Pacific Gas & Electric Co. v. Lynch

216 F. Supp. 2d 1016, 53 Fed. R. Serv. 3d 1376, 2002 U.S. Dist. LEXIS 13895, 2002 WL 1821474

• Court: District Court, N.D. California
• Decided: July 25, 2002
• Docket: C-01-3023 VRW
• Status: Published

Opinion

ORDER

WALKER, District Judge.

By its complaint, Pacific Gas and Electric Company (PG & E) attacks the regulatory restructuring scheme California developed and later implemented for its electrical public utilities. PG & E names as defendants Loretta M Lynch, Henry M Duque, Carl W Wood, Geoffrey F Brown and Michael R Peevey 1 , in their„ official capacities as Commissioners of the California Public Utilities Commission (CPUC). PG & E has moved for summary judgment on its first and second claims for relief (Doc # 111) while defendants have filed cross-motions for summary judgment or, in the alternative, for partial summary judgment on PG & E’s preemption claims. Doc # 104. Applicant in intervention The Utility Reform Network (TURN) also moves for summary judgment against PG & E. Doc #119.

I

A

PG & E filed its original complaint against defendants on November 8, 2000, in this court, bringing the same claims as in its present complaint, with the addition of an equal protection claim. See Compl in PG & E v. Lynch, et al, No C-00-4128 (SBA) (NDCal), in PRJN I (Doc # 49, Exh # 3). That action was subsequently transferred to Judge Lew in the Central District of California, who was presiding over a similar lawsuit filed by Southern California Edison (SCE). After PG & E amended its complaint, defendants moved to dismiss. See Def Mot in PG & E v. Lynch, et al, No C-01-1083-RSWL (SHx) (CDCal), in PRJN I (Doc # 50, Exh # 11). On May 2, 2001, Judge Lew granted defendants’ motion to dismiss without prejudice on ripeness grounds, because “many of the decisions to which PG & E refers in its [first amended complaint] as violating federal law are non-final interim orders that will become final upon a grant or denial of rehearing.” 5/2/01 Order at 38, attached in PRJN I (Doc # 49, Exh # 2). Judge Lew noted that PG & E could refile its complaint once the “CPUC interim orders it challenges become final decisions.” Id at 39.

On August 8, 2001, PG & E filed the instant action before Judge Hamilton in this court, asserting that two of the orders central to its complaint had become final under state law. See Compl in PG & E v. Lynch, et al, No C-01-3023-PJH (Doc # 1). On September 24, 2001, defendants moved to dismiss PG & E’s complaint. Doc # 24. Also on September 24, TURN moved to intervene and to dismiss PG & E’s complaint. Docs ## 18 and 20. On December 18, 2001, the undersigned determined that C-01-3023-PJH was related to a bankruptcy appeal brought by PG & E and pending before the undersigned, C-01-2490-VRW, and C-01-3023-PJH was reassigned to the undersigned.

The court heard oral argument on defendants’ and TURN’S motions to dismiss on February 7, 2002. See Doc # 85. At the March 7, 2002, case management conference, the court determined that the court’s consideration of the issues raised would benefit from a further development of the record and set a hearing date on any summary judgment motions for May 24, 2002. See Doc #90.

B

The instant action is one of many filed in response to California's attempt to restructure its regulatory scheme for the generation and sale of electricity. As codified in Assembly Bill 1890 (AB 1890), California’s restructuring reflected the CPUC’s determination that:

the interests of the ratepayers and the state as a whole will be best served by moving from the regulatory framework *** in which retail electricity service is provided principally by electrical corporations subject to an obligation to provide ultimate consumers in exclusive service territories with reliable electric service at regulated rates, to a framework under which competition would be allowed in the supply of electric power and customers would be allowed to have the right to choose their supplier of electric power.

CalPubUtilCode § 330.

California’s restructuring scheme involved the creation of two new non-governmental corporations to orchestrate the transmission and sale of electricity, organized under California law but regulated by the Federal Energy Regulatory Commission (FERC): the Independent System Operator (ISO) and the California Power Exchange (PX).

Before it ceased operation, the PX operated a continuous state-wide auction, matching bids for the sale and purchase of wholesale electricity.. Bidders of supply into the PX included independent generators of electricity, and other entities that had purchased electricity from such generators for resale. The PX matched these supply bids with requirements for the delivery of electricity, as expressed by demand bids from buyers. Starting in July 1999, a division of the PX, CalPX Trading, operated a block forward market (BFM), an exchange that matched bids to buy specified amounts of power for various time periods with offers to sell power for the same periods in advance of the contracted delivery date. BFMs provided a degree of predictability in the future cost of power.

The ISO, which continues to operate, assumed control over the transmission systems of all three of California’s investor-owned utilities (IOUs): PG & E, SCE and San Diego Gas & Electric Co. (SDG & E). The ISO operates the electrical grid for the state and purchases power as neces-. sary to ensure non-discriminatory access and system reliability. Although PG & E continues to own its transmission system, the ISO has operational control. At all times relevant to PG & E’s complaint, if PG & E’s customer demand was not met by scheduled supplies into the PX or other sources, the ISO was required to procure additional electricity to serve PG & E’s requirements and maintain the stability of the grid. See Compl (Doc #1) at ¶ 20.

Prior to August 3, 2000, the CPUC required PG & E to procure electricity solely through the PX, unless, as discussed above, PG & E’s customer demand could not be met by the scheduled power supply available on the PX. After August 3, 2000, the CPUC authorized PG & E, and the other California IOUs, to purchase a limited amount of wholesale electricity through bilateral contracts outside the PX and ISO markets, subject to certain restrictions. See Kubitz Decl (Doc # 115) at ¶¶ 4-9. As a consequence of this regulatory change, PG & E has divided its preemption claim into two parts: one concerning the period before August 3 and one concerning the period after.

In order for AB 1890 to be implemented and for the PX and ISO to begin operation, the FERC, which has jurisdiction to regulate the sale of electricity in interstate commerce, was required to approve certain filings by the ISO, the PX and the IOUs. Beginning in late 1996, as part of the shift to a competitive electricity market, the FERC granted a series of requests by owners of generation plants, including the IOUs and recent purchasers of plants previously owned by the IOUs, for authorization to sell electricity on the PX and other wholesale markets at market-based rates. PG & E during restructuring would be both a purchaser and a seller of electric wholesale power.

Prior to 1996, by contrast, the CPUC regulated nearly every aspect of PG & E’s vertically integrated electricity business. PG & E owned and operated assets used in generation, transmission and distribution of electricity to retail customers. The CPUC established and regulated the retail rates that PG & E could charge its customers, setting these rates at a level suffi cient to allow PG & E to recover the costs of generation, transmission, distribution and other ancillary functions, as well as allowing PG & E a reasonable rate of return on its investment in the capital required to perform these functions. In restructuring its electricity markets, California sought to separate the utilities’ vertically-integrated generation, transmission and distribution functions with the goal of providing consumer access to competitively priced generation. On the retail side, this envisioned eventually replacing the regulation of retail rates based on cost and a reasonable rate of return with competitively determined market rates, subject to certain limitations in order to protect certain constituencies.

In enacting AB 1890, the California legislature did not effect an immediate transition to this new regime. Rather, California imposed a rate freeze on retail rates during a transition period. This transition period was set to end the earlier of March 31, 2002, or the date that the IOUs recovered so-called “stranded costs.”

Addressing stranded costs is a central problem in restructuring electricity markets. Stranded costs are historic investments and contractual obligations of the utilities that exceed the value of the underlying assets in a competitive environment. Examples of stranded costs are the expenses of certain generating plants or long-term power supply contracts that cannot be recovered from customers through competitive electrical prices. Historic investments in transmission lines and facilities, which, in a competitive market must be made available to competitors, are also considered stranded costs. Because such costs were incurred during a period when IOUs operated pursuant to a eost-of-ser-viee regulatory scheme, IOUs would be burdened upon the introduction of a competitive retail market with costs not borne by other entries into the market. Bearing these stranded costs placed IOUs at a competitive disadvantage to electricity generators not similarly saddled with such costs. AB 1890 sought to level the playing field.

One way of addressing stranded costs— the route chosen by California — is to allow IOUs a limited opportunity to recover stranded costs before the introduction of competition. It would, of course, be possible to permit such a transition period to extend indefinitely, until stranded costs were completely recovered. This route, however, would delay the introduction of market incentives for an indeterminate period. California’s approach was to introduce both market incentives, in the form of a time constraint, and state mandated retail pricing. Under this regime, PG & E, in theory, had both the opportunity and incentive to maximize its surplus during the transition period.

