Metropolitan Edison Co. v. Pennsylvania Public Utility Commission

OPINION OF THE COURT

This case requires us to decide the pre-clusive effect of a state utility agency’s ruling, which has been affirmed by Pennsylvania’s Commonwealth Court and denied review by the Pennsylvania Supreme Court and the United States Supreme Court. Although the Appellants, electric utility companies Metropolitan Edison Co. (“Mei>-Ed”) and Pennsylvania Electric Co. (“Penelec”) (collectively, the “Companies”), also, in effect, invite us to review the agency’s ruling on the merits, we need not and do not take that step.

The Companies’ end-game appears to be to recoup from their customers more than $250 million in costs associated with “line losses”—ie., energy that is lost when electricity travels over power lines—and interest related to those costs. For reasons we will explain, the Companies’ line loss costs had increased pursuant to a mandate by the Federal Energy Regulatory Commission (“FERC”), and the Companies’ ability to recover those costs depended on whether line-loss costs were classified as a cost of electricity generation or as a cost of electricity transmission on their customers’ utility bills. In a prior proceeding, the Pennsylvania Public Utility Commission (“PUC”) rejected the Companies’ proposal to classify line-loss costs as a cost of transmission, thereby preventing the Companies from passing those costs through to their customers. The Companies then pressed their arguments and lost in the Pennsylvania state courts and were denied review by the United States Supreme Court.

The Companies now seek declaratory judgment and injunctive relief in federal court against the PUC and its Commis*341sioners in their official capacities, which would effectively set aside the result of the earlier state proceeding. The United States District Court for the Eastern District of Pennsylvania held that the Companies’ unsuccessful pursuit of relief in the state proceeding precluded their effort to claim relief in federal court. In short, none of the Companies’ claims survived application of the doctrine of issue preclusion. We agree and will affirm the District Court’s order of dismissal.

I. BACKGROUND1

To understand the issues raised in this appeal, it is helpful to first look at the legislative and administrative framework of electricity regulation and how that framework affects the parties before us.

A. The Federal Power Act and the Filed Rate Doctrine

In 1935, Congress enacted the Federal Power Act (“FPA”), 16 U.S.C. § 791a et seq., which authorized “federal regulation of the expanding business of transmitting and selling electric power in interstate commerce.” New York v. FERC, 535 U.S. 1, 6, 122 S.Ct. 1012, 152 L.Ed.2d 47 (2002) (internal quotation marks and citation omitted). As it stands today, the FPA grants FERC jurisdiction over “the transmission of electric energy in interstate commerce and the sale of such energy at wholesale in interstate commerce,” 16 U.S.C. § 824(a), and requires “[a]ll rates and charges ... subject to the jurisdiction of the Commission” to be “just and reasonable,” id. § 824d(a).2 The scope of that authority, broad though it is, is meant “to extend only to those matters which are not subject to regulation by the States.” Id. § 824(a).

The so-called “filed rate doctrine” is an application of the FPA’s statutory grant of authority to FERC. See Borough of Ellwood City v. FERC, 583 F.2d 642, 648 (3d Cir.1978) (calling the filed rate doctrine “not so much a judicially created ‘doctrine’ as an application of explicit statutory language”). It may be understood for our purposes as the rule that “interstate power rates filed with FERC or fixed by FERC must be given binding effect by state utility commissions determining intrastate rates.” Nantahala Power & Light Co. v. Thornburg, 476 U.S. 953, 962, 106 S.Ct. 2349, 90 L.Ed.2d 943 (1986). The filed rate doctrine thus “concern[s] the pre-emptive impact of federal jurisdiction ... on state regulation.” Miss. Power & Light Co. v. Mississippi, 487 U.S. 354, 371, 108 S.Ct. 2428, 101 L.Ed.2d 322 (1988). The doctrine of federal pre-emption, in turn, is rooted in the Supremacy Clause of the Constitution, which provides that federal law “shall be the supreme Law of the Land[,] ... any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.” U.S. Const, art. VI, cl. 2; see also Nantahala, 476 U.S. at 963, 106 S.Ct. 2349 (stating that the application of the filed rate doctrine to state tribunals is “a matter of enforcing the Supremacy Clause”).

*342B. The Market for Electricity

Before the passage of the FPA, electricity was usually sold by vertically integrated electric utilities that controlled their own generators, transmission lines, and local distribution networks.3 New York, 535 U.S. at 5, 122 S.Ct. 1012; see also ARIPPA v. Pa. Pub. Util. Comm’n, 792 A.2d 636, 642 (Pa.Commw.Ct.2002) (noting that, historically, electric utilities in Pennsylvania were vertically integrated). Services were typically “bundled” together, “meaning consumers paid a single price for generation, transmission, and distribution.” Midwest ISO Transmission Owners v. FERC, 373 F.3d 1361, 1363 (D.C.Cir.2004); see also 66 Pa. Cons.Stat. Ann. § 2802(13) (stating that the same was the case in Pennsylvania). “Although there were some interconnections among utilities, most operated as separate, local monopolies subject to state or local regulation.” New York, 535 U.S. at 5, 122 S.Ct. 1012.

Advances in technology since the enactment of the FPA have resulted in “[transmission grids [that] are now largely interconnected, which means that ‘any electricity that enters the grid immediately becomes a part of a vast pool of energy that is constantly moving in interstate commerce.’ ” N.J. Bd. of Pub. Utils. v. FERC, 744 F.3d 74, 81 (3d Cir.2014) (quoting New York, 535 U.S. at 7, 122 S.Ct. 1012). “[T]he development of a national, interconnected grid has made it possible for a generator in one state to serve customers in another, thus opening the door to potential competition that did not previously exist.” Id. Nevertheless, electric utilities maintained ownership of transmission lines, and, thus, “the ability to stifle competition from new generators by ‘refusing] to deliver energy produced by competitors or [by] deliver[ing] competitors’ power on terms and conditions less favorable than those they applied] to their own transmissions.’ ” Id. (alterations in original) (quoting New York, 535 U.S. at 8-9, 122 S.Ct. 1012). As a result, for many years, monopolistic tendencies still restrained competition in the market for electricity.

In 1996, FERC issued Order No. 888, a landmark ruling aimed at encouraging competition and lowering electricity rates. See Promoting Wholesale Competition Through Open Access Non-Diseriminatory Transmission Services by Public Utilities, 61 Fed.Reg. 21,540, 21,541 (May 10, 1996) [hereinafter Order No. 888], ajfd in relevant part, Transmission Access Policy Study Grp. v. FERC, 225 F.3d 667 (D.C.Cir.2000), aff'd sub nom. New York v. FERC, 535 U.S. 1, 122 S.Ct. 1012, 152 L.Ed.2d 47 (2002). Significantly for this case, that Order requires the “unbundling” of wholesale generation and wholesale transmission services. Id. at 21,558, 21,-571, 21,577-78. Each electric utility must apply the same rate for wholesale transmission services to itself and others so as to provide open access to transmission services. Id. at 21,541. Although FERC noted that unbundling retail services would also be helpful to encouraging competition, Order No. 888 only required the unbun-dling of wholesale transmission from wholesale generation. Id. at 21,577.

That same year, Pennsylvania enacted the Electricity Generation Customer Choice and Competition Act (the “Electric *343Competition Act”), 66 Pa. Cons.Stat. Ann. § 2801 et seq., which deregulated the business of electricity generation within the Commonwealth. The Electric Competition Act was designed to promote competition in the electricity market and lower retail rates for electric energy. See 66 Pa. Cons. Stat. Ann. § 2802(4), (7) (noting the relatively high rates for electricity in Pennsylvania and the importance of transitioning to “greater competition in the electricity generation market”); see also ARIPPA, 792 A.2d at 642 (stating the rationale behind the Electric Competition Act). The Act “requires electric utilities to unbundle their rates and services and to provide open access over their transmission and distribution systems to allow competitive suppliers to generate and sell electricity directly to consumers in this Commonwealth.” 66 Pa. Cons.Stat. Ann. § 2802(14). Under the law, customers in Pennsylvania can purchase generation services directly from licensed “electric generation suppliers” rather than just from electric utilities. Id. Electric utilities, however, continue to provide the transmission and distribution of electricity, and “[i]f consumers d[o] not choose to or [a] re unable to purchase power from another supplier, the local utility [i]s still required to provide electricity to them as the Provider of Last Resort.”4 ARIPPA, 792 A.2d at 642 (citing 66 Pa. Cons.Stat. Ann. § 2802(16)).

