggestion that we hold this appeal in abeyance until the Eighth Circuit has resolved that question. We realize that should we apply the rules and decide this case now, the Eighth Circuit may later invalidate some or all of the rules.

We therefore consider whether making a decision today will work any fundamental unfairness on GTE. We turn first to the interconnection agreements between GTE and the CLECs, which specifically provide for modification should the governing law or regulations change. As counsel for one of the CLECs conceded at oral argument, if the invalidation of a rule requires a change in prior rates, the parties will “settle up,” a regular occurrence in public utility ratemaking.

Thus, if the Eighth Circuit invalidates any rule on which we rely, GTE can recover any resulting shortfall in prices through the process of settling up. Moreover, our refusal to stay this appeal will have no immediate effect on GTE’s obligations. Since the SCC approval of the interconnection agreements, GTE has been required to lease its network elements and make services available at the prices set by the SCC arbitration. Even if we stayed this appeal, GTE would continue to make its network and services available at those prices.

Moreover, the prices under review here are interim, not permanent prices, and the interconnection agreements themselves expire on May 1, 2000. In these circumstances, issuing a stay now would serve little purpose. In addition, Congress intended that competition under the Telecommunications Act take root “as quickly as possible.” H.R.Rep. No. 104-204 at 89 (1995). Withholding our decision until the Eighth Circuit and (perhaps) the Supreme Court review the validity of the FCC’s rules could result in a considerable delay.

Issuing this opinion without delay is consistent with Congress’s express policy to bring about competition quickly in the local telephone markets. Finally, we reject GTE’s suggestion that our inability to review the rules before we apply them violates due process. First, GTE has not been denied judicial review on the validity of the rules, it has only been denied review in this forum. Clearly, Congress “may prescribe the procedures and conditions under which, and the courts in which, judicial review of administrative orders may be had.” City of Tacoma v.

Taxpayers of Tacoma, 357 U.S. 320, 336, 78 S.Ct. 1209, 2 L.Ed.2d 1345 (1958). Second, GTE is now obtaining judicial review of the rules in the Eighth Circuit, the forum selected according to Congress’s prescription in 28 U.S.C. § 2112(a). Third, when Congress has provided an exclusive mechanism for review of an agency rule, due process is not violated by enforcing that rule in another forum that cannot consider its validity. See Yakus v. United States, 321 U.S. 414, 432-34, 64 S.Ct. 660, 88 L.Ed. 834 (1944).

Contrary to GTE’s assertions, that review need not take place before the rule is enforced. “Where only property rights are involved, mere postponement of the judicial enquiry is not a denial of due process, if the opportunity given for the ultimate judicial determination of the liability is adequate.” Bowles v. Willingham, 321 U.S. 503, 520, 64 S.Ct. 641, 88 L.Ed. 892 (1944) (quoting Phillips v. Commissioner, 283 U.S. 589, 596-97, 51 S.Ct. 608, 75 L.Ed. 1289 (1931)). See also Yakus, 321 U.S. at 436, 64 S.Ct. 660; cf. Mathews v.

Eldridge, 424 U.S. 319, 339-49, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976)(upholding termination of Social Security benefits prior to evidentiary hearing). Here, if we apply the rules and the Eighth Circuit invalidates them, GTE can recover any underpayment by settling up. If it cannot settle up through negotiations, it can arbitrate before the SCC. If it is dissatisfied with the arbitration result, it can sue in district court under § 252(e)(6) of the Act.