During the transition period, AB 1890 temporarily froze the utilities’ retail rates at the levels in effect on June 10, 1996, subject to a 10% rate reduction for residential and smaller business customers. The rate reduction and retail price freeze was predicated on an assumption that proved in retrospect to be wildly inaccurate: namely, wholesale prices achieved under FERC’s transition to market-based rates would be sufficiently lower than in recent years as to provide “headroom” (i e, excess of retail over wholesale prices). The purpose of this “headroom” was to allow utilities to recover their stranded costs.

AB 1890 charged the CPUC with implementing the ratemaking elements of the bill. Among its responsibilities, the CPUC was directed to determine the stranded costs eligible to be recovered during the transition period and the methods by which the utilities could recover their stranded costs. CalPubUtilCode §§ 367- 369. Since 1996, the CPUC has issued a series of decisions interpreting AB 1890 and determining the mechanisms for the recovery and accounting of the recovery of stranded costs, as well as the costs of providing electric service during the transition period.

The CPUC established two accounts as the primary mechanisms for tracking costs and revenues during the rate freeze. The first, the transition recovery account (TRA), is used to record operating costs and retail revenues. The second, the transition cost balancing account (TCBA), tracks the recovery of stranded costs. During the rate freeze, retail revenue in excess of cost recorded in the TRA is charged to customers, appearing on customer bills as a competitive transition charge (CTC). When the retail rates during the rate freeze exceeded the cost of providing electricity, as was supposed regularly to occur, California customers paid higher energy bills than they would have before the beginning of the transition period, thereby subsidizing the shift to a competitive market.

The original accounting rules, promulgated by the CPUC in 1997 in Resolution E-3514, provided that both debit balances (liabilities) and credit balances (headroom) would be transferred from the TRA to the TCBA each month during the rate freeze. See Res E-3514, in DRJN (Doc # 106, Exh F). In 1998, the CPUC adopted Resolution E-3527, which changed this rule retroactively and specified that only credit balances in the TRA would be transferred to the TCBA at the end of each month. See Res E-3527, in DRJN (Doc # 106, Exh G). Under this accounting system, PG & E could pay down its stranded costs in months in which revenues exceeded costs. When revenues from frozen rates were insufficient to cover operating costs recorded in the TRA, however, the TRA account accumulated an “undercollection” which was carried over to the following month for recovery and was not set against previous stranded cost recovery. Defendants assert, and PG & E does not dispute, that in the first two years of the transition period, before the prices of wholesale energy in California soared, PG & E was able to transfer billions of dollars of excess revenues from the TRA to the TCBA, paying down stranded costs. See DefBr (Doc # 104) at 8.

Beginning in June of 2000, however, the market-based prices of wholesale electricity in California soared dramatically, setting off a crisis in the California energy market. As wholesale prices soared, PG & E’s ability to pay down the approved stranded costs in the TCBA suffered and instead, PG & E began to accumulate massive deficits in its TRA. Between June 2000 and January 31, 2001, PG & E alleges that its wholesale energy costs exceeded the amount available in frozen retail rates by approximately $8.3 billion. As PG & E began to accumulate crippling debt in order to finance the cost of buying electricity on the expensive wholesale market, PG & E’s credit rating deteriorated. See PlOppBr (Doc # 148) at 7. PG & E defaulted on commercial obligations and ultimately was forced to seek protection from its creditors by filing for bankruptcy on April 6, 2001.

PG & E blames its debt and bankruptcy on the CPUC’s refusal to dissolve the rate freeze and raise retail rates concurrently with the spike in electricity wholesale prices. Defendants, on the other hand, contend that PG & E accepted the risk that prices would exceed frozen rates and expressly relied on the existence of that risk in its appeal to the FERC for market-based wholesale pricing. Defendants also contend that PG & E, in fact, did not suffer any total debt during the period of the rate freeze, despite the spike in whole sale prices, and was, in fact, able to pay-down some substantial portion of its stranded costs. This contention is based, in part, on an adjustment by the CPUC in the accounting rules governing the rate freeze, adopted during the energy crisis.

In that time frame, specifically, on October 17, 2000, TURN, an advocacy group, filed a petition with the CPUC, requesting that the CPUC modify the accounting rules for tracking PG & E’s recovery of stranded costs and, correspondingly, determining the end of the rate freeze. TURN’S proposal, which it refers to as a “true-up,” would require that, beginning January 1, 1998, both negative and positive balances in the TRA would be transferred to the TCBA on a monthly basis, as opposed to only positive balances (headroom) being transferred. This accounting change was to have a couple of effects. First, it would require excess revenue previously applied to pay down stranded costs instead first to be applied to offset under-collection caused by the soaring prices in the wholesale energy market. Second, by increasing the balance to be paid down in the TCBA in months in which the TRA was undereollected, the TURN accounting change would extend the period of the rate freeze.

One effect of the California energy crisis is that it reversed the incentives of some market actors. When AB 1890 was initially approved, PG & E benefited from the transition period, which allowed PG & E an extended opportunity to recover the costs of its historic investments that would otherwise be uneconomic upon an immediate transition to a competitive regime. When the frozen rates became insufficient to cover increasing energy costs, however, PG & E, on the retail side at least, would benefit from an immediate end to the retail rate freeze. As a result, on November 8, 2000, while TURN’S proposal was pending with the CPUC, PG & E filed its initial complaint in this matter, alleging that the CPUC’s actions, including its consideration of the TURN “true-up,” exceeded its authority by fading to allow PG & E to recover its energy procurement costs concurrently in retail rates.

During this same period, PG & E repeatedly petitioned the CPUC for rate increases. In November 2000, PG & E sought the CPUC’s approval of an emergency rate stabilization plan, including a proposal to increase retail rates to recover PG & E’s undercollection. After holding emergency hearings and ordering outside audits of PG & E’s financial condition, the CPUC rendered a decision on January 4, 2001, ordering a $.01/kWh emergency surcharge to help pay for future procurement costs. See D01-01-018 in PRJN II (Doc # 117, Exh # 14).

On March 27, 2001, the CPUC issued D01-03-082, which is the central CPUC decision challenged by PG & E. See D01-03-082 in PRJN II (Doc # 117, Exh # 12). This decision approved a further $.03/kWh surcharge increase for the electric utilities’ retail rates. More significantly for present purposes, however, D01-03-082 also adopted the TURN proposal for the accounting “true-up.” As a result, the CPUC modified its current accounting rules to require that each month the balance in the TRA be transferred to the TCBA, whether positive or negative. See id. at 30. The effective date of the accounting changes was January 1, 1998, the date when Resolution E-3527 took effect. In enacting the decision, the CPUC noted that, in retrospect, the accounting rules of Resolution E-3527 contravened the principles of AB 1890, and that the “true-up is necessary to correct inequities in the current accounting rules which make it appear that the utilities have fully collected their stranded capital costs, while at the same time recording monthly liabilities of billions of dollars in operating costs.” Id. at 28.

The true-up accounting decision is central to the instant dispute as it requires PG & E, in general terms, to set its total losses against its total revenues before applying any surplus revenue to paying down its stranded costs. Besides resulting in the recovery of fewer stranded costs, therefore, the enactment of this decision was likely to extend the period of the rate freeze. As the decision was adopted during the wholesale price spike, PG & E was understandably against its adoption.

Judge Lew dismissed PG & E’s original complaint largely because D01-03-082 was not yet final. After this decision became final, PG & E re-filed its complaint, with minor adjustments. In its complaint, PG & E alleges that state law, as interpreted and applied by defendants, commissioners of the CPUC, caused it severe financial harm. Compl (Doc # 1) at ¶ 2. PG & E seeks injunctive and declaratory relief to prevent and restrain defendants from “continuing to violate federal law by denying PG & E recovery of its wholesale electricity procurement and transmission costs in retail rates.” Id. at ¶ 3. PG & E’s complaint is built on its preemption claims, which place heavy reliance on a somewhat esoteric, although tremendously important, regulatory doctrine: the “filed rate doctrine.” In particular, PG & E argues that because its electricity costs were incurred pursuant to rate schedules filed with the FERC, the CPUC was required, effectively, to end the rate freeze and raise retail rates once wholesale costs exceeded frozen rates. PG & E also alleges that defendants’ actions violate the Takings Clause of the Constitution, the Due Process Clause and the Commerce Clause.

II

Before addressing the substance of the summary judgment motions, the court must first address several preliminary matters.

First, the court must address TURN’S motion to intervene (Doc # 18), filed concurrently with its motion to dismiss. Doc #20. TURN is a nonprofit organization that is, in its words, “devoted to protecting the interests of residential and small-commercial consumers of electricity, natural gas and telephone services.” TURN Inter Br (Doc # 18) at 1-2. TURN has been an active participant in the proceedings forming the basis of PG & E’s lawsuit, as well as all litigation stemming from these proceedings. TURN played an active role in the debate over AB 1890, both before and after its enactment. Id at 2. Although TURN generally opposed the CPUC’s decision to restructure the energy market, as AB 1890 legislation became inevitable, TURN redirected its efforts to push for the inclusion of rate relief for consumers and small business as part of the restructuring. Id at 3.