As a result of introducing competition into the market for electricity generation services, the Electric Competition Act left electric utilities with “transition,” or “stranded,” costs, which are defined as “known and measurable” generation-related costs that “traditionally would be recoverable under a regulated environment but which may not be recoverable in a competitive electric generation market and which the [PUC] determines will remain following mitigation by the electric utility.”5 66 Pa. Cons.Stat. Ann. § 2803 (defining “[tjransition or stranded costs”). In other words, stranded costs are costs that were incurred while an electric utility developed as a generator and supplier of power within a regulated market but that will no longer be recoverable in a more competitive market. Indianapolis Power & Light Co. v. Pa. Pub. Util. Comm’n, 711 A.2d 1071, 1074 (Pa.Commw.Ct.1998); see also Roger A. Greenbaum, Annotation, Recovery of “Stranded Costs” by Utilities, 80 A.L.R. 6th 1 (2012) (“ ‘Stranded costs’ represent that portion of ... a utility’s generation assets not yet recovered through [regulated rates] that has become unrecoverable in a deregulated environment.”). For example, stranded costs may include a long-term investment in a generation facility that is no longer used due to deregulation of the market or other transition costs like the cost of retraining employees. 66 Pa. Cons.Stat. Ann. § 2803; see also PECO Energy Co. v. Commonwealth, 591 Pa. 405, 919 A.2d 188, 189 n. 2 (2007) (“Stranded costs ... often [involve] assets with high construction costs which were due to be recuperated through the rate *344guaranteed under the previous monopoly-system and which now will operate at a loss.”); Indianapolis Power & Light, 711 A.2d at 1074 n. 4 (explaining the main categories of stranded costs). The Electric Competition Act allows electric utilities to recover certain stranded costs through a “charge applied to the bill of every customer accessing the transmission or distribution network,” separate from the charge for the actual amount of electricity consumed. ARIPPA, 792 A.2d at 643 (internal quotation marks omitted) (citing 66 Pa. Cons.Stat. Ann. §§ 2803, 2806(c), 2808).

To ease transition to a competitive market, the Electric Competition Act required electric utilities in the Commonwealth to submit “restructuring plans,” including proposed rate schedules and plans for the recovery of stranded costs, for approval by the PUC. 66 Pa. Cons.Stat. § 2806(d)-®. The Act outlined some restructuring standards, such as “caps” on service rates for certain periods of time in exchange for electric utilities being able to recover their stranded costs. Id. § 2804(4). The rate caps allowed customers to obtain electricity at the capped rates, which put downward pressure on any market rate above that level. Cf. ARIPPA, 792 A.2d at 643 (noting that customers would buy from an electric utility as the provider of last resort if market rates rose above the capped rates). Electric utilities could seek approval from the PUC for exceptions to the rate-cap standards. 66 Pa. Cons.Stat. Ann. § 2804(4)(iii).

C. The Companies’ Settlement Agreement

The Companies provide electricity and associated services to customers in their prescribed territories within Pennsylvania. Pursuant to passage of the Electric Competition Act, they filed restructuring plans with the PUC in 1997. In 1998, they jointly and voluntarily entered into an omnibus settlement agreement (the “Settlement Agreement”) that resolved disputes related to their restructuring plans and to pending litigation in the United States District Court for the Eastern District of Pennsylvania. Of importance in the present matter, the Companies agreed to caps on “Transmission and Distribution (T & D) Charges” through December 31, 2004, as well as caps on “Generation rates” through December 31, 2010. (J.A. at 115.) Compared to the standard time-frames for rate caps under the Electric Competition Act, the periods for those agreed-upon rate caps represented extensions of three-and-a-half years on the transmission rate cap and five years on the generation rate cap. 66 Pa. Cons.Stat. Ann. § 2804(4)(i), (ii). In exchange for accepting those extensions, the Companies were given additional time to recover certain stranded costs from their customers. The PUC entered a final order approving the Settlement Agreement in October 1998.6

D. The Companies’ Line-Loss Costs

The Companies’ distribution facilities are connected to an interstate transmission grid that is overseen by PJM Interconnection, LLC (“PJM”). PJM is a regional transmission organization, a voluntary association “to which transmission providers ... transfer operational control of their facilities for the purpose of efficient coordination” of the wholesale electricity market. Morgan Stanley Capital Grp. Inc. v. Pub. Util. Dist. No. 1., 554 U.S. 527, 536, 128 S.Ct. 2733, 171 L.Ed.2d 607 (2008). *345Among other things, PJM ensures that there is a sufficient amount of electricity in its regional transmission system, which reaches the District of Columbia and thirteen Mid-Atlantic and Midwest states, including Pennsylvania. N.J. Bd. of Pub. Utils., 744 F.3d at 79, 82. FERC regulates the wholesale rates that PJM charges the Companies, and those rates are set forth in PJM’s Open Access Transmission Tariff (“PJM’s Tariff’), which is on file with FERC. Among the costs that the Companies are billed by PJM are the costs for line losses.7 As noted earlier, line losses represent the energy that is lost when electricity travels over power lines. PJM bills the Companies for line losses as a discrete line item within the charge for “transmission” service. (J.A. at 41-42 (Amended Complaint); id. at 481, 483, 486, 488,191-92 (PJM’s Tariff).)

Until June 30, 2007, PJM calculated and billed for line losses using what is called the “average loss” methodology. See Atl. City Elec. Co. v. PJM Interconnection, LLC (Atlantic City I), 115 FERC ¶ 61,-132, p. 61,473 (2006), reh’g denied, 117 FERC ¶ 61,169. As the name suggests, PJM charged its customers for line losses “equal to the average loss cost”—PJM recovered line-loss costs by allocating the cost to all of its customers equally. Id. at 61,473. As a result, line-loss costs did not depend on the distance between the point of electricity generation and the point of electricity delivery. Id. at 61,473-74.

On March 2, 2006, several electric utilities (but not the Companies) filed a complaint with FERC alleging that, under an agreement appended to PJM’s Tariff, PJM was required to switch from the average loss methodology to a “marginal loss” methodology to calculate the cost of line losses. Id. at 61,473. “Under the marginal loss method, the effect of losses on the marginal cost of delivering energy is factored into the energy price ... at each location.” Id. at 61,474. Thus, “[o]ther things being equal, customers near generation centers pay prices that reflect smaller marginal loss costs while customers far from generation centers pay prices that reflect higher marginal loss costs.” Id.; see also Sacramento Mun. Util. Dist. v. FERC, 616 F.3d 520, 524 (D.C.Cir.2010) (describing the marginal loss methodology as a rate structure in which “prices are designed to reflect the least-cost of meeting an incremental [energy] demand at each location on the grid, and thus prices vary based on location and time”). After issuing notice of the complaint, FERC solicited comments from numerous electric utilities and customer coalitions. Atlantic City I, 115 FERC at 61,474-77.

In an order issued in 2006, FERC held that the agreement appended to PJM’s Tariff required PJM to use the marginal loss methodology once it was technologically feasible to do so and that PJM had conceded that it possessed the necessary technology. Id. at 61,477. FERC also noted that using marginal loss pricing would result in cost savings to PJM and efficiencies in resource allocation. Id. at 61,474, 61,477-78. Accordingly, FERC required PJM to switch from using the average loss methodology to the marginal loss methodology of calculating line losses. Id. at 61,478. The Companies did not participate in the comments process before *346FERC or challenge the resulting order. Id. at 61,474-77.

A few months later, FERC denied rehearing requests but granted a request to delay implementation of the marginal loss methodology to June 2007. Atl. City Elec. Co. v. PJM Interconnection, LLC (Atlantic City II), 117 FERC ¶ 61,169, pp. 61,-858, 61,861 (2006). The Companies did not directly challenge that order either; in fact, they assert that “no one did.” (Appellants’ Opening Br. at 36.) PJM’s implementation of FERC’s orders to change the calculation of line-loss costs, which orders we will refer to collectively as the Atlantic City decision, decreased the line-loss costs for some electric utilities. However, it increased the line-loss costs that PJM billed to the Companies.

II. PROCEDURAL HISTORY

Not surprisingly, the Companies eventually sought to recover their increased line-loss costs by asking the PUC to allow them to pass the expense through to their customers. A “transmission rider,” which was approved by the PUC in 2006 after the Companies’ transmission rate cap had lapsed, allowed the Companies to pass through various proposed transmission costs to their customers and to engage in an annual updating and reconciliation process in order to recover projected transmission costs and adjust for the over-or under-collection of past transmission costs. Pa. Pub. Util. Comm’n v. Metro. Edison Co., Nos. R-00061366C0001 et al., 2007 WL 496359 (Pa.P.U.C. Jan. 11, 2007), aff'd sub nom. Met-Ed Indus. Users Grp. v. Pa. Pub. Util Comm’n, 960 A.2d 189 (Pa. Commw.Ct.2008). Under that annual process, the Companies proposed for the first time in April 2008 to charge their customers for the higher line-loss costs that the Companies incurred after PJM’s implementation of the Atlantic City decision. Because the generation rate cap under the Settlement Agreement was still in effect at that time, the Companies could only recover the line-loss costs if granted approval to bill them to customers as a cost of transmission.

A. The PUC Order

Pennsylvania’s Office of Consumer Advocate and Office of Small Business Advocate 8 and two groups known as the MetEd Industrial Users Group and the Pene-lec Industrial Users Alliance (collectively, the “Customer Groups”)—all representing the interests of various customers—filed complaints before the PUC to contest the Companies’ proposed rate increase. They argued that line-loss costs should properly be viewed as a generation cost, not a transmission cost, and, thus, could not be increased due to the Settlement Agreement’s generation rate cap in effect through the end of 2010. The Customer Groups’ complaints were consolidated for a hearing before a PUC administrative law judge (“ALJ”).9 An evidentiary hearing was held after the Companies and Customer Groups completed briefing.