In sum, because GTE can obtain any relief to which it is entitled through the combination of its properly filed petition for review in the Eighth Circuit and its remedies under the interconnection agreements and the Act, our application of the FCC’s pricing rules without considering their validity would not violate due process. We will therefore apply the rules. V. A. We are almost ready to discuss the merits, but we must first consider the applicable standards of review. We review de novo the district court’s grant of

summary judgment. See United States v. Leak, 123 F.3d 787, 791 (4th Cir.1997). As for the SCC’s determinations, § 252(e)(6) of the Act provides for “judicial review of the [state] Commission’s actions,” but it does not prescribe any standard for that review. Moreover, because the SCC is not an “agency” as defined by the Administrative Procedure Act, the standards provided by that act are not directly applicable. See 5 U.S.C. § 701(b)(l)(defining agency as an “authority of the Government of the United States”). Absent a statutory command, general standards for judicial review of agency action apply. A “state agency’s

interpretation of federal statutes is not entitled to the deference afforded a federal agency’s

interpretation of its own statutes under Chevron [U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)].” Orthopaedic Hasp. v. Belshe, 103 F.3d 1491, 1495-96 (9th Cir.1997) (citation omitted). See also MCI v. Bell-Atlantic, 36 F.Supp.2d 419, 422-23 (D.D.C.1999); AT & T Communications of South Central States, Inc. v. BellSouth Telecommunications, Inc., 20 F.Supp.2d 1097, 1100-01 (E.D.Ky.1998); US West Communications, Inc. v. MFS Intelenet, No. C97-222WD, 1998 WL 350588, at *2 (W.D.Wash. Jan.7, 1998). Thus, we review de novo the SCC’s

interpretations of the Telecommunications Act. See James Madison Ltd. v. Ludwig, 82 F.3d 1085, 1096 (D.C.Cir.1996). Turning to the standard for our review of SCC factfinding, we note first that the Act does not require us to sit as a super public utilities commission. “The appraisal of cost figures is itself a task for [agency expertise], since these costs involve many estimates and assumptions and, unlike a problem in calculus, cannot be proved right or wrong. They are, indeed, only guides to judgment. Their weight and significance require expert appraisal.” New York v.

United States, 331 U.S. 284, 328, 67 S.Ct. 1207, 91 L.Ed. 1492 (1947). Consequently, we review the. factfindings of the SCC under the substantial evidence standard. This standard, as we have noted in similar contexts, “is consistent with our precedent concerning federal judicial review of state-agency decisions.” AT & T Wireless PCS, Inc. v. Winston-Salem Zoning Bd. of Adjustment, 172 F.3d 307, 314 (4th Cir.1999). See also Clark v. Alexander, 85 F.3d 146, 151-152 (4th Cir.1996)(reviewing factfinding of Virginia local housing authority for substantial evidence).

The substantial evidence standard is, by the way, the same standard we would apply in cases where arbitration under the Act is conducted by the FCC, rather than a state utilities commission. See 47 U.S.C. § 252(e)(5)(requiring FCC to arbitrate if state commission fails to do so); 5 U.S.C. § 706(2)(E)(requiring that agency action be supported by substantial evidence when “reviewed on the record of an agency hearing provided by statute”).

In applying the substantial evidence standard, a “court is not free to substitute its judgment for the agency’s ...; it must uphold a decision that has ‘substantial support in the record as a whole’ even if it might have decided differently as an original matter.” AT & T Wireless PCS, Inc. v. City Council of City of Virginia Beach, 155 F.3d 423, 430 (4th Cir.1998), quoting NLRB v. Grand Canyon Mining Co., 116 F.3d 1039, 1044 (4th Cir.1997). With these standards in mind, we turn, at last, to the merits. B.

On the merits, GTE first argues that the SCO’s methodology for setting prices for unbundled network elements violates the FCC’s rules. The SCC invited GTE and the CLECs to submit models to generate prices for access to these elements, and both sides complied. The CLECs offered the Hatfield model, and GTE offered a model employing the M-ECPR methodology. The SCC staff proposed a third model based on Hatfield, with some modifications. The SCC ultimately adopted the staffs modified Hatfield model as “the only reasonable option” given the “limited record” in the arbitration proceeding.

GTE claims that this model does not comply with the methodology required by the FCC’s rules. The Act requires that state commissions determine “just and reasonable” prices for access to unbundled network elements. 47 U.S.C. § 252(d)(1). Those prices are to be “based on the cost (determined without reference to a rate-of-return or other rate-based proceeding) of providing the ... network element.” Id. § 252(d)(1)(A). The FCC’s First Report and Order provides the methodology that state commissions are to use under the Act in arbitrating disputes over prices for unbundled network elements.