PG & E filed a statement of non-opposition to TURN’S motion to intervene. This statement, however, amounts only to a non-opposition to TURN’S motion to the extent TURN seeks to intervene permissively pursuant to FRCP 24(b). Doc # 44. In fact, PG & E opposes TURN’S motion to intervene as of right, pursuant to FRCP 24(a), relying largely on the contention that Judge Lew’s decision that TURN may only permissively intervene precludes TURN from intervening as of right in this matter. The court will not, however, apply collateral estoppel to TURN’S motion. When Judge Lew ruled on TURN’S motion to intervene, the accounting changes advocated by TURN had not yet been finally implemented.

FRCP 24(a) establishes four requirements for intervention as of right: timeliness; an interest relating to the subject matter of the action; practical impairment of the party’s ability to protect that interest; and inadequate representation by the parties to the suit. Idaho Farm Bureau Fed’n v. Babbitt, 58 F.3d 1392, 1397 (9th Cir.1995), citing United States v. Oregon, 913 F.2d 576, 587 (9th Cir.1990). TURN moved to intervene less than two months after the filing of PG & E’s complaint in this matter, before discovery began or the court made any substantive rulings. TURN’S motion is timely. See e.g., Idaho Farm, 58 F.3d at 1397.

“A public interest group is entitled as a matter of right to intervene in an action challenging the legality of a measure it has supported.” Id., citing Sagebrush Rebellion, Inc. v. Watt, 713 F.2d 525, 527 (9th Cir.1983); Washington State Bldg. & Constr. Trades Council v. Spellman, 684 F.2d 627, 630 (9th Cir.1982). Beyond supporting the measures in dispute, TURN was the acknowledged author and leading proponent of the true-up proposal, adopted in D01-03-082, which is one of the central actions by the CPUC challenged by PG & E. TURN has an interest relating to the subject of the present litigation.

The present action could substantially affect TURN’S interest as the disposition of the present action could substantially affect the electrical rates charged to consumers and small business owners. Finally, the court determines that TURN is not adequately represented by defendants. The burden of making this showing is minimal. See e.g., Sagebrush Rebellion, 713 F.2d at 528, citing Trbovich v. United Mine Workers, 404 U.S. 528, 538 n. 10, 92 S.Ct. 630, 30 L.Ed.2d 686 (1972). As evidenced by TURN’S adaptation of positions relative to the actions taken by the CPUC, TURN and the CPUC do not have coextensive interests and serve different, if overlapping, constituencies.

As a result, the court grants TURN’S motion to intervene as of right, pursuant to FRCP 24(a). In the alternative, the court grants TURN permission to intervene permissively, pursuant to FRCP 24(b).

The parties have requested that the court take judicial notice of an assortment of documents. See Does##106 and 116. With respect to documents not referenced in PG & E’s complaint, the court may take judicial notice of adjudicative facts, which are those to which the law is applied in the process of adjudication. See Adv Notes to FRE 201. A judicially noticed fact may not be subject to reasonable dispute and must be relevant. See FRE 201(b). “A judicially noticed fact must be one not subject to reasonable dispute in that it is either (1) generally known within the territorial jurisdiction of the trial court or (2) capable of accurate and ready determination by resort to sources whose accuracy cannot reasonably be questioned.” Id. The court may take judicial notice of pleadings, orders and statutes from other jurisdictions, including agency decisions, if the documents are public records and subject to confirmation by sources that cannot reasonably be questioned. See e.g., United States ex rel Robinson Rancheria v. Borneo, Inc., 971 F.2d 244, 248 (9th Cir.1992).

The only challenge to documents submitted for judicial notice is brought by defendants, who argue that the court should decline to take judicial notice of the stipulated judgment and settlement agreement, see PRJN II (Doc # 117, Exh # 39), between defendants and SCE in a related action in the Central District of California. Doc # 142. Defendants argue that settlement agreements and related documents are inadmissable under FRE 408, which provides:

Evidence of (1) furnishing or offering or promising to furnish, or (2) accepting or offering or promising to accept, a valuable consideration in compromising or attempting to compromise a claim which was disputed as to either validity or amount, is not admissible to prove liability for or invalidity of the claim or its amount. Evidence of conduct or statements made in compromise negotiations is likewise not admissible.

One of the principles underlying FRE 408 is that evidence of a settlement is generally not relevant, because settlements may be motivated by a variety of factors unrelated to liability. See Hudspeth v. CIR, 914 F.2d 1207, 1213-14 (9th Cir.1990). The court agrees with defendants that documents relating to the settlement between defendants and SCE are not relevant to the instant dispute and, therefore, the court declines to take judicial notice of exhibit # 39 to PG & E’s second set of request for judicial notice. All other documents submitted for judicial notice, however, meet the requirements of FRE 201(b) as they are not subject to reasonable dispute and are in the official public records of the CPUC, the California legislature, federal agencies or federal courts and the court will take judicial notice of them.

C

PG & E advances a series of evidentiary objections. First, PG & E objects to the declaration of Douglas Long (Doc # 105), submitted in support of defendants’ motion for summary judgment. Doc # 151. PG & E also objects to the declaration of Matthew Freedman (Doc # 121), submitted in support of TURN’S motion. Doc # 151. Long is the program manager of the CPUC’s energy division. Long Decl (Doc # 105) at ¶ 1. Long’s declaration discusses the enactment of AB 1890, various CPUC decisions, the accounting true-up and other elements of the rate freeze generally. PG & E objects to Long’s declaration on the basis of lack of personal knowledge and lack of foundation. This objection (Doc # 151) is DENIED. Long sets forth the foundation of his statements. Long was personally involved in many aspects of CPUC activities for a number of years and, pursuant to the broad conception of “expert” embodied in FRE 702, Long appears qualified to testify as witness knowledgeable about the regulatory and accounting aspects of electrical generation, transmission and marketing. See Thomas v. Newton International Enterprises, 42 F.3d 1266, 1269 (9th Cir.1994).

Long’s declaration is, however, argumentative and not very helpful on the issues the court must decide. For example, Long argues that PG & E supported enactment of AB 1890. The import of any such support on the issues at bar is not apparent; CPUC evidently also supported this legislation. That the scheme enacted by AB 1890 turned out badly may well bear on the issues to be decided, but why PG & E’s (or the CPUC’s) support for the legislation matters is not readily discerned and Long’s declaration does not enlighten the reader. Furthermore, Long’s declaration advances legal conclusions as assertions of fact. Most notable in this regard is Long’s contention that “AB 1890 only afforded the utilities an opportunity to recover their stranded costs, not a guarantee ***,” (Doc # 105 at ¶ 15), which Long seems to suggest should be the conclusion dx-awn by this court on the legal issues hei'e. These reservations aside, the court will overrule PG & E’s evidentiary objections as it has considered the Long declaration for what evidentiary value it may have.

PG & E similarly seeks to" strike Freedman’s declaration (Doc # 121). Freedman summarizes PG & E’s regulatory filings in support of TURN’S contention that PG & E, in fact, recovered its wholesale procurement costs over the period of the rate freeze. PG & E contends that because Freedman did not make these regulatory filings himself, he is not competent to testify about them on the basis of his personal knowledge. Freedman’s declaration is highly conclusory and seeks to have the court accept a number of matters that appear to the court to be in dispute. Although Freedman’s declaration is not helpful for these reasons and PG & E’s objection appears to result from its concern with Freedman’s conclusions about the evidence, not a proper ground for objection, PG & E’s objection to Freedman’s declaration will be DENIED, notwithstanding the court’s determination that this declaration contributes little, if anything, of value.

PG & E also moves to strike four declarations submitted by defendants and TURN in opposition to PG & E’s motion. Doc # 167. PG & E moves to strike the declaration of Peter Bradford (Doc # 155). Although Bradford’s declaration does contain some facts and some opinions about utility restructuring generally that could qualify as opinions, Bradford’s declaration simply expands TURN’S legal argument that the filed rate doctrine does not, as a legal matter, apply to California’s rate freeze. Because this declaration primarily contains legal argument rather than evidentiary matter, PG & E’s motion to strike Bradford’s declaration (Doc # 167) is GRANTED.

Similarly, the declaration of Michael Florio (Doc # 156), a staff attorney for TURN, contains little more than an elaboration of TURN’S legal arguments about the proper time period over which to consider whether PG & E has recovered its costs and the regulatory bargain to which PG & E allegedly agreed. If such “expert” testimony were permitted, the page requirements for briefs filed with the court would become, effectively, moot. PG & E’s motion to strike Florio’s declaration (Doc # 167) is also GRANTED.