The ALJ recommended dismissing the Customer Groups’ complaints and approving the Companies’ requests to recover line-loss costs as a transmission cost. In re Pa. Elec. Co. Transmission Serv. Charge, Nos. M-2008-2036188 et al., 2009 Pa. PUC LEXIS 2328 (July 24, 2009). *347The Customer Groups filed exceptions to the ALJ’s recommendation, triggering review by the Commissioners of the PUC. See 66 Pa. Cons.Stat. Ann. § 332(h) (providing procedure for excepting to an ALJ’s recommendation).

The Customer Groups argued to the PUC that line-loss costs should not be billed to them as transmission costs because (1) line losses have historically been recognized as part of the cost of electricity generation; (2) how PJM bills the Companies for line losses is irrelevant to whether those losses should be billed to the Companies’ customers as a generation or transmission cost; and (3) the Companies themselves have historically treated line-loss costs as generation costs. The Companies responded by (1) emphasizing how line losses are related to transmission, ie., as electricity is transmitted over longer distances, line losses increase; (2) pointing to the FERC-approved definition of “transmission losses” in PJM’s Tariff;10 and (3) arguing that PJM bills the Companies for line losses as a cost of transmission service. The Companies also claimed that they did not initially seek to recover line-loss costs as a transmission cost because, at the time, FERC had not yet mandated the use of marginal loss pricing.

The PUC in a split decision entered March 3, 2010 (the “PUC Order”) ultimately rejected all of the Companies’ arguments and agreed with the Customer Groups. The PUC did not adopt the ALJ’s recommendation that line losses be considered a transmission cost, concluding instead that the Companies’ line losses were generation costs subject to the Settlement Agreement’s generation rate cap that was in effect through 2010. As the merits of the PUC Order are not before us, suffice it to say that the PUC thoroughly reviewed all of the Companies and the Customer Groups’ arguments. By a three-to-two vote of the Commissioners, the agency required the Companies to file tariff supplements consistent with the majority’s decision.11

B. Review of the PUC Order

The Companies petitioned the Pennsylvania Commonwealth Court for review of the PUC Order to the extent it denied their request to classify line-loss costs as a transmission cost.12 In June 2011, the Commonwealth Court, sitting en banc, affirmed that aspect of the PUC Order in a unanimous opinion and order. The Commonwealth Court reviewed whether the PUC’s findings of fact—“including the [PUC’s] finding that line loss costs were and are being recovered in [the] Compa*348nies’ generation rates”—were supported by substantial evidence. (J.A. at 176.) The court also reviewed whether the PUC erred as a matter of law in concluding that line-loss costs are a generation cost. It found no reversible error in either regard.

Important for purposes of this appeal, the Commonwealth Court addressed the Companies’ argument that classifying line-loss costs as a generation cost for purposes of retail billing “violates the Filed Rate Doctrine and is inconsistent with ... FERC’s characterization of line losses.” (J.A. at 183.) The Companies had cited FERC decisions that allegedly treated line losses as a cost of transmission, but the Commonwealth Court held that those decisions “do not unambiguously state that such costs are transmission-related.” (J.A. at 188.) As the court saw it, several of those FERC decisions included language tying line losses to the costs of generating electricity. The court thus concluded that FERC’s decisions did not create any “direct conflict” with the classification of the Companies’ line-loss costs as generation costs. (J.A. at 189.)

Furthermore, the Commonwealth Court held that, for two reasons, there was no impermissible “trapping” of the Companies’ costs. Cost trapping, in this context, refers to a state “bar[ring] regulated utilities from passing through to retail consumers FERC-mandated wholesale rates.” Miss. Power & Light, 487 U.S. at 372, 108 S.Ct. 2428. First, the court stated that the Companies’ trapping argument was “premised on the [rejected] assumption that line loss costs are transmission-related.” (J.A. at 191.) Second, it determined that any alleged trapping was resolved by the Settlement Agreement “because [the] Companies voluntarily extended th[e] [generation] rate cap through December 31, 2010 ... in exchange for recovering stranded costs,” thus assuming the risk that any increased costs would not be recoverable. (Id.) The Commonwealth Court therefore affirmed the PUC Order in relevant part, holding that the Order was not inconsistent with FERC precedent, did not run afoul of the filed rate doctrine, and did not improperly trap the Companies’ costs.

The Pennsylvania Supreme Court subsequently denied the Companies’ petition for allowance of appeal, and the United States Supreme Court denied the Companies’ petition for a writ of certiorari. The Commonwealth Court’s decision (the “State Decision”) affirming the classification of line-loss costs for retail billing purposes thus became final.

C. The Federal Action

On July 13, 2011, while their petition for review before the Pennsylvania Supreme Court was pending, the Companies filed the present action in the District Court, naming as defendants the PUC and PUC Commissioners Robert F. Powelson, John F. Coleman, Jr., Pamela A. Witmer, Wayne E. Gardner,13 and James H. Caw-ley in their official capacities (collectively, the “PUC Defendants”). As originally filed, the suit claimed that the PUC Defendants had violated the FPA and the filed rate doctrine, as well as the Companies’ property interests under the Fourteenth Amendment. The Companies later filed an amended complaint to add a claim that the Electric Competition Act is unconstitutional as applied. Pennsylvania’s Office of *349Small Business Advocate, the Met-Ed Industrial Users Group, and the Penelec Industrial Users Alliance filed motions to intervene, which the District Court granted, permitting them “to intervene as defendants.” (J.A. at 5 (Dkt.41).)

The gravamen of the Companies’ amended complaint is that the outcome of the state proceeding resulted in unlawful trapping of the line-loss costs that PJM charged them pursuant to FERC-ap-proved tariffs. The Companies ultimately seek to recover the line-loss costs they incurred between 2007 and 2010.14 Those disputed costs, including interest, allegedly total more than $250 million.15

Count I of the Companies’ amended complaint asserts that the alleged cost-trapping violates the FPA and the filed rate doctrine. Count II alleges that the PUC Order “imposes a confiscatory rate on the Companies” by depriving them of a property interest in recovering line-loss costs and, thus, violates the Due Process Clause of the Fourteenth Amendment and, by extension, the FPA’s requirement that rates be just and reasonable. (J.A. at 50.) Count III claims that the Electric Competition Act is unconstitutional as applied because it is pre-empted by federal law. In sum, the Companies allege that, by barring them from recovering the line-loss costs that PJM charged them under a FERC-mandated methodology, the PUC Order violates the filed rate doctrine, the Supremacy Clause of the Constitution, the Fourteenth Amendment, and the FPA, and, to the extent the PUC and the Commonwealth Court relied on the Electric Competition Act, that statute, as applied, is pre-empted by federal law.

The PUC Defendants moved to dismiss the amended complaint,16 arguing that the Companies’ claims are barred by issue preclusion, claim preclusion, abstention principles, and judicial estoppel.17 With respect to preclusion, the Companies responded with three arguments for why their claims are not barred by preclusion principles: the state proceeding was legislative, rather than judicial, in nature; the Commonwealth Court reviewed the PUC’s ruling under the wrong standard; and the PUC Order was facially pre-empted by FERC’s exclusive jurisdiction.

After hearing oral argument on the renewed motion to dismiss, the District Court dismissed all of the Companies’ claims on the basis of issue preclusion. The Companies then timely filed this appeal.18

*350III. DISCUSSION19

At the outset, it is worth emphasizing what is and is not at issue here. The question before us is whether the State Decision—ie., the Commonwealth Court’s decision that the PUC’s classification of line-loss costs did not violate the filed rate doctrine or impermissibly trap costs—bars litigation of the claims in this federal action. It is not whether the PUC correctly classified the Companies’ line-loss costs as generation costs in the first instance.

The Companies offer several arguments for denying the State Decision any preclu-sive effect, based on what they call exceptions to the application of the Full Faith and Credit Statute, 28 U.S.C. § 1738. (Appellants’ Opening Br. at 29-44.) They also argue that the District Court misinterpreted the reach of the State Decision to preclude all of their claims. The PUC Defendants respond that the principles of issue preclusion properly bar the present case and, in the alternative, that dismissal would be proper under claim preclusion, abstention principles, and judicial estoppel.

A. Issue Preclusion

The District Court viewed the State Decision as having preclusive effect because the Commonwealth Court addressed the Companies’ arguments that the PUC Order violated the filed rate doctrine and impermissibly trapped costs. Under the doctrine of issue preclusion, also referred to as collateral estoppel, “once a court has decided an issue of fact or law necessary to its judgment, that decision may preclude relitigation of the issue in a suit on a different cause of action involving a party to the first case.” Allen v. McCurry, 449 U.S. 90, 94, 101 S.Ct. 411, 66 L.Ed.2d 308 (1980). Federal courts give preclusive effect to issues decided by state courts, to “not only reduce unnecessary litigation and foster reliance on adjudication, but also promote the comity between state and federal courts that has been recognized as a bulwark of the federal system.” Id. at 95-96, 101 S.Ct. 411. The preclusive effect of a state court judgment in a subsequent federal lawsuit is determined by the Full Faith and Credit Statute, which provides, in relevant part, that state judicial proceedings “shall have the same full faith and credit in every court within the United States ... as they have by law or usage in the courts of such State ... from which they are taken.” 28 U.S.C. § 1738. That statute has been interpreted by the Supreme Court to require a federal court to look to state law to determine the preclusive effect of a prior *351state judgment. Marrese v. Am. Acad. of Orthopaedic Surgeons, 470 U.S. 873, 380, 105 S.Ct. 1327, 84 L.Ed.2d 274 (1985).