See First Report and Order at ¶ 618 (“[T]he rules we adopt today set forth the methodological principles for states to use in setting prices.”). The Supreme Court has confirmed the FCC’s authority to issue rules “to design a pricing methodology” under the Act. Iowa Utils. Bd., 119 S.Ct. at 733. Under the FCC rules a state commission must set the price for an unbundled network element based on the element’s forward-looking Total Element Long Run Incremental Cost (TELRIC). See First Report and Order at ¶ 672.

A “forward-looking” approach looks to the costs a new entrant (CLEC) would incur in constructing a functionally comparable network today. “Long run” is a period “so long that all of the firm’s present contracts will have run out, its present plant and equipment will have been worn out or rendered obsolete and will therefore need replacement, etc.” Id. at ¶ 677 n. 1682. “Incremental costs” are “the additional costs ... that a firm will incur as a result of expanding the output of a good or service by producing an additional quantity of the good or service.” Id. at ¶ 675. Applying these

definitions, the FCC concluded that an unbundled network element’s TELRIC “should be based on costs that assume that wire centers will be placed at the incumbent LEC’s current wire center locations, but that the reconstructed local network will employ the most efficient technology for reasonably foreseeable capacity requirements.” Id. at ¶ 685.

See also id. at ¶ 690 (“Costs must be based on the incumbent LEC’s existing wire center locations and most efficient technology available.”); 47 C.F.R. § 51.505(b)(1) (1999) (requiring costs to be based on “the most efficient telecommunications technology currently available and the lowest cost network configuration, given the existing location of the incumbent LEC’s wire centers.”). GTE points to no specific aspect of the SCO’s modified Hatfield model that violates these rules.

That model estimates the cost of providing unbundled network elements based on the hypothetical construction and operation of the most efficient local network conceivable, with one limitation. Specifically, it takes only the location of an incumbent’s switches or wire centers as a given (the “scorched node” approach) and otherwise assumes a network comprised of the most efficient wiring configuration and most efficient technology available. This approach is expressly endorsed in the FCC’s pricing rules. See First Report and Order at ¶ 685.

In fact, the FCC in choosing a methodology that calculates costs on a forward-looking basis, cited the very version of Hatfield that the CLECs placed before the SCC in the arbitration. See id. at ¶ 834 (“We believe that the generic forward-looking costing models, in principle, appear best to comport with the preferred economic cost approach discussed previously. Several such models were placed in the [FCC rulemaking] record, including Hatfield 2 [and] Hatfield 2.2.”).

It is clear, then, that the modified Hatfield methodology used by the SCC complied with the FCC’s requirements that the costs of unbundled network elements be calculated on a forward-looking, long run basis, using the most efficient telecommunications technology currently available. According to GTE, it is not enough that Hatfield’s methodology complies with the FCC’s rules. GTE goes on to attack the accuracy of the Hatfield model. Complaints about accuracy are, of course, assertions of factual error, which we review under the substantial evidence standard.

GTE argues that the FCC, in an order issued after this appeal was filed, “rejected even a significantly updated and improved version of Hatfield for use in determining the costs of universal service.” Appellant’s Br. at 60. GTE refers to the FCC’s discussion in In the Matter of Federal-State Joint Board on Universal Service, Fifth Report and Order, 13 F.C.C.R. 21323 (1998) (Fifth Report and Order). Based on that order’s identification of flaws in the updated version of Hatfield, GTE contends that “the SCO’s use of an earlier and even more flawed version of Hatfield must be reversed.” Appellant’s Br. at 61.

The Fifth Report and Order, unlike the adversarial arbitration proceedings challenged here, was the result of a notiee-and-comment rulemaking in which the FCC considered the merits of several forward-looking cost models: an updated version of Hatfield (called HAI), the Benchmark Cost Proxy Model (BCPM), and the Hybrid Cost Proxy Model (HCPM). Recognizing that each had advantages and disadvantages, the FCC ultimately endorsed a synthesized model that incorporated elements of HAI, BCPM, and HCPM. See Fifth Report and Order at ¶ 92.