PG & E also objects to the declaration of James Loewen. Loewen’s declaration discusses the procurement options available to PG & E in the block forward market. See Loewen Decl (Doc # 138). Again, PG & E’s objections to Loewen’s declaration appear driven by a disagreement with his conclusions and while Loewen’s declaration is unhelpful, it does contribute some factual information to place the issues in context. Accordingly, PG & E’s motion to strike Loewen’s declaration (Doc # 167) is DENIED.

PG & E also contends that Long’s declaration in support of defendants’ opposition (Doc # 137) is vague and conclusory. Again, this is little more than an attack on Long’s statements. The motion to strike Long’s declaration (Doc # 167) is DENIED.

In the main, the declarations and objections are a distracting side show to the central matters at bar. The court hopes that as this litigation proceeds the parties will avoid submissions of lengthy and tendentious declarations that are little more than legal arguments masquerading as factual assertions.

Defendants seek to file the declaration of David E Effross under seal, as it contains information designated as confidential by PG & E. Doc # 140. For good cause shown, this motion (Doc # 140) is GRANTED. The clerk is directed to file the lodged document under seal.

Finally, the parties’ stipulated protective order (Doc # 103) is hereby ENTERED.

III

PG & E moves for summary judgment on its first and second claims for relief. In these claims, PG & E asserts that state law, as interpreted and applied by the CPUC is preempted to the extent it prohibits PG & E from recovering in retail rates expenses incurred in procuring and providing electricity. See Compl (Doc # 1) at ¶¶ 60-78. Claim one asserts preemption prior to August 3, 2000, and claim two asserts preemption from August 3, 2000, through January 19, 2001. The significance of the August 3 date is that on August 3, 2000, the CPUC granted limited authorization for PG & E and the other utilities to enter into contracts for the purchase .of wholesale energy outside the PX.

Defendants move for summary judgment on PG & E’s complaint or, in the alternative, on PG & E’s preemption claims. Doc # 104. Attempting fully to cover their bases, defendants have also filed a FRCP 56(f) request to continue PG & E’s motion for judgment on its preemption claims. Doc # 143. In this motion, defendants contend that if the court is not prepared to deny PG & E’s motion, the court should permit defendants to conduct more discovery on PG & E’s ability to procure electricity through cheaper sources, under the so-called Pike County exception to the filed rate doctrine. See Pike County Light & Power Co. v. Pennsylvania Pub. Util. Comm’n, 77 Pa.Cmwlth. 268, 465 A.2d 735 (1983).

TURN moves for summary judgment on PG & E’s complaint, although its motion is, like that of the other parties, overwhelmingly directed at PG & E’s preemption claims. Doc # 119. The state of California has also filed an amicus brief in support of defendants’ motion for summary judgment. Doc # 110.

The court first notes that in their pending motion to dismiss (Docs ## 24 and 77), defendants argued that PG & E’s complaint was barred by the doctrine of state sovereign immunity. “Though not jurisdictional in the traditional sense,” sovereign immunity defenses usually represent a threshold issue, to be reviewed before considering other non-jurisdictional defenses. Agua Caliente Band of Cahuilla Indians v. Hardin, 223 F.3d 1041, 1045 (9th Cir.2000). In this instance, however, the court finds it appropriate to address PG & E’s preemption claims before considering in detail defendants’ state sovereign immunity defense, because the court’s determination of the validity of PG & E’s preemption claims will greatly inform the court’s determination of the validity of defendants’ assertion of state sovereign immunity.

Pursuant to the Supreme Court’s analysis of state sovereign immunity in Idaho v. Coeur d’Alene Tribe, 521 U.S. 261, 117 S.Ct. 2028, 138 L.Ed.2d 488 (1997), as interpreted by the Ninth Circuit, in considering whether the Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908), exception to state sovereign immunity applies, the court must examine, among other things, the intrusiveness of the relief requested by PG & E. Moreover, in general, a valid claim that state officials are engaging in ongoing behavior in violation of or preempted by federal law will fall within the Ex parte Young exception to state sovereign immunity. See, e.g., Green v. Mansour, 474 U.S. 64, 68, 106 S.Ct. 423, 88 L.Ed.2d 371 (1985) (“the Eleventh Amendment does not prevent federal courts from granting prospective injunctive relief [against state officials] to prevent a continuing violation of federal law.”). See also Duke Energy Trading & Mktg. LLC v. Davis, 267 F.3d 1042, 1052- 1055 (9th Cir.2001). It would not make sense, however, to permit a plaintiff to bring a suit against state officials in federal court that would otherwise be barred, merely by pleading a meritless preemption claim. In this case, therefore, a determination of the merits of defendants’ state sovereign immunity defense requires a pri- or consideration of the substance and merit of PG & E’s preemption claims.

Similarly, in their motion to dismiss, defendants contend that the Johnson Act, 28 USC § 1342, bars the award of PG & E’s requested relief. The Johnson Act provides:

The district courts shall not enjoin, suspend or restrain the operation of, or compliance with, any order affecting rates chargeable by a public utility and made by a State administrative agency or a ratemaking body of a State political subdivision, where:

(1) Jurisdiction is based solely on diversity of citizenship or repugnance of the order to the Federal Constitution; and

(2) The order does not interfere with interstate commerce; and

(3) The order has been made after reasonable notice and hearing; and

(4) A plan, speedy and efficient remedy may be had in the courts of such State.

A preemption claim based on the filed rate doctrine is not based solely on repugnance to the Supremacy Clause of the Constitution, but also to a statutory provision, the Federal Power Act, as well as a federal agency determination. If such a claim is valid, therefore, the Johnson Act poses no impediment to the award of an injunction. See, e.g., Public Serv. Co. v. Patch, (Patch V), 167 F.3d 29, 33 (1st Cir.1998). If PG & E’s preemption claims are unmeritorious, however, the Johnson Act may well pose a formidable hurdle to PG & E’s success on its remaining causes of action.

Defendants’ abstention arguments, too, will become more or less persuasive depending on the court’s determination of the viability of PG & E’s preemption claims. See Public Serv. Co. v. Patch, (Patch VII), 221 F.3d 198, 203 (2000). For these additional reasons, therefore, a thorough consideration of PG & E’s preemption claims must take precedence to the resolution of PG & E’s other claims, as well as to the court’s estimation of many of the strongest defenses advanced by defendants and TURN. PG & E’s «preemption claims are, in fact, potentially dispositive; if they fail, PG & E’s complaint will face formidable obstacles. Defendants’ sovereign immunity arguments become forceful absent a valid filed rate claim. The Johnson Act erects a formidable hurdle and abstention arguments gain force under those circumstances. But if preemption claims have merit, these defenses largely fade away.

PG & E’s preemption claims require the court to consider application of the “filed rate doctrine,” a regulatory doctrine with a lengthy historical pedigree, to a contemporary regulatory context, in which the continuing application of this doctrine is less than obvious. The regulatory context in which the actions giving rise to this lawsuit occurred involved substantial cooperation between federal and state regulators, each operating largely within their respective sphere of regulatory jurisdiction, in an attempt to restructure the market for electricity and make this market, among other things, less reliant on traditional public utility rate-making.

As discussed above, one of the central goals of this restructuring was to expose wholesalers of electricity and providers of electricity to the competitive forces of the market. As amply evinced in California’s attempt to achieve this exposure, one great virtue of markets is their ability to defy conventional norms. The increased efficiency and reduced prices associated with competition comes as a result of exposure to risk. Whether the exposure to the risk of energy price fluctuations is desirable in the context of electricity is a matter for debate among policymakers and whether federal law forbids state regulators from requiring a utility to sell electricity at rates below wholesale costs sanctioned under federal law is the central legal question before the court.

The filed rate doctrine, which governs the relationship between service providers and end-users in regulated industries, dates back to the nineteenth century. The doctrine developed in the context of the regulation of interstate railroads, in the years after the Civil War, when a developing interstate railroad system dominated interstate commerce. 2 This dominance, in turn, resulted in widespread discontent with railroad carriers, who were accused of using market power to charge discriminatory rates and to wield immoderate economic influence in transportation and other markets. This discontent produced a rich and colorful literature. See, e.g., Frank Norris, The Octopus (Penguin Books, 1901); Ida M Tarbell, The History of the Standard Oil Company (McClure, 1904).