Here, there is no dispute that Pennsylvania’s preclusion law applies. The Pennsylvania Supreme Court has established a five-prong test providing that issue preclusion will apply when:

(1) the issue decided in the prior case is identical to the one presented in the later action;

(2) there was a final adjudication on the merits;

(3) the party against whom the plea is asserted was a party or in privity with a party in the prior case; (4) the party ... against whom the doctrine is asserted had a full and fair opportunity to litigate the issue in the prior proceeding; and (5) the determination in the prior proceeding was essential to the judgment! [20]

Office of Disciplinary Counsel v. Kiesewetter, 585 Pa. 477, 889 A.2d 47, 50-51 (2005).

As noted before, Count I of the amended complaint alleges that the PUC Order trapped the Companies’ line losses in violation of the filed rate doctrine and, by extension, in violation of the FPA and the Supremacy Clause of the Constitution. The Companies do not appear to dispute that Count I meets all five of the requirements for issue preclusion under Pennsylvania law. That is wise, since (1) the Commonwealth Court squarely decided that the PUC Order did not violate the filed rate doctrine or impermissibly trap costs; (2) the court’s decision was on the merits and final after both the Pennsylvania Supreme Court and the United States Supreme Court denied petitions to review the State Decision;21 (3) the Companies were parties to the underlying state proceeding; (4) the Companies were given the opportunity to fully and fairly litigate the issue, as they were represented by counsel, filed multiple briefs, pointed to evidence from the PUC proceeding, and presented oral argument to the en banc Commonwealth Court;22 and (5) the determination was essential to the judgment because, had the Commonwealth Court decided that there was a violation of the filed rate doctrine, it could not have affirmed the PUC Order as it did. Absent some exception, Count I is therefore barred by issue preclusion.

According to the Companies, however, their claims in Counts II and III—which allege a confiscatory taking and federal pre-emption of the Electric Competition Act, respectively—do not meet the five-prong issue preclusion test under Pennsyl*352vania law. They argue that those claims raise new issues that were not decided in the state proceedings and that the Companies were not given a full and fair opportunity to litigate them. The PUC Defendants argue that the Companies failed to object to the application of issue preclusion to Counts II and III before the District Court, thereby waiving their arguments against preclusion of those two counts. The PUC Defendants further submit that Counts II and III, like Count I, require adjudication of the very issues that were fully litigated and decided in the state proceedings. We consider the waiver argument first.

1. Waiver

“[Fjailure to raise an issue in the district court constitutes a waiver of the argument.” Gass v. V.I. Tel. Gorp., 311 F.3d 237, 246 (3d Cir.2002) (internal quotation marks and citation omitted). “We only depart from this rule when manifest injustice would result from a failure to consider a novel issue.” Id. (internal quotation marks and citation omitted). The Companies do not attempt to show that manifest injustice would result from a failure to consider their arguments regarding Counts II and III. Rather, they claim that there is no waiver because they “provided [the PUC Defendants) with fair notice and the grounds on which Counts II and III separately rested.” (Appellants’ Reply Br. at 24 (citing Phillips v. Cnty. of Allegheny, 515 F.3d 224, 233-35 (3d Cir.2008)).) That argument misses the mark because, even if it were factually accurate, it relates to pleading requirements. It does not show that the Companies preserved their arguments for appeal by raising them in the District Court, and indeed they did not.

The Companies claim that they did not waive their arguments because their “[b]rief ... explained] why Counts II and III were not commingled with Count I.” (Appellants’ Reply Br. at 25.) But that argument is unavailing because the brief that they cite to is the opening brief before us, not anything that they filed in the District Court.23 The Companies also argue that they did not litigate the merits of Count III in the state proceeding, but that is an argument that Count III is not precluded; it is not a justification for failing to raise arguments specific to Count III in response to the motion to dismiss in the District Court.

The only colorable argument that the Companies make to rebut waiver is that the PUC and its Commissioners, in their motion to dismiss, “did not argue [in the first place] that Count II was barred by issue preclusion.” (Appellants’ Reply Br. at 25.) In that regard, the Companies are correct. As a consequence, we are not prepared to say that they were required, at the risk of waiver, to argue that issue preclusion does not apply to Count II. We will not consider the Companies’ issue preclusion arguments with respect to Count II waived. The PUC and its Commissioners did, however, argue in the District Court that issue preclusion bars Counts I and III. As the Companies did not attempt to distinguish Count III in the District Court in response to the issue preclusion arguments, they waived at least their arguments as to that count.24

*3532. Issue preclusion analysis

In any event, as the PUC Defendants argue, Counts II and III of the Companies’ amended complaint are both barred by issue preclusion, absent any exceptions that would preserve them. Count II alleges that the PUC Order “imposes a confiscatory rate on the Companies in violation of the Constitution because it deprives the Companies of their property right to recover their federally-approved costs of providing electric service, which includes marginal transmission line loss charges, to their Pennsylvania customers.” (J.A. at 50.) Count II further alleges that the PUC Order is confiscatory because it violates the FPA’s requirement for rates to be just and reasonable. In other words, Count II is premised on the success of the argument that the PUC Order violated the filed rate doctrine and, thus, impermissibly “trapped” the Companies’ line-loss costs— the same argument that the Companies raise in Count I and that is precluded by the State Decision, absent an applicable exception. Without a legal determination that their costs were impermissibly “trapped,” the Companies have no basis for asserting an unconstitutional deprivation of any property interest. Because Count II depends entirely on the same issues that were already litigated to finality in the state proceeding, it is foreclosed by issue preclusion.

A similar fate would befall Count III, even if the Companies’ arguments regarding that count were not waived. Count III relates to the constitutionality of the Electric Competition Act as applied to the Companies, to the extent the PUC Order “disregarded] FERC orders or ... interpreted] FERC tariffs” in violation of the filed rate doctrine. (J.A. at 51.) Although the constitutional challenge to the Electric Competition Act was not raised until the PUC made its decision, it depends, like Count II, on the Companies being able to establish that the PUC Order violated the filed rate doctrine. Again, the State Decision expressly held that there was no violation of the filed rate doctrine, so Count III would also be precluded, absent any exception.25

B. Exceptions to the Full Faith and Credit Statute

Although issue preclusion would typically foreclose their claims, the Companies argue that three exceptions to the Full Faith and Credit Statute apply to render the State Decision devoid of any preclusive effect: (1) the state proceeding was legislative rather than judicial in nature; (2) the Companies had a substantially higher burden of persuasion in the Commonwealth Court than they do in this federal action; and (3), under the filed rate doctrine, the PUC and the Commonwealth Court infringed on FERC’s exclusive jurisdiction. [Appellants’ Opening Br. at 24-26.] We are not persuaded that any of those exceptions apply to foreclose the application of issue preclusion in this ease.

1. Whether the state proceeding was legislative or judicial in nature

The Full Faith and Credit Statute, by its terms, applies only to “judicial proceedings.” 28 U.S.C. § 1738. The Companies *354argue that the state proceeding was legislative, not judicial, in nature, so the Full Faith and Credit Statute—and the principles of preclusion that stem from it—do not apply.

The parties do not dispute that the Supreme Court has counseled federal courts to defer to each state’s characterization of its own proceedings. See Okla. Packing Co. v. Okla. Gas & Elec. Co., 309 U.S. 4, 7, 60 S.Ct. 215, 84 L.Ed. 537 (1940) (looking to “[t]he pronouncements of the Oklahoma Supreme Court concerning the character of ... a [prior] determination”); Okla. Natural Gas Co. v. Russell, 261 U.S. 290, 291, 43 S.Ct. 353, 67 L.Ed. 659 (1923) (“The Constitution of Oklahoma [ ] ... gives an appeal to the Supreme Court of the State, acting in a legislative capacity ..., with power to substitute a different order and to grant a supersedeas in the meantime.”); cf. Prentis v. Atl. Coast Line Co., 211 U.S. 210, 226, 29 S.Ct. 67, 53 L.Ed. 150 (1908) (“We shall assume that when [ ] ... a state Constitution sees fit to unite legislative and judicial powers in a single hand, there is nothing to hinder, so far as the Constitution of the United States is concerned.”). In addition, the Supreme Court in New Orleans Public Service, Inc. v. Council of New Orleans (“NOPSI”), 491 U.S. 350, 109 S.Ct. 2506, 105 L.Ed.2d 298 (1989), said that the proper characterization of an agency’s actions “depends not upon the character of the body but upon the character of the proceedings. ... The nature of the final act determines the nature of the previous inquiry.” Id. at 371, 109 S.Ct. 2506 (first alteration in original) (internal quotation marks and citation omitted). NOPSI teaches that

[a] judicial inquiry investigates, declares and enforces liabilities as they stand on present or past facts and under laws supposed already to exist. That is its purpose and end. Legislation on the other hand looks to the future and changes existing conditions by making a new rule to be applied thereafter to all or some part of those subject to its power.