GTE is correct in asserting that the FCC has identified specific defects in the Hatfield model that cause it to under-estimate costs.

See, e.g., First Report and Order at ¶ 794 (“We do not believe, however, that these model outputs by themselves necessarily represent accurate estimates of the absolute magnitude of loop costs.... [F]urther analysis is necessary in order to evaluate fully the procedures and input assumptions that the models use in order -to derive cost estimates.”); Fifth Report and Order at ¶ 58 (“The HAI model also sacrifices accuracy by assuming that customers are dispersed uniformly within its distribution areas.”).

However, GTE failed to provide evidence to the SCC that would have permitted the commission to assess more accurately the cost of providing network elements. Rather, GTE presented a model that the SCC staff concluded was a “black box,” in that its operations and assumptions could not be tested. The SCC staff complained that GTE failed to suggest alternate input data or assumption changes for the Hatfield model, that its documentation was incomplete, that its model included data not specific to Virginia, and that it relied on assumptions and inputs that could not be verified or supported.

Both the SCC and its staff noted that the Hatfield model, unlike GTE’s model, had been subject to scrutiny by the FCC, as well as by parties to numerous arbitration proceedings around the country. Consequently, the generally understood Hatfield model was less likely than the untested GTE model to have fundamental problems that had not been discovered. The SCC concluded that the criticisms of the different cost models offered by GTE and the CLECs had merit. Consequently, the SCC refused to accept either model presented by the parties.

See Order Resolving Rates for Unbundled Network Elements and Interconnection, Wholesale Discount for Services Available for Resale, and Other Matters, PUC 960117, PUC 960118, PUC 960124, PUC 960131 (Dec. 11, 1996) (“We find that the evidence presented by the parties was not adequate to allow us to choose either the GTE or the AT & T/MCI Hatfield model results for determining costs to be used for setting permanent rates.”). Instead, the SCC adopted the Hatfield model, as modified by the staff to address GTE’s concerns about fill factors and the cost of capital.

In doing that, the SCC repeated its concerns about the models put forward by the parties, but observed that it was required by the Act to resolve the arbitrated issues within nine months of the CLECs’ request for interconnection. See 47 U.S.C. § 252(b)(4)(C).

Given the limited record, limited time, demonstrated flaws in each side’s model, and the absence of any opportunity to adequately test GTE’s model, the SCC set interim rates based on the modified Hatfield model, the “only reasonable option.” GTE concludes its briefing with the suggestion that any failure on its part to provide the SCC with a verifiable pricing proposal or data that supported it is “irrelevant” because the SCC “was still obligated to set rates that met the Act’s commands.” Appellant’s Reply Br. at 29.

The Act does not command the SCC to independently acquire evidence that GTE did not provide, and we will affirm the SCC determinations so long as they find support in the record as a whole. See 47 U.S.C. § 252(b)(4)(B) (“If any party refuses or fails unreasonably to respond on a timely basis to any reasonable request from the State commission, then the State commission may proceed on the basis of the best information available to it from whatever source derived.”).

In the face of conflicting evidence and a statutory deadline for decision, the SCC relied on the best information available to set prices for network elements: the Hatfield model as adjusted to incorporate specific objections based on verifiable evidence. We conclude that the SCC’s decision to set interim prices based on that information and the commission’s resulting decisions on prices all have substantial support in the record as a whole. We therefore affirm the district court’s judgment upholding the SCC’s network element pricing determinations. C.

GTE also challenges the SCC’s methodology for determining the wholesale prices for GTE’s retail products. The SCC assumed, for purposes of determining wholesale prices, that GTE would exit the retail market altogether and thus incur no costs for marketing, billing, collection, and other such items.