Eventually this discontent gave birth to legislation and regulation. A railroad, of course, enjoys certain locational advantages and, to the extent alternative routes for the carriage of goods (and, in an earlier era, people, as well) are unavailable, certain features of a monopoly. A persistent construct underlying legislation and early regulation of railroads was the notion that prices should reflect the cost of producing the services subject to such regulation. From this construct flows the notion of “price discrimination.” If alternative routes were unavailable, a monopoly carrier “is likely to fix different prices to different purchasers depending not on the costs of selling to them, which are the same, but on the elasticity of their demands for [service]. This is price discrimination.” R Posner, Economic Analysis of Law, 4th Edl992, § 9.4 at 281. Focusing on costs, it seemed “discriminatory” to charge a shipper more for moving goods from Elk-hart to Chicago than from New York to Chicago unless, of course, the carrier could establish cost savings associated with greater traffic densities for the latter services.

Associated with this were other arrangements thought to represent market dysfunction, including “a wide variety of kickbacks, gratuities, and other devices that agitated much of the public.” Kearney & Merrill, 98 ColumLRev at 1883 (cited in note 2). Even “[m]ore importantly, some railroads simply entered the business of buying and selling products. In such cases, railroad carriers could grant themselves discounts on shipments and obtain either a competitive advantage over suppliers of the same goods or an outright monopoly in the market for those products.” Johnson, 68 AmBankrLR at 321 (cited in note 2).

In response to the appearance of monopoly abuses and widespread price discrimination, Congress passed the Interstate Commerce Act of 1887(ICA), the first major federal regulatory statute in the United States and the model for future regulation of common carriers and, in due course, public utilities. The purpose of the ICA, as articulated by the Supreme Court not long after its enactment, was to:

secure equality of rates as to all, and to destroy favoritism, these last being ac- , complished by requiring the publication of tariffs, and by prohibiting secret departures from such tariffs, and forbidding rebates, preferences, and all other forms of undue discrimination.

NYNH v. Hart, 200 U.S. 361, 391, 26 S.Ct. 272, 50 L.Ed. 515 (1906).

The ICA placed the railroad industry under the regulatory authority of the Interstate Commerce Commission (ICC), the nation’s first major regulatory agency. The first formulation of what became known as the “filed rate doctrine” was found in § 6(7) of the ICA:

[W]hen any such common carrier shall have established and published its rates, fares, and charges in compliance with the provisions of this section, it shall be unlawful for such common carrier to charge, demand, collect, or receive from any person or persons a greater or less compensation for the transportation of passengers or property, or for any services in connection therewith, than is specified in such published schedule of rates, fares, and charges as may at the time be in force.

ICA, ch 104, § 6(7), 24 Stat 379, 381 (1887).

This requirement — that regulated providers charge end-users only the rate on file with the appropriate regulatory agency — is, at its essence, the filed rate doctrine, from which all other variations and applications flow.

As noted, the ICA and the filed rate doctrine expressed therein became the prototype for subsequent regulation by Congress of a range of other industries. By 1938, Congress had applied this model to the interstate components of the shipping, stockyard, telephone, telegraph, trucking, electric, gas and aviation industries, reflecting a consensus that heavy government oversight of such industries and published, managed pricing was the appropriate means to achieve the goals of reasonable prices, non-discrimination and reliability. See Kearney & Merrill, 98 Co-lumLRev at 1334 (cited in note 2).

With experience, these regulatory regimes produced numerous regulatory failures rivaling or exceeding dysfunctions in markets. One by one in each of these industries the consensus that favored heavy governmental oversight has broken down. The role of private actors, the market and government regulators have all been re-evaluated. The railroad, airline, motor carrier and telecommunications industry, among others, have witnessed fitful moves to one form of deregulation or the other beginning in the 1970s with the airline industry. The instant dispute, similarly, results from the joint determination of state and federal regulators to move toward competition, in this case in the generation, transmission and sale of electricity.

Historically, the filed rate doctrine’s primary application has been to prevent price discrimination. In the case of common carrier regulation, such discrimination was perceived if large suppliers and shippers negotiated discounted fares or tariffs. This, of course, was seen as a classic route to monopoly. See, e.g., Standard Oil Co. v. United States, 221 U.S. 1, 32-33, 31 S.Ct. 502, 55 L.Ed. 619 (1911) (“[T]he bill alleged that the combination and its members obtained large preferential rates and rebates in many and devious ways over their competitors from various railroads companies, and that by means of the advantage thus obtained many, if not virtually all competitors were forced either to become members of the combination or were driven out of business ***.”). The requirement that regulated carriers charge and receive only the filed rate has been consistently and rigidly applied, with, at times, somewhat counter-intuitive results. In Louisville & Nashville Railroad Co v. Maxwell, 237 U.S. 94, 35 S.Ct. 494, 59 L.Ed. 853 (1915), for example, a passenger negotiated a ticket for a rate of $49.50, although the published rate for the route was $78.65. After discovering this error, the railroad sued the passenger for the difference and the Supreme Court upheld an award against the passenger, declining to countenance any deviation from the filed rate.

Early regulation of public utilities at the federal and state level was motivated more by the goal of preventing undue price discrimination than by the fear of high prices. 1 AJG Priest, Principles of Public Utility Regulation, 285, 289 (Michie, 1969) (“Prevention of discrimination has been a vital regulatory function since federal and state statutes which deal with ‘natural monopolies’ first acquired teeth. *** Even exorbitant rates probably have generated less irritation and exasperation than discriminatory practices. *** Discrimination raises its grizzly head both when utility service is sold to a large consumer at less than cost and when any charge or practice imposes unreasonable burdens on other customers.”).

Especially in the case of electricity regulation, the filed rate doctrine came to have a subsidiary application demarking the boundary between federal and state regulatory authority. Early on, electricity generation, transmission and sale were almost exclusively confined to one locality and, in this regard, electricity generation, transmission and sale differed from, say, rail transportation which by the time it came under regulatory purview was largely interstate in character. The ability, developed on a large scale in the past few decades, to transmit electricity beyond state boundaries at low cost demanded federal as well as state action if electricity were to maintain under effective regulatory control. Federal policy fostered the availability of electricity generated at widely dispersed locations and from many sources. See Public Utility Regulatory Policies Act, PubL 95-617 § 2, Nov 9, 1978, 92 Stat 3117, codified at 16 USC § 2601 et seq and Powerplant and Industrial Fuel Use Act, PubL 95-620, Title I, § 102, Nov 9, 1978, 92 Stat 3291, PubL 100-42 § 1(c)(1), May 21, 1987, 101 Stat 310, codified at 42 USC § 8301 et seq. The filed rate doctrine came to assist in setting the boundaries of the respective spheres of regulation over the increasingly multifaceted generation, transmission, distribution and sale of electricity.

The filed rate doctrine operates in the electricity context to ensure that rates charged for wholesale electricity are on file with and approved by federal regulators; but, perhaps more importantly, the doctrine also operates to prevent state regulators, as well as courts, from taking action that fails in any manner to account for the fact that in most instances today a utility must purchase the power delivered to consumers pursuant to the rate filed with the appropriate federal agency.

Regulatory jurisdiction over electricity is divided by federal law into two rather arbitrary spheres of authority: states have regulatory authority over retail sales of electricity and the federal government has authority over interstate, i.e. wholesale, sales. This scheme for dual authority is codified in the Federal Power Act (FPA), 16 USC §§ 824-824m. As the Supreme Court has observed, in an oft quoted phrase, in the FPA, “Congress meant to draw a bright line easily ascertained, between state and federal jurisdiction.” Nantahala Power & Light Co. v. Thornburg, 476 U.S. 953, 966, 106 S.Ct. 2349, 90 L.Ed.2d 943 (1986), quoting FPC v. Southern Cal. Edison Co., 376 U.S. 205, 215-16, 84 S.Ct. 644, 11 L.Ed.2d 638 (1964). “This was done in the Power Act by making [FERC] jurisdiction plenary and extending it to all wholesale sales in interstate commerce except those which Congress has made explicitly subject to regulation by the States.” Id.

Under this scheme, individual states are empowered to regulate retail sales as well as local distribution services involving electric power, but may not intrude on the federal government’s plenary power to regulate interstate transmission and wholesale sales of electricity in interstate commerce. The federal government exercises its jurisdiction by delegating authority to the FERC, which has exclusive jurisdiction over all facilities for interstate transmission and sale of electric energy. Pursuant to this grant of jurisdiction, FERC has the authority to regulate the rates, terms and conditions of interstate transmission, transportation and wholesale sales by nongovernmental entities. Under the FPA, all rates for the transmission and sale of wholesale electricity must be filed with FERC and published for public review. 16 USC § 824d(c). FERC has the obligation to ensure that all such rates are just and reasonable, § 824d(a), and are applied in a non-discriminatory manner.

The filed rate doctrine, in the electricity context, results from FERC’s responsibility for setting and ensuring compliance with just and reasonable rates of wholesale electricity sale and transmission. At a high level of generality, the filed rate doctrine prevents actors from failing to adhere to the wholesale electricity rates filed with FERC. This prohibition applies to both private and governmental parties. Wholesale electricity suppliers are prohibited from charging anything other than the filed rate; utilities purchasing wholesale energy for re-sale are prohibited from negotiating a different price or entitlement to allocation. But, as will be further discussed below, courts, both state and federal, are prohibited from considering any rate other than that filed with FERC to be the appropriate wholesale rate.' And more significantly for present purposes, state regulators, when setting retail rates, are prohibited by the filed rate doctrine from considering any rate other than that filed with FERC as a reasonably incurred wholesale cost.

The filed rate doctrine, in conjunction with its corollary, the rule against retroactive ratemaking, also imposes restrictions on FERC’s actions. Because a utility may charge only the rates that are on file, even if FERC subsequently determines that those rates were unreasonable, FERC generally may not order retroactive refunds. See 96 FERC ¶ 61,120 at 61,505 (2001).

In all these applications, however, the filed rate doctrine is basically the same in the electricity context as in every other: the filed rate, and no other, must be charged and this charge must be considered valid.

Because the filed rate doctrine is a product of federal regulatory jurisdiction staked out in the FPA, judicial enforcement of the filed rate doctrine is “a matter of enforcing the Supremacy Clause.” Nantahala Poioer & Light, 476 U.S. at 963,106 S.Ct. 2349. The application of the doctrine by the Supreme Court in several seminal cases further clarifies the typical reach of the doctrine when applied by federal courts.

“As developed for the purposes of the Federal Power Act, the ‘filed rate’ doctrine has its genesis in Montana-Dakota Utilities Co. v. Northwestern Public Service Co., 341 U.S. 246, 251-52, 71 S.Ct. 692, 95 L.Ed. 912 (1951).” Nantahala Power & Light, 476 U.S. at 962, 106 S.Ct. 2349. In Montana-Dakota, two power companies, with interlocking directorate and joint corporate officers, each purchased power from the other at rates the Federal Power Commission (FPC), FERC’s predecessor, determined were reasonable. After the companies’ management separated, one company sued the other in federal court, asserting that it had been paid unreasonably low rates for the electricity that it provided. In dismissing the claim, the Supreme Court held that Congress had given the FPC the exclusive right to determine reasonable rates:

We hold that the right to a reasonable rate is the right to the rate which the Commission files or fixes, and that, except for review of the Commission’s orders, the court can assume no right to a different one on the ground that, in its opinion, it is the only or the more reasonable one.

Montana-Dakota, 341 U.S. at 251-52, 71 S.Ct. 692.

Since Montana-Dakota, the Court has applied the filed rate doctrine to preclude not only federal court review of FERC reasonableness determinations but also to preclude state court review. See e.g., Arkansas Louisiana Gas Co. v. Hall, 453 U.S. 571, 101 S.Ct. 2925, 69 L.Ed.2d 856 (1981). Subsequently, in Nantahala Power & Light, the Court held that just as courts could not fail to give binding effect to a FERC determination of reasonable rates, so too were state agencies prohibited from concluding that FERC-approved wholesale rates were unreasonable.

In Nantahala Power & Light, the Court considered the preemptive effect of a FERC order that reallocated the respective shares of two affiliated companies’ entitlement to low cost power. Under an agreement between two affiliated companies, Nantahala, a public utility, was allocated 20% of the low cost power purchased from the Tennessee Valley Authority (TVA), while the other company, Tapoco, received 80% of the low cost power. Determining that this agreement was unfair to Nantahala, FERC ordered Nantahala to file a new wholesale rate schedule based on an entitlement to 22.5% of the low cost power purchased from the TVA. In a subsequent retail rate proceeding, however, the Utilities Commission of North Carolina (NCUC) reexamined FERC’s determination and ordered Nantahala to calculate its costs for retail ratemaking purposes as though it had received 24.5% of the low cost power. As a result, Nantahala was forced to incur costs of procuring wholesale power that it could not recover through its retail rates. The Court determined that the NCUC’s allocation was preempted by federal law, holding that:

Once FERC sets [a wholesale] rate, a State may not conclude in setting retail rates that the FERC-approved wholesale rates are unreasonable. A State must rather give effect to Congress’ desire to give FERC plenary authority over interstate wholesale rates, and to ensure that the States do not interfere with this authority.

Id. at 966,106 S.Ct. 2349.

In broad terms, the Court also noted that the filed rate doctrine was violated when a purchaser of wholesale electricity is required to sell that electricity at less than the purchase price:

The filed rate doctrine ensures that sellers of wholesale power governed by FERC can recover the costs incurred by their payment of just and reasonable FERC-set rates. When FERC sets a rate between a seller of power and a wholesaler-as-buyer, a State may not exercise its undoubted jurisdiction over retail sales to prevent the wholesaler-as-seller from recovering the costs of paying the FERC-approved rate. *** Such a “trapping” of costs is prohibited.

Id. at 970,106 S.Ct. 2349.

In Mississippi Power & Light Co. v. Mississippi (MP & L), 487 U.S. 354, 108 S.Ct. 2428, 101 L.Ed.2d 322 (1988), the Court reaffirmed that the filed rate doctrine also applies to FERC determinations of the proper allocation of wholesale power. FERC required the Mississippi Power & Light Company (MP & L) to purchase 33% of the plant’s output at rates determined by FERC to be just and reasonable. The Mississippi Public Service Commission (MPSC) subsequently granted MP & L an increase in its retail rates to cover its required purchases. On appeal, however the Mississippi Supreme Court held that it was error for the MPSC to grant such an increase in rates without first conducting a prudence inquiry into the reasonableness of the rates. In reversing, the Supreme Court held that such a prudence inquiry was preempted by the FERC’s determination that the rates set were reasonable.

Although Montanar-Dakota Utilities, Nantahala Pmver & Light and MP & L well illustrate the basic contours of the filed rate doctrine as applied by federal courts in the electricity context, the principles of these cases do not seamlessly translate to the present dispute. The classic model for regulating the relationship between providers and end-users of electricity, which provided the backdrop for all these cases, involves (1) an initial FERC determination of a reasonable price for and/or allocation of entitlement to wholesale energy and (2) a subsequent state determination of the permissible retail rate charged to end-users, based on the cost incurred pursuant to the FERC-filed rate, ancillary costs and a reasonable rate of return. This model is often referred to as a “cost-of-service” regulatory regime. In a cost-of-service regulatory regime, the typical application of the filed rate doctrine is relatively straightforward: when setting retail rates, state regulators may not determine that costs incurred pursuant to a FERC-filed rate were unreasonable and set retail rates as if the full cost of wholesale energy had not been incurred.

As noted, however, the cost-of-service model has fallen from favor. At the time of restructuring, electricity rates in California were approximately twice the national average. See 93 FERC ¶ 61,121 at 61,351 (2000). AB 1890 was intended as a step towards a “market-based” regulatory regime, in which neither wholesale nor retail rates would be, after the transition period, ordained by regulators, but determined by market forces. This new regime, it was thought, would bring California’s electricity rates more in line with average rates.

Defendants and TURN stress that the typical process of regulatory action in classic filed rate cases was, in California, reversed. Specifically, in this case, state regulators instituted a rate freeze and created two new entities, the ISO and PX, upon which electricity sales were to occur and only afterwards did FERC authorize the sale of wholesale energy in these new entities at market-based rates. PG & E, defendants and TURN assert, is, in essence, challenging the operation of a rate freeze that occurred before the federal action which supposedly preempts it. In contrast, under the cost-of-service model, the preemptive action (a FERC-filed rate) occurs before the preempted action (a state’s setting of an insufficient retail rate).

Defendants and TURN further contend that FERC’s decision to allow the sales of wholesale electricity at market-based rates was made, in substantial part, in reliance upon the existence of the state mandated rate freeze. This point is, in particular, stressed by TURN.

FERC began to approve the sale of wholesale energy at market-based rates in the early 1990s in- an effort to develop competitive bulk power markets and to encourage the growth of the electricity generation sector. See Promoting Wholesale Competition Through Open Access Non-diseriminatory Transmission Services by Public Utilities: Recovery of Stranded Costs by Public Utilities and Transmitting Utilities, Order No 888, 61 FedReg 21,540 at 21,545 (1996). Previously, the sale of wholesale energy also operated under a cost-of-service regime, which entailed procedural delays in requesting and achieving approvals for changes in tariffs. One advantage of market-based rates at the wholesale level is that market players may more readily take advantage of fluctuating market opportunities. While the cost-of-service model may have fit vertically integrated utilities, the development of entities that either only generated electricity, to be sold on the market, or did not generate any electricity, but simply wholesaled it, made this model somewhat problematic.

Before approving market-based wholesale rates FERC requires' that the seller of electricity and its affiliates either lack market power or take adequate measures to mitigate any market power they possess. In particular, FERC is concerned with the ability of sellers to limit competition and thereby drive up prices. Of special concern are sellers who own the transmission lines of interstate commerce. In Order No 888, FERC ordered that all public utilities who own transmission lines in interstate commerce have on file open access non-diseriminatory transmission tariffs.

In approving the sale of electricity on the ISO and PX, FERC faced a request by the IOUs, who would be both buyers and sellers of electricity in the restructured market, as well as other electricity generators, to be permitted to sell electricity based on market prices. FERC, concerned that the IOUs had a troublesome amount of market power, instructed the IOUs to file market mitigation proposals. FERC was concerned both that disproportionate market power would allow the IOUs to inflate prices on the PX and/or that the IOUs could engage in predatory pricing, artificially depressing prices to force out other market entrants. FERC, in other words, was concerned about the same type of price discrimination that had animated much of railroad regulation in an earlier era.

PG & E’s market mitigation proposal, as that of the other IOUs, relied upon the existence of the rate freeze and the limited transition period during which to recover stranded costs as measures sufficient to allay FERC’s concerns. First, PG & E argued that as it would most often be a net-buyer of energy, it would'not have an incentive to inflate market prices because, due to the rate freeze, it could not pass on higher costs to the consumer:

In this situation, any increase in revenues that PG & E would obtain as a seller of energy into the PX will be more than offset by the increased costs it bears as a customer purchasing all its energy requirements from the PX. Increasing prices during such periods could benefit PG & E only if PG & E could pass such cost increases through to its retail customers. This is not the case, however, because of the rate freeze and the fixed CTC recovery period.

PG & E Phase II filing, in DRJN (Doc # 106, Exh E) at 8-9, emphasis in original.

PG & E also asserted that it would not have the incentive to engage in predatory pricing as a net buyer, because although this would allow for the faster accumulation of headroom, PG & E would not be able to recover operating losses incurred by bidding below cost. When it was a net seller, PG & E noted, the only risk of market abuse would be inflating prices, but this risk, PG & E stated, was offset by the abundance of low-cost electricity. Moreover, PG & E noted that it was unlikely to be a net-seller of electricity.

Ultimately, FERC approved the sale of electricity at market based rates on the PX, in part in reliance upon the mitigating impact of “the existence of the retail rate freeze under the Restructuring Legislation [AB 1890] during the transition period and the mandatory sale of energy by the Companies into the PX.” 81 FERC ¶ 61,122 at 61,546, in PRJN II (Doc # 117, Exh # 10).

In light of this sequence of events, TURN contends that the court should apply judicial estoppel to PG & E’s claims, to preclude “PG & E from gaining an advantage by taking a position in one tribunal and then seeking a second advantage by taking an incompatible position in a second tribunal.” TURN Br (Doc # 119) at 7, citing Helfand v. Gerson, 105 F.3d 530, 534 (9th Cir.1997). TURN asserts that “PG & E may not now be heard to repudiate its former position that gained it access to what proved to be one of the most lucrative unregulated markets in history.” Id. Ultimately, however, the court is not convinced that PG & E’s identification of the incentives created by the rate freeze and its subsequent attempt to be relieved of those risks present an “inconsistency of factual or legal position,” thereby requiring the court to apply judicial estoppel. Yanez v. United States, 989 F.2d 323, 326 (9th Cir.1993). While the inconstancy of PG & E’s positions underscores the novelty of the claim it makes in this tribunal, defendants demonstrate flexibility no less striking.

To the court’s knowledge, no court has yet been asked to apply the filed rate doctrine to circumstances such as those at bar. The closest parallels are found in the First Circuit which addressed the filed rate doctrine in the context of the restructuring of a state’s electricity market. See Public Serv. Co. v. Patch, 962 F.Supp. 222 (D.N.H.1997) (Patch I); 173 F.R.D. 17 (D.N.H.1997) (Patch II); 136 F.3d 197 (1st Cir.1998) (Patch III); 167 F.3d 15 (1st Cir.1998); (Patch IV); 167 F.3d 29 (1st Cir.1998) (Patch V); 202 F.3d 29 (1st Cir.2000) (Patch VI); 87 FSupp2d 57 (D.N.H.2000) (Patch VII); 221 F.3d 198 (1st Cir.2000) (Patch VIII). But the central filed rate claim in these cases involved a relatively more straightforward application of classic filed rate doctrine. In these cases, Connecticut Valley, an electric utility providing service to end-users in New Hampshire, was denied a rate increase by the New Hampshire Public Utilities Commission (NHPUC), after the cost of procuring electricity pursuant to a long-term wholesale contract filed with FERC increased. See Patch V, 167 F.3d at 32. The NHPUC disallowed the rate increase because it determined that Connecticut Valley should have exercised its option to terminate the FERC rate schedule. See id. át 58-60.

Plaintiff argued that this was a failure to give binding effect to a FERC determination of a reasonable filed rate. Notably, the First Circuit determined that the ability of plaintiff to terminate its rate schedule in the contract filed with FERC could possibly permit the NHPUC to prohibit plaintiff from recovering all of its costs under the rate schedule, as a valid exercise of the NHPUC’s power to conduct a prudence review. See Patch V, 167 F.3d at 35-36. During the course of the litigation, however, the district court noticed that “buried in the mass of paper submitted” to the court was a fact that blunted defen dant’s position as a matter of law. See Patch VII, 87 F.Supp.2d at 64. Specifically, during the litigation, the NHPUC had rescinded its determination that plaintiff must terminate the rate schedule, because it wished to avoid the termination fee. In light of this, the district court concluded that:

defendant is attempting to whipsaw plaintiff by assuming polar positions. Plaintiff is prohibited from terminating the contract because defendant fears the imposition of an exit fee by FERC but yet cannot recover the wholesale rates it has to pay pursuant to that contract in its retail rates.

Id.

The district court, therefore, entered a preliminary injunction, mandating that defendant allow plaintiff to pass through its costs incurred pursuant to the rate schedule. This determination was affirmed by the court of appeals, which noted that the NHPUC’s “prior justification for disallowing the pass-through has evaporated.” Patch VIII, 221 F.3d at 202. The Patch cases would seem to represent a more or less typical application of the filed rate doctrine: a state agency is not permitted to determine that unavoidable FERC-ap-proved rates are unreasonable, when setting retail rates in a cost-based regime. Yet the Patch cases, by virtue of arising in a context of deregulation, more closely resemble the present dispute than any other filed rate cases in the electricity context.

Nevertheless, despite the lack of precedent specifically on point and the unique features of the regulatory context underlying the instant dispute, the court finds that the filed rate doctrine applies here in much the same way as it does under a cost-of-service regime. The rule adopted by the court may be stated as follows: costs of wholesale energy, incurred pursuant to rate tariffs filed with FERC, whether these rates are market-based or cost-based, must be recognized as recoverable costs by state regulators and may not be trapped by excessively low retail rates or other limitations imposed at the state level.

In light of this rule, the novel features of California’s regulatory scheme are in some ways ultimately irrelevant. Utilities must be able to recover their wholesale costs incurred pursuant to FERC-filed tariffs, even when FERC allows sales of wholesale electricity at prices the market will bear, even when this federal approval is based in part on a retail rate freeze and even when, as here, FERC subsequently has determined that the market-based rates were, at times, unreasonable. The retail rate freeze established by the CPUC was, therefore, intrinsically inconsistent with the FERC-mandated regime for wholesale prices. Undoubtedly, all concerned parties assumed when the rate freeze began that frozen rates would be sufficiently high to avoid the trapping of wholesale costs. When that assumption proved erroneous, however, the CPUC could not avoid its responsibility to give effect to costs incurred within FERC’s regulatory jurisdiction by continued, unyielding reliance upon past enactment of a rate freeze.

In reaching this conclusion the court is mindful of several considerations.

First, from the inception of national regulation of major industry, across all regulated industries to which it pertains, the filed rate doctrine has been strictly and rigidly applied, without concern for the equities of application. This rigid application pays appropriate deference to the determination of Congress in passing the great regulatory acts applying to common carriers that unwavering adherence to rates filed with the appropriate federal regulatory body is the central method by which to avoid price discrimination and unfair market practices and to ensure reliability. This principle was the method chosen by Congress in the FPA for the regulation of public utilities and remains the command of that act. FPA section 205(c) states:

Under such rules and regulations as the Commission may prescribe, every public utility shall file with the Commission, within such time and in such form as the Commission may designate *** schedules showing all rates and charges subject to the jurisdiction of the Commission ***.

Whenever the question has arisen, courts have recognized that under the dual sphere of regulation of electricity, adherence to the filed rate requirement of the FPA, in conjunction with the requirement that utilities provide electricity to end-users, prohibits state regulators from trapping the costs prudently incurred pursuant to FERC-filed tariffs. See, e.g., Nantahala Power & Liqht, 476 U.S. at 970, 106 S.Ct. 2349. Defendants and TURN forcefully argue that the novel regulatory scheme pursued by California regulators relaxes this absolute requirement, but passing fashions in regulatory schemes of the day cannot rework the structure of regulation established by Congress in the FPA. Such arguments — that the introduction of competition into a regulated industry brings into question the continuing application of the filed rate doctrine — have been, to the court’s knowledge, uniformly rejected by courts in the regulatory contexts in which they have been raised. As the Supreme Court stated in another regulatory context (and referencing yet another), in considering a challenge to a decision by the Federal Communications Commission (FCC) to authorize permissive detar-iffing, also justified in light of the introduction of competition in the market:

[Pjetitioners earnestly urge that their interpretation *** furthers the Communications Act’s broad purpose of promoting efficient telephone service. They claim that although the filing requirement prevented price discrimination and unfair practices while AT & T maintained a monopoly over long-distances service, it frustrates those same goals now that there is greater competition in that market. *** But our estimations, and the [FCCj’s estimations, of desirable policy cannot alter the meaning of the Federal Communications Act of 1934. For better or worse, the Act establishes a rate-regulation, filed-tariff system for common-carrier communications, and the [FCC]’s desire “to ‘increase competition’ cannot provide [it] authority to alter the well-established statutory filed rate requirements,” Maislin Industries, US, Inc. v. Primary Steel, Inc., 497 U.S. 116, 135, 110 S.Ct. 2759, 111 L.Ed.2d 94. As we observed in the context of a dispute over the filed-rate doctrine more than 80 years ago, “such considerations address themselves to Congress, not to the courts,” Armour Packing Co. v. United States, 209 U.S. 56, 82, 28 S.Ct. 428, 52 L.Ed. 681 (1908).

MCI Tel. Corp. v. American Tel. & Tel. Co., 512 U.S. 218, 234, 114 S.Ct. 2223, 129 L.Ed.2d 182 (1994).

In strikingly similar language, the Seventh Circuit rejected a challenge to the continuing application of the filed rate doctrine to a deregulated railroad industry in Louisville and Nashville RR Co v. Mead Johnson, 737 F.2d 683 (7th Cir.1984). This case involved a dispute over shipments of milk product by Mead Johnson from Indiana to New Jersey. Originally, Mead Johnson shipped its product on a northern route, but due to delays, several shipments became damaged by cold weather. The parties thereafter agreed to make shipments along a southern route, which had, for reasons not relevant here, a higher rate tariff than the northern route. Mead Johnson was charged, however, the lower northern route rate. Upon discovery of this deviation from the filed rate, the railroad sued Mead Johnson for the undercharge. Despite acknowledging the harsh result and the consequent windfall to the railroad, the Seventh Circuit upheld the lower court in finding for the railroad and ordering Mead Johnson to pay the undercharge. In language with resonance to the present dispute, the court of appeals stated:

Recent legislation moving toward modified deregulation of the railroad industry (involving greater reliance on competitive forces) might suggest the need for a second look at policies requiring strict adherence to filed tariffs. Greater rate flexibility, for example, might seem appropriate where competitive forces are at work and monopoly or monopsony power has abated. However, Congress has not yet amended the Act to permit greater flexibility in situations like the one before us, and we are unwilling to engage in deregulation by adjudication.

Id. at 690 n. 5.

True, the prohibition of trapping wholesale electricity costs by state regulators is not, itself, made explicit in the FPA. Nevertheless, this prohibition has been long recognized as a means to “give effect to Congress’ desire to give FERC plenary authority over interstate wholesale rates, and to ensure that the States do not interfere with this authority.” Nantahala Power & Light, 476 U.S. 953, 106 S.Ct. 2349.

Second, the system of dual regulation established by the FPA, combined with a state’s requirement of the provision of service by IOUs, imposes a special obligation on state regulators to ensure that public utilities are not rendered insolvent because of unavoidable costs. See Board of Public Utility Comm’rs v. New York Tel. Co., 271 U.S. 23, 46 S.Ct. 363, 70 L.Ed, 808 (1926). PG & E was required to provide service to end-users within its geographic area during the rate freeze. Because state regulators require the provision of electricity, an independent, affirmative duty is imposed upon state regulators to permit the recovery of the costs of procurement. In light of this duty, the absence of an explicit determination by the CPUC that FERC-filed rates were unreasonable is of no consequence.

In other words, because of the provision of service requirement of state law, the duties of state and federal regulators under the FPA are not reciprocal. Federal regulators may consider state rates when regulating wholesale rates, see Federal Power Comm’n v. Conway Corp., 426 U.S. 271, 273, 96 S.Ct. 1999, 48 L.Ed.2d 626 (1976), but federal regulators are not obligated to adjust wholesale rates to harmonize with retail rates. In the instant context, therefore, although FERC initially relied, in part, on the existence of a retail rate freeze in approving market based wholesale rates, this approval was not ultimately beholden to the existence of the rate freeze. Although state regulators, on the other hand, imposed a retail rate freeze before market based wholesale rates were approved, the continuing constitutionality of the retail rate freeze depended, in part, on the magnitude of fluctuations in wholesale rates.

Finally, the court notes that the concern in other regulatory contexts that rigid application of the filed rate doctrine results in a windfall to negligent actors is not implicated here. The prohibition of trapping wholesale costs does not result in unwarranted enrichment to a public utility. Even, as here, where market fluctuations led to historically high costs of energy, allowing a utility to pass through these costs to consumers — if that is what is required — would not provide a windfall to the utility, but would merely properly alio- cate the burden of responsibility for the expense of providing a mandated service to the public. “[Cjustomers are entitled to demand service and the company must comply. The company is entitled to just compensation and, to have the service, the customers must pay for it. *** The revenue paid by the customers for service belongs to the company.” New York Tel. Co., 271 U.S. at 31, 46 S.Ct. 363.

Besides determining that, as a general matter, the filed rate doctrine continues to apply in the regulatory context involved here, the court must also reject the more specific claim of defendants and the state of California in an amicus brief that the filed rate doctrine does not apply to the particular market-based tariffs at issue. California asserts that the market-based tariffs and quarterly reports filed by generators and marketers selling power into the PX and ISO do not comply with section 205(c) of the FPA and that, accordingly, the filed rate doctrine is not implicated. But the tariffs at issue in the present dispute are the PX and ISO tariffs under which PG & E purchased its wholesale power, not the generator tariffs pursuant to which power was sold into these entities. Revealingly, California relies on cases in which the filed rate doctrine operates to preclude fraud and antitrust damage claims stemming from the payment of a validly filed rate. See, e.g., AmRepBr (Doc # 171) at 4, citing Florida Municipal Power Agency v. Florida Power & Light Co., 64 F.3d 614, 616 (11th Cir.1995). This issue may arise in California’s attempt to recover damages from generators for sales of wholesale energy, but it is-not implicated here.

Defendants and California also contend that the “pro forma market-based” PX and ISO tariffs are not governed by the filed rate doctrine. See DefBr (Doc # 104) at 20; AmRepBr (Doc # 171) at 5. California argues that these tariffs “do not state the rates to be charged in the market[;] they simply establish a process by which bids submitted by suppliers and purchasers of electricity and ancillary services can be aggregated to generate the market price for a given service in a given interval.” Id. In Duke Energy Trading, 267 F.3d 1042, however, the Ninth Circuit determined that the filed rate doctrine applied to the block-forward contracts, administered by the PX and subject to PX tariffs. The court is not inclined to second-guess this determination. Nor can Duke Energy be distinguished because the relevant provisions of the tariffs at issue there, default mitigation and security provisions, are different from those at issue here. If the filed rate doctrine applies to PX and ISO tariffs, there is no reason that it would apply so selectively. That FERC has decided to set tariffs based on market factors and not with fixed prices does not remove the requirement that costs incurred pursuant to FERC’s regulatory authority must be recovered. See also Transmission Agency of N Cal. v. Sierra Pac. Power Co., 287 F.3d 771, 783 (9th Cir.2002) (noting that pursuant to Order No 888 FERC “has functionally accomplished FERC regulation of rates with FERC regulation of transmission capacity” and determining that the filed rate doctrine applies to this setting of rates “at a competitive level.”).

The rule adopted by the court does not, however, resolve PG & E’s preemption claims; but it does bring into focus the relevant factual issues that must be resolved before a disposition may be reached. Contrary to PG & E’s contention that this matter involves only pure questions of law, factual issues emerge. The CPUC did not, in fact, sit idly by when wholesale rates spiked in the summer of 2000, but took a series of actions intended to address the challenges