Id. at 370-71, 109 S.Ct. 2506 (internal quotation marks and citation omitted).

The Companies argue that Pennsylvania has not clearly decided whether the Commonwealth Court’s review of a PUC order is legislative or judicial, while the PUC Defendants counter that the Pennsylvania Administrative Law and Procedure Act and the Pennsylvania Judicial Code unequivocally call appellate review of PUC proceedings “judicial.” (PUC Defendants’ Br. at 44.) The District Court concluded that the Commonwealth Court’s review of the PUC Order was judicial in nature because the Commonwealth Court’s authority to review PUC orders under 2 Pa. Cons.Stat. § 704 “is limited to determining whether a constitutional violation, an error of law, or a violation of PUC procedure has occurred and whether necessary findings of fact are supported by substantial evidence.” 26 (J.A. at 30 (quoting Popowsky *355v. Pa. Pub. Util. Comm’n, 589 Pa. 605, 910 A.2d 38, 48 (2006)) (internal quotation marks omitted).) The District Court further found support for the judicial nature of the proceeding in “the Commonwealth Court[’s] reli[ance] upon past facts (as found in the proceeding before the [] PUC) and existing law (as the Commonwealth Court interpreted it) to resolve a challenge to the legality of a prior action (the [ ] PUC ... Order).” (Id.)

The Companies contend that the District Court’s reasoning was erroneous because “[i]t cannot be true that [the] commonplace standard of agency review—one that applies to both ratemaking and non-ratemak-ing agencies alike—makes the Commonwealth Court’s decision here judicial.” (Appellants’ Opening Br. at 51.) In other words, they argue that the scope of the Commonwealth Court’s review, alone, cannot determine whether such review is judicial or legislative in nature. That argument fails, however, because the scope of agency review is not the sole basis for concluding that the State Decision was judicial rather than legislative. Other aspects of the state proceeding also indicate that it was judicial in nature.

The Companies rely on two Pennsylvania cases from the 1950s to argue that Pennsylvania courts consider their review of a state agency’s rate-making to be legislative in nature. The two are a 1954 Pennsylvania Superior Court case, Duquesne Light Co. v. Pennsylvania Public Utility Commission, which includes the comment that “[r]ate making is an exercise of the legislative power, delegated to the Commission,” 176 Pa.Super. 568, 107 A.2d 745, 755 (1954), and a Pennsylvania Supreme Court opinion from 1956, Pennsylvania State Chamber of Commerce v. Torquato, that says “[t]he [United States Supreme] Court has permitted resort to a federal court of equity where a state was enforcing confiscatory rates and by its law precluded a stay ... until the state courts ‘acting in a legislative capacity’ had taken final action,” 386 Pa. 306, 125 A.2d 755, 763 (1956) (quoting Aircraft & Diesel Equip. Corp. v. Hirsch, 331 U.S. 752, 773 n. 38, 67 S.Ct. 1493, 91 L.Ed. 1796 (1947)). Du-quesne, however, relates to the nature of certain PUC rate-making; it does not dictate that all PUC actions are legislative in nature, let alone hold that the Commonwealth Court’s review of a PUC decision is a legislative act. And the Pennsylvania Supreme Court in Torquato simply recognized that a state court acting in a legislative capacity does not necessarily establish precedent that prevents resort to a federal court; it did not hold that review of a PUC action is by definition legislative.

We recognized in Kentucky West Virginia Gas Co. v. Pennsylvania Public Utility Commission, 837 F.2d 600 (3d Cir. 1988), that PUC proceedings may be judicial in nature: “When a state agency acting in a judicial capacity resolves disputed issues of fact properly before it which the parties have had an adequate opportunity to litigate, federal courts must give the agency factfinding the same preclusive effect to which it would be entitled in the state’s courts.” Id. at 611 (emphasis added). Moreover, Pennsylvania law recognizes that PUC action and subsequent court review can be judicial in nature. See 2 Pa. Cons.Stat. Ann. § 704 (Pennsylvania Administrative Law and Procedure Act describing the various dispositions when a court reviews a state agency’s “adjudication”). As the PUC Defendants point out, PUC decisions can be “the product of a quasi-judicial, on-the-record proceeding that includes a presiding ALJ who has the *356power to administer oaths, conduct eviden-tiary hearings, allow for cross-examination, rule on motions, review briefs submitted by the parties, and issue recommended decisions with findings of fact and conclusions of law.” (PUC Defendants’ Br. at 44 (citing 66 Pa. Cons.Stat. Ann. § 331; 2 Pa. Cons.Stat. Ann. §§ 504-507).) By implication, if a state agency proceeding is judicial, appellate review of that proceeding is also judicial.

A straightforward application of the distinction between judicial and legislative inquiry outlined in NOPSI confirms that the Commonwealth Court decision at issue here is judicial in nature. As the District Court held, “the Commonwealth Court of Pennsylvania did not conduct an independent, forward-looking ... investigation.” (J.A. at 30.) Instead, the Commonwealth Court, like the PUC, referred to and endeavored to enforce (whether correctly or not is immaterial at this juncture) the preexisting Settlement Agreement. The Commonwealth Court further made a determination specific to the Companies. It determined that there was no violation of the filed rate doctrine with respect to how the PUC required the Companies to classify their line losses, which involved a review of the record regarding how the Companies, specifically, had treated line losses in the past. At bottom, both the PUC and the Commonwealth Court adjudicated the adversarial dispute between the Customer Groups and the Companies after considering those parties’ respective legal arguments. We have no difficulty holding that the state proceeding was judicial, not legislative. The nature of the state proceeding therefore does not bar the application of issue preclusion in this case.

2. Whether the Companies’ burdens before the Commonwealth Court and in the instant case are different

The Companies also argue that the so-called “difference-in-burden exception” bars giving the State Decision any preclusive effect. (Appellants’ Opening Br. at 44.) They rely on Section 28(4) of the Restatement (Second) of Judgments, which states that preclusion does not apply when

[t]he party against whom preclusion is sought had a significantly heavier burden of persuasion with respect to the issue in the initial action than in the subsequent action; the burden has shifted to his adversary; or the adversary has a significantly heavier burden than he had in the first action.

Restatement (Second) of Judgments § 28(4) (1982). The Companies do not argue that the burden of proof ever shifted to their adversaries. (See Oral Argument Transcript (“Tr.”) at 29:11-12 (“We have the burden—either way we have the burden.”).) Rather, they argue that, in reviewing the PUC Order, the Commonwealth Court applied the wrong standard of review and placed a substantially more onerous burden of persuasion on them than the Companies would face in this action. The PUC Defendants respond by arguing that “the use of the [difference-in-burden] exception is not ‘well-established’ in relevant ease law,” and that, in any event, the Companies confuse the concept of a party’s burden of proof with a court’s standard of review. (PUC Defendants’ Br. at 42.)

According to the Companies, Section 28(4) of the Restatement is well-established because it provides the basis for the axiomatic rule that, “ ‘even when the parties are the same, an acquittal in a criminal proceeding is not conclusive in a subsequent civil action arising out of the same event.’ ” (Appellants’ Opening Br. at 44 (quoting Restatement (Second) of Judgments § 28 cmt. f).) A comment to Sec*357tion 28 of the Restatement explains that, “[t]o apply issue preclusion in the cases described in Subsection (4) would be to hold, in effect, that the losing party in the first action would also have lost had a significantly different burden been imposed.” Restatement (Second) of Judgments § 28 emt. f.

However, we need not decide whether Pennsylvania recognizes the difference-in-burden exception, wherein a party that lost on an issue in a first proceeding is nevertheless permitted to relitigate the issue in a second proceeding if its burden of proof is lower in the second proceeding.27 Assuming such an exception exists in Pennsylvania law, the Companies have failed to show any relevant difference in burden here. They argue that a federal district court reviews an issue of federal pre-emption de novo as a question of law, whereas the Commonwealth Court afforded deference to the PUC’s factual findings underlying the determination that line losses are not a transmission cost. Contrary to the Companies’ position, the District Court was not reviewing the merits of the PUC Order, so it makes little sense to speak of the Companies’ burden of persuasion in the District Court in terms of de novo “review.” What the Companies point to is the Commonwealth Court’s use of an allegedly incorrect standard of review, not a change in their own burden of proof on the merits. To the extent the Companies complain that the Commonwealth Court applied the incorrect standard of review, that argument was something to be remedied on direct appeal, not something that opens the PUC Order to collateral attack in federal court.28 See Del. River Port Auth. v. Fraternal Order of Police, 290 F.3d 567, 576 (3d Cir.2002) (“Error in a prior judgment is not a sufficient ground for refusing to give it preclusive effect.”). The Companies’ reliance on the Restatement (Second) of Judgments § 28(4) is therefore unpersuasive.

3. Whether the PUC and the Commonwealth Court were without jurisdiction

That brings us to the Companies’ only remaining argument that the State Decision lacks any preclusive effect: that “[t]he PUC and [the] Commonwealth Court lacked subject matter jurisdiction to construe the nature of new charges imposed by a FERC transmission tariff.” (Appellants’ Opening Br. at 24.) We have recognized that Pennsylvania’s preclusion law appears to require subject matter jurisdiction in the first proceeding for a decision made in that proceeding to have preclusive effect, McCarter v. Mitcham, 883 *358F.2d 196, 199 (3d Cir.1989), and the PUC Defendants do not dispute that jurisdiction is a prerequisite to the application of issue preclusion in this case.

To be clear, the Companies’ position is that the State Decision is “void ab initio for want of subject matter jurisdiction and not merely voidable as wrongly decided on the merits.” (Appellants’ Supp. Br. at 10.) They argue that the PUC and the Commonwealth Court “invaded th[e] exclusive federal scheme [of power regulation] by purporting to reclassify FERC-mandated interstate transmission rates as generation charges.” (Appellants’ Opening Br. at 32.) In other words, the Companies’ jurisdictional argument is premised on the outcome of the merits in the state proceeding being adverse to them. Notably, they do not dispute that the Commonwealth Court had jurisdiction to consider the import of the filed rate doctrine to the classification of line losses. (Id. at 33-34 (“The Companies did not contend that the Commonwealth Court lacked subject matter jurisdiction to address the Companies’ filed rate doctrine claim.”).) They only dispute that the PUC and the Commonwealth Court had jurisdiction to say they lose.

We begin by emphasizing “the limited scope of review one court may conduct to determine whether a foreign court had jurisdiction to render a challenged judgment.” Underwriters Nat’l Assurance Co. v. N.C. Life & Accident & Health Ins. Guar. Ass’n, 455 U.S. 691, 706, 102 S.Ct. 1357, 71 L.Ed.2d 558 (1982). Generally, when fully and fairly litigated to finality, “a tribunal’s determination of its own jurisdiction is accorded the same status for issue preclusion purposes as the merits of a dispute.” Crossroads Cogeneration Corp. v. Orange & Rockland Utils., 159 F.3d 129, 135 (3d Cir.1998); see also Durfee v. Duke, 375 U.S. 106, 111, 84 S.Ct. 242, 11 L.Ed.2d 186 (1963) (“[A] judgment is entitled to full faith and credit—even as to questions of jurisdiction—when ... those questions have been fully and fairly litigated and finally decided in the court which rendered the original judgment.”).

With respect to its jurisdiction, the PUC held:

[I]t is within the [PUC’s] discretion whether and how to allocate costs via [the Transmission Rider] or otherwise. And, we believe it is unreasonable to suggest that the [PUC] is required to rubber stamp recovery of such costs simply because they are imposed by PJM, even when the Companies voluntarily (and properly) sought approval of their recovery from [the PUC] acting within its jurisdiction to set just and reasonable retail rates for jurisdictional transmission and distribution facilities.

(J.A. at 154.) In short, the PUC concluded that it had jurisdiction not only to consider how to classify line losses for the Companies’ retail rate structure but also to resolve the classification of costs under the Settlement Agreement as it did. As the Companies have conceded, they challenged the PUC’s exercise of jurisdiction on direct appeal to the Commonwealth Court and lost. (See Appellants’ Supp. Br. at 10 (“The basis for the Companies’ appeal— forum and field preemption under the FPA and filed rate doctrine—was jurisdictional, not factual.”).)

Under Pennsylvania law, the Commonwealth Court has “jurisdiction of appeals from final orders of ... the [PUC].” 42 Pa. Cons.Stat. Ann. § 763(a). The Commonwealth Court affirmed the PUC, holding that the PUC Order “was not inconsistent with FERC precedent, did not violate the Filed Rate Doctrine, and did not improperly prevent [the] Companies from recovering trapped costs.” (J.A. at 191.) On application for discretionary review to both the Pennsylvania Supreme Court and the *359United States Supreme Court, the Companies again argued that the state tribunals lacked authority to decide the matter adversely to the Companies. (Tr. at 22:6-11 (confirming that the Companies’ petitions for discretionary review in the state proceeding sought a determination that the PUC lacked authority to make the decision that it did).) Both courts denied discretionary review, and the PUC’s determination of its own jurisdiction stood as final. Typically, we would afford that determination of jurisdiction preclusive effect, and that would be the end of it.

The Companies, however, submit that their argument raises a question that we reserved in Crossroads Cogeneration v. Orange & Rockland: whether “an exception to the rule [of according preclusive effect to a tribunal’s determination of jurisdiction] applies in a case ... where a federal statute ... preempts [a] state agency from acting altogether.” 159 F.3d at 135. But we again do not need to reach that question because we conclude that, contrary to the Companies’ position, the PUC and the Commonwealth Court were not divested of jurisdiction to act altogether in the state proceeding.

a. Whether the state tribunals have been divested of jurisdiction

The Companies maintain that the result of the state proceeding is void for lack of jurisdiction, and it is true that “[a] void judgment is a legal nullity.” United Student Aid Funds, Inc. v. Espinosa, 559 U.S. 260, 270, 130 S.Ct. 1367, 176 L.Ed.2d 158 (2010). To be deemed void, a judgment must be “so affected by a fundamental infirmity that the infirmity may be raised even after the judgment becomes final.” Id. “Federal courts considering [whether] a judgment is void because of a jurisdictional defect generally have reserved relief only for the exceptional case in which the court that rendered judgment lacked even an arguable basis for jurisdiction.” Id. at 271, 130 S.Ct. 1367 (internal quotation marks and citation omitted) (discussing a motion filed under Rule 60(b)(4) of the Federal Rules of Civil Procedure to render a judgment void).

Showing that a state tribunal lacked even an arguable basis for jurisdiction over a federal question is difficult because, under the principles of federalism, there is a “deeply rooted presumption in favor of concurrent state court jurisdiction.” Tafflin v. Levitt, 493 U.S. 455, 459, 110 S.Ct. 792, 107 L.Ed.2d 887 (1990). Federal and state law “together form one system of jurisprudence, which constitutes the law of the land for the State; and the courts of the two jurisdictions are ... courts of the same country, having jurisdiction partly different and partly concurrent.” Claflin v. Houseman, 93 U.S. 130, 137, 23 L.Ed. 833 (1876). The concurrent jurisdiction of the States is “subject only to limitations imposed by the Supremacy Clause.” Tafflin, 493 U.S. at 458, 110 S.Ct. 792; see also Del. River Port Auth., 290 F.3d at 576 (noting that it is well-settled that “[s]tate courts may answer federal questions”). Indeed, “[s]o strong is the presumption of concurrency that it is defeated only in two narrowly defined circumstances: first, when Congress expressly ousts state courts of jurisdiction, and second, ‘[w]hen a state court refuses jurisdiction because of a neutral state rule regarding the administration of the courts.’ ”29 Haywood v. Drown, 556 U.S. *360729, 735, 129 S.Ct. 2108, 173 L.Ed.2d 920 (2009) (second alteration in original) (citations omitted). The second circumstance is not relevant here, so we focus on the first, which is typically stated in unmistakable terms:

In the standard fields of exclusive federal jurisdiction, the governing statutes specifically recite that suit may be brought “only” in federal court, Investment Company Act of 1940, as amended, 84 Stat. 1429, 15 U.S.C. § 80a-35(b)(5); that the jurisdiction of the federal courts shall be “exclusive,” Securities Exchange Act of 1934, as amended, 48 Stat. 902, 15 U.S.C. § 78aa; Natural Gas Act of 1938, 52 Stat. 833, 15 U.S.C. § 717u; Employee Retirement Income Security Act of 1974, 88 Stat. 892, 29 U.S.C. § 1132(e)(1); or indeed even that the jurisdiction of the federal courts shall be “exclusive of the courts of the States,” 18 U.S.C. § 3231 (criminal cases); 28 U.S.C. §§ 1333 (admiralty, maritime, and prize eases), 1334 (bankruptcy cases), 1338, (patent, plant variety protection, and copyright cases), 1351 (actions against consuls or vice consuls of foreign states), 1355 (actions for recovery or enforcement of fine, penalty, or forfeiture incurred under Act of Congress), 1356 (seizures on land or water not within admiralty and maritime jurisdiction).

Tafflin, 493 U.S. at 471,110 S.Ct. 792.

The Companies are correct that the FPA grants FERC exclusive jurisdiction over certain matters, but the relevant question here is whether Congress divested state utility agencies or state courts of jurisdiction to hear cases requiring an adjudication of the filed rate doctrine’s scope, and the answer to that is no. The FPA plainly leaves a role for states in electricity regulation.30 While section 201(b) of the FPA grants federal regulatory authority as to “the transmission of electric energy in interstate commerce and to the sale of electric energy at wholesale in interstate commerce,”31 16 U.S.C. § 824(b)(1), at the *361same time, it provides that federal regulation is “to extend only to those matters which are not subject to regulation by the States,” id. § 824(a). Thus, in enacting the FPA, Congress expressly envisioned a role for state utility agencies in electricity regulation, which may well require consideration of the import of the filed rate doctrine. Cf. Crossroads, 159 F.3d at 135 (“Given the substantial role given state utility agencies by Congress in enacting [the Public Utility Regulatory Policies Act], we conclude Congress did not intend to prevent application of common law rules of preclusion.”).

Nevertheless, the Companies submit that the PUC and Commonwealth Court so exceeded the scope of their authority under the “preemptive force of the federal regulatory scheme” of the FPA and the filed rate doctrine that those tribunals utterly lacked jurisdiction. (Appellants’ Opening Br. at 29.) The Companies point out that a federal statute or regulation may pre-empt state regulation in three ways. First, under express preemption, Congress can preempt state law by explicit statutory language. Barnett Bank of Marion Cnty., N.A. v. Nelson, 517 U.S. 25, 31, 116 S.Ct. 1103, 134 L.Ed.2d 237 (1996). Second, under field pre-emption, Congress can enact a regulatory scheme “so pervasive as to make reasonable the inference that Congress left no room for the States to supplement it.” Id. at 31, 116 S.Ct. 1103 (internal quotation marks and citation omitted). And third, “federal law may be in ‘irreconcilable conflict’ with state law,” which creates what is known as conflict pre-emption.32 Id. (internal quotation marks and citation omitted). The Companies have cast their net widely, arguing that “[t]his case concerns all three” types of pre-emption.33 (Appellants’ Supp. Br. at 1.) Not ones to shy from emphatic declarations, they submit that the filed rate doctrine is “a uniquely *362sweeping and clear manifestation of field preemption” that divests states of jurisdiction to classify line losses as generation costs in a retail rate structure. (Appellants’ Opening Br. at 29.) We cannot concur.

As we have recently noted, preemption arguments do not ordinarily raise issues of subject matter jurisdiction. Harris v. Kellogg Brown & Root Servs., Inc., 724 F.3d 458, 464 n. 1 (3d Cir.2013) (“[W]e must clarify that our prior decision did not imply ... that Rule 12(b)(1) [of the Federal Rules of Civil Procedure] is the right vehicle for ordinary preemption arguments.”). That is because “[preemption arguments, other than complete preemption, relate to the merits of the case.” Id. (citing In re U.S. Healthcare, Inc., 193 F.3d 151, 160 (3d Cir.1999)); see also Joyce v. RJR Nabisco Holdings Corp., 126 F.3d 166, 171 (3d Cir.1997) (pointing out the distinction “between the complete preemption doctrine for jurisdictional purposes and ordinary preemption, which merely constitutes a defense to a state law cause of action”).

While the Supreme Court has said that “[d]octrines of federal pre-emption ... may in some contexts be controlling” over “the general rule of finality of jurisdictional determinations,” Durfee, 375 U.S. at 114, 84 S.Ct. 242, this case does not present such an exception. In the Atlantic City decision on which the Companies so heavily rely, FERC required PJM to factor marginal line losses into the energy price at each location. Atlantic City I, 115 FERC at 61,473-74. Certain FERC language from that decision certainly does highlight the connection between line losses and the transmission of electricity. See, e.g., id. at 61,473 (“As in the case of all electric transmission, there is some loss ... as ... power is transmitted from the point of generation to the point of delivery.”). But the agency did not say that line losses should be categorized as a cost of transmission, and indeed it made comments that can be read as supporting the view that line losses could be understood as a factor in electricity generation. It noted, for example, that “[s]uch loss[es] result [ ] in a cost PJM incurs to maintain the level of the scheduled power and to deliver it under conditions of system reliability.” Id. at 61,473.34 In the end, the FERC orders that the parties point us to require PJM to calculate line losses in a certain way but do not make the kind of categorical statements that lead to preemption and override the finality of the state ruling the Companies themselves sought. That is in sharp contrast with a case like Nantahala Power & Light Co. v. Thornburg, in which the Supreme Court held that FERC’s express allotment of entitlement power to two owners of hydroelectric power plants pre-empted a state agency’s retail rate-making order allocating entitlement power differently. 476 U.S. at 955, 958, 106 S.Ct. 2349.

The Companies also try to rely on “complete pre-emption,” which is jurispruden-tially distinct from the three “ordinary” types of pre-emption—express, field, and conflict pre-emption—described above. Ry. Labor Exec. Ass’n v. Pittsburgh & Lake Erie R.R. Co., 858 F.2d 936, 939 (3d Cir.1988); see also Lontz v. Tharp, 413 F.3d 435, 440 (4th Cir.2005) (“[W]e may not conflate ‘complete preemption’ with ... ‘ordinary’ pre-emption. While these two concepts are linguistically related, *363they are not as close kin jurisprudentially as their names might suggest. Complete preemption is a ‘jurisdictional doctrine,’ while ordinary preemption simply declares the primacy of federal law, regardless of the forum or the claim.”). Under complete preemption, “the pre-emptive force of a statute is so ‘extraordinary’ that it ‘converts’ ” a state-law complaint into a federal one. Caterpillar Inc. v. Williams, 482 U.S. 386, 393, 107 S.Ct. 2425, 96 L.Ed.2d 318 (1987) (quoting Metro. Life Ins. Co. v. Taylor, 481 U.S. 58, 65, 107 S.Ct. 1542, 95 L.Ed.2d 55 (1987)).

Complete pre-emption, however, stands as a limited exception to the well-pleaded complaint rule, i.e., the rule that “a case may not be removed to federal court on the basis of a federal defense, including the defense of pre-emption, even if the defense is anticipated in the plaintiffs complaint, and even if both parties concede that the federal defense is the only question truly at issue.” Id. Complete pre-emption, in other words, arises in the context of removal jurisdiction. It serves as a basis for federal jurisdiction over causes of action that may appear, on their face, to be based on state law but that are in truth only actionable under federal law due to Congress’s clear intent “to completely pre-empt a particular area of law.” U.S. Healthcare, 193 F.3d at 160 (internal quotation marks and citation omitted). It does not resolve whether state tribunals have been wholly divested of jurisdiction to hear the federal cause of action.35 Cf. Avco Corp. v. Aero Lodge No. 735, Int'l Ass’n of Machinists & Aerospace Workers, 390 U.S. 557, 560 n. 2, 88 S.Ct. 1235, 20 L.Ed.2d 126 (1968) (recognizing that a state court may retain jurisdiction over an action that is completely pre-empt-ed if the defendant does not elect to have the case removed to federal court). The Companies have cited no cases to indicate otherwise. Perhaps recognizing that the doctrine is not the best vehicle for their argument, they did not even raise complete pre-emption, used as a term of art, until oral argument. (Tr. at 5:22-24 (saying that field pre-emption and conflict preemption in this case “add up to complete preemption”).) That has the look of a waiver, but even assuming, arguendo, that the Companies have not waived their argument, complete pre-emption has no place in this discussion.

Furthermore, history matters here. The Supreme Court has recognized, without indicating that there were any jurisdictional defects, that “state courts have examined th[e] interplay [of the filed rate doctrine] in determining the effect of FERC-approved wholesale power rates on retail rates for electricity.” Nantahala, 476 U.S. at 964-65, 106 S.Ct. 2349; see also Gen. Motors Corp. v. Ill. Commerce Comm’n, 143 Ill.2d 407, 158 Ill.Dec. 537, 574 N.E.2d 650, 655 (1991) (deciding *364whether a state utility agency’s action violated the filed rate doctrine); Me. Yankee Atomic Power Co. v. Me. Pub. Utils. Comm’n, 581 A.2d 799, 804-05 (Me.1990) (same); Pa. Power Co. v. Pa. Pub. Util. Comm’n, 127 Pa.Cmwlth. 97, 561 A.2d 43, 49-52 (1989) (same). Binding precedent instructs that, “when a state proceeding presents a federal issue, even a pre-emption issue, the proper course is to seek resolution of that issue by the state court.” Chick Kam Choo v. Exxon Corp., 486 U.S. 140, 149-50, 108 S.Ct. 1684, 100 L.Ed.2d 127 (1988). Thus, despite the Companies’ attempt to craft a way for us to review whether the State Decision complies with their interpretation of the FPA and the filed rate doctrine, we cannot say that the PUC and the Commonwealth Court “lacked even an arguable basis for jurisdiction,” Espinosa, 559 U.S. at 270, 130 S.Ct. 1367, to decide the merits of classifying line losses for purposes of a retail rate structure. As the PUC and the Commonwealth Court were not divested of authority to act altogether, the result of the state proceeding is not void on that ground.

b. Whether the state proceedings were an impermissible “collateral attack” on a FERC decision

The Companies also argue that the FPA explicitly proscribes the state agencies and courts, as improper forums, from resolving the dispute between the Companies and the Customer Groups such that the state proceedings were an impermissible “collateral attack” on a FERC decision. The United States Supreme Court in City of Tacoma v. Taxpayers of Tacoma, 357 U.S. 320, 78 S.Ct. 1209, 2 L.Ed.2d 1345 (1958), held that, pursuant to FPA § 313(b), “Congress ... prescribed the specific, complete and exclusive mode for judicial review of the Commission’s orders,” which consists of direct review by a federal circuit court of appeals and, possibly, the United States Supreme Court.36 Id. at 336, 78 S.Ct. 1209. Direct review of FERC’s orders

necessarily preclude^] de novo litigation between the parties of all issues inhering in the controversy, and all other modes of judicial review. Hence, upon judicial review of the Commission’s order, all objections to the order, to the license it directs to be issued, and to the legal competence of the licensee to execute its terms, must be made in the Court of Appeals or not at all.

Id. (footnote omitted). Emphasizing that the rule bars tribunals—with the exception of federal circuit courts and the United States Supreme Court—from hearing di*365rect challenges to FERC orders, the Companies claim it shows a jurisdictional deficiency with the state proceeding. Their argument is akin to what we have referred to as “forum pre-emption”:

When Congress intends a particular forum to have exclusive jurisdiction ..., that policy decision deprives other fora of subject matter jurisdiction. This doctrine of “forum preemption” implements Congressional determinations that development of the substantive law in a particular area should be left to a particular administrative agency created for that purpose.

Ry. Labor Execs. Ass’n v. Pittsburgh & L.E.R. Co., 858 F.2d 936, 943 (3d Cir.1988); see also Int’l Longshoremen’s Ass’n v. Davis, 476 U.S. 380, 388, 106 S.Ct. 1904, 90 L.Ed.2d 389 (1986) (“It is clearly within Congress’ powers to establish an exclusive federal forum to adjudicate issues of federal law in a particular area that Congress has the authority to regulate under the Constitution.”).

The Companies argue that, to the extent the Customer Groups had any grievances regarding the proposed fine losses, they could and should have brought their grievances in a federal court of appeals on direct appeal of a FERC order, rather than waiting to contest the Companies’ proposed rates before the PUC in a separate proceeding. However, the issue in the state proceeding—whether the Companies could classify line losses as transmission charges—was not an issue arising from any FERC order that the Companies have identified. To the extent the Companies complain that the Customer Groups should have directly appealed the Atlantic City decision, their argument is misplaced. The Customer Groups did not challenge how FERC has mandated PJM to ealcu-late its line losses. If anything, the classification of the Companies’ line-loss costs for retail billing was an issue made relevant by the voluntarily agreed-upon terms of the Settlement Agreement, which provided different end dates on transmission rate caps and generation rate caps.37

c. Conclusion on state jurisdiction

Ultimately, for purposes of jurisdiction, we need not resolve whether the Companies are correct that their interpretation of line losses is required under FERC’s regulatory scheme or that the Commonwealth Court improperly deferred to certain aspects of the PUC Order. Cf. Decker v. Nw. Envtl. Defense Ctr., — U.S. -, 133 S.Ct. 1326, 1335, 185 L.Ed.2d 447 (2013) (“For jurisdictional purposes, it is unnecessary to determine whether [the respondent] is correct in arguing that only its readings of the [relevant] Rule is permitted under the [Clean Water] Act.”); Avco, 390 U.S. at 561, 88 S.Ct. 1235 (“Any error in granting ... relief does not go to the jurisdiction of the court.” (internal quotation marks and citation omitted)).

The Companies have not cited a single instance in which a party has been allowed to litigate a substantive issue all the way through the state courts and a petition for certiorari to the United States Supreme Court and then subsequently argue that the state courts lacked jurisdiction in the first place. The closest case is Southern Union Co. v. FERC, 857 F.2d 812 (D.C.Cir.1988), in which the United States Court of Appeals for the District of Columbia Circuit declined to apply issue preclusion “with its full rigor” and decided that a state court had no power to enforce a damage award that effectively awarded a price for interstate gas that was under *366FERC’s exclusive jurisdiction. Id. at 816. However, Southern Union is distinguishable because the D.C. Circuit’s rationale for not applying issue preclusion rested on “the distinct possibility that the [United States Supreme] Court may have declined to issue ... [a] writ [of certiorari] in deference to the pendency of the proceedings [in FERC].” Id. We have no such indication here prompting us to set aside the result of a state proceeding that has been litigated to finality and denied review by the United States Supreme Court.38

The Companies also cite several Supreme Court decisions in which actions by state utility agencies were held to be preempted by FERC actions. See Entergy La., Inc. v. La. Pub. Serv. Comm’n, 539 U.S. 39, 123 S.Ct. 2050, 156 L.Ed.2d 34 (2003); Miss. Power & Light, 487 U.S. at 356-57, 108 S.Ct. 2428; Nantahala, 476 U.S. at 955, 106 S.Ct. 2349. But those decisions were all made on direct review from state agency decisions. Entergy, 539 U.S. at 49-50, 123 S.Ct. 2050; Miss. Power & Light, 487 U.S. at 373-75, 108 S.Ct. 2428; Nantahala, 476 U.S. at 970-72, 106 S.Ct. 2349. Here, the United States Supreme Court denied discretionary review, rendering the State Decision final. We have held that, “[i]f [a state tribunal] answered federal questions erroneously, it remained for state appellate courts, and ultimately for the United States Supreme Court, to correct any mistakes. Error in a prior judgment is not a sufficient ground for refusing to give it preclusive effect.”39 Del. River Port Auth., 290 F.3d at 576. The Supreme Court’s decisions in Entergy, Mississippi Power & Light, and Nantaha-la support the conclusion that any error in the application of the filed rate doctrine should have been corrected on direct appeal of the PUC Order.

Moreover, “[t]here is ... no reason to believe that Congress intended to provide a person claiming a federal right an unrestricted opportunity to relitigate an issue already decided in state court simply because the issue arose in a state proceeding in which he would rather not have been engaged at all.” San Remo Hotel, L.P. v. San Francisco, 545 U.S. 323, 343, 125 S.Ct. 2491, 162 L.Ed.2d 315 (2005) (internal quotation marks and citation omitted). Here, the Companies have even less reason to complain, as they affirmatively chose to litigate their case through the state system. They admit that “[t]here was nothing preventing [them] from going to FERC” and that, had they obtained a favorable ruling from FERC, they could have enforced it. (Tr. at 7:3—8:9; see also id. at 21:8-19 (stating that the Companies could go to FERC, even at this point).) In other words, the Companies chose their forum for litigation and lost. As the Supreme Court has stated,

*367[pjublic policy [ ] ... dictates that there be an end of litigation; that those who have contested an issue shall be bound by the result of the contest; and that matters once tried shall be considered forever settled as between the parties. We see no reason why this doctrine should not apply in every case where one voluntarily appears, presents his case and is fully heard, and why he should not, in the absence of fraud, be thereafter concluded by the judgment of the tribunal to which he has submitted his cause.

Durfee, 375 U.S. at 111-12, 84 S.Ct. 242 (internal quotation marks and citation omitted).40

The Companies could have withdrawn their federal issues from the state proceeding and brought them in federal court, as has been done before. See Ky. W. Va. Gas v. Pa. Pub. Util. Comm’n, 837 F.2d 600, 604 n. 2 (3d Cir.1988) (noting that a gas utility that had appealed a PUC denial of a pass-through rate to the Commonwealth Court had withdrawn its constitutional claims from the state proceeding and brought them in federal court). The only reason the Companies proffered for not withdrawing their federal pre-emption issues was that they had to keep those issues before the Commonwealth Court to complete “[t]he legislative process.” (Tr. at 23:1-2.) As we have explained, however, the state proceeding was a judicial process, not a legislative one, and the Companies’ excuses now for not pursuing their claims in federal court in the first instance have the ring of post-hoc rationalization.41

In the end, we are compelled to reject the Companies’ efforts to pose their merits-based pre-emption arguments—the same ones that were rejected in the State Decision—as jurisdictional arguments. They would like, as the saying goes, to have it both ways—if they had obtained approval to charge their customers line-loss costs as a transmission cost, the PUC and the Commonwealth Court would have had jurisdiction to approve their proposed rates; otherwise, as they perceive it, the PUC and the Commonwealth Court must lack jurisdiction, and the Companies get a “do-over” with a clean slate in federal court. It is the classic “heads I win, tails you lose” approach to dispute resolution.42 *368(Tr. at 5:9.) And it must fail because there is no sound justification for a rule that provides for jurisdiction in a state tribunal only when a pre-ordained merits outcome is reached by that tribunal.

IV. CONCLUSION

The Companies chose to challenge the PUC Order on direct appeal, and they must abide by the result.43 The operative concern before us is not whether the result of the state proceeding “got it right” but whether the Companies litigated the merits of the underlying issues legitimately and to finality. They did. To refuse to give the State Decision any preclusive effect would be a violation of the Full Faith and Credit Statute, which we cannot endorse. Cf. Underwriters Nat’l, 455 U.S. at 694, 102 S.Ct. 1357 (concluding that a state court’s refusal to accord preclusive effect to another state’s prior judgment was a violation of the Full Faith and Credit Clause and its implementing federal statute).

We will therefore affirm the District Court’s dismissal of the Companies’ amended complaint.