Under the Act wholesale prices are determined on the basis of the incumbent LEC’s “retail rates charged to subscribers for the telecommunications service requested, excluding the portion thereof attributable to any marketing, billing, collection, and other costs that will be avoided by the local exchange carrier.” 47 U.S.C. § 252(d)(3). According to GTE, the SCC’s assumption that GTE will exit the retail market entirely is contrary to fact, resulting in a discount that is not limited to the costs that “will be avoided” by GTE when it begins wholesaling.

In other words, GTE says that when it wholesales only some of its retail services, the costs it will avoid will be smaller on a per unit basis than the costs it would avoid if it abandoned retailing entirely. In its First Report and Order the FCC issued the following guide to state commissions for setting wholesale prices: “the portion [of the retail rate] ... attributable to costs that will be avoided” includes all of the costs that the LEC incurs in maintaining a retail, as opposed to wholesale, business.

In other words, the avoided costs are those that an incumbent LEC would no longer incur if it were to cease retail operations and instead provide all of its services through resellers. Thus, we reject the arguments of incumbent LECs and others who maintain that the LEC must actually experience a reduction in its operating expenses for a cost to be considered “avoided” for purposes of

section 252(d)(3). First Report and Order at ¶ 911 (emphasis added). The SCC’s assumption in setting wholesale prices — that GTE would exit the retail market altogether' — is completely consistent with the FCC’s rule. We therefore affirm the district court judgment upholding the wholesale prices set by the SCC. Of course, if the Eighth Circuit strikes down the FCC’s wholesale pricing rule, GTE will have the opportunity to modify its interconnection agreements and seek recovery of any underpayment in the process of settling up. See

part IV, supra. VI. The district court’s judgment upholding the SCC’s pricing methodology and determinations is affirmed. AFFIRMED . If the state commission elects not to assume its role as arbitrator, the Act shifts this responsibility to the FCC. See °Tl U.S.C. § 252(e)(5). . In light of the history and substance of the orders and opinions of the Eighth Circuit and the Supreme Court, we reject GTE's alterna-live argument that the pricing rules are not yet in effect. .

GTE also relies on cases creating a limited exception to the Hobbs Act's filing deadline: when a party did not seek direct review to challenge an agency’s rule within sixty days of its issuance, and the agency later applies the rule in an enforcement action against the party, the court of appeals reviewing the enforcement order also has jurisdiction to hear the party’s substantive challenge to the rule. See, e.g., Commonwealth Edison Co. v. NRC, 830 F.2d 610, 614 (7th Cir.1987). This judi-daily created exception simply waives the Hobbs Act’s sixty-day filing deadline in a narrow circumstance.

The exception is of no help to GTE because it does not alter the exclusive jurisdiction of the courts of appeals, which may be invoked only by the filing of a petition for review. See, e.g., JEM Broad. Co. v. FCC, 22 F.3d 320, 324-25 (D.C.Cir.1994); Texas v. United States, 749 F.2d 1144, 1146 (5th Cir.1985). . For example,

section AJ of the GTE-Cox agreement reads as follows: The terms and conditions of this Agreement shall be subject to any and all applicable laws, rules, regulations or guidelines that subsequently may be prescribed by any federal, state or local governmental authority. To the extent required by any such subsequently prescribed law, rule, regulation or guideline, the parties agree to modify, in writing, the affected term(

s) and condition(

s) of this Agreement to bring them into compliance with such law, rule, regulation or guideline. . We are aware that in reviewing state commission arbitration decisions under the Act, some other courts, including the district court in this case, have used the "arbitrary and capricious” standard of review. See, e.g., U.S. West Communications v. MFS Intelenet, Inc., 193 F.3d 1112, 1116 (9th Cir.1999); GTE South, Inc. v. Morrison, 6 F.Supp.2d 517, 523 (E.D.Va.1998). With respect to review of factfindings, there is no meaningful difference between this standard and the substantial evidence standard we apply.

See James City County v. EPA, 12 F.3d 1330, 1338 n. 4 (4th Cir.1993); Association of Data Processing Serv. Orgs. v. Board of Governors, 745 F.2d 677, 683-84 (D.C.Cir.1984) (Scalia, J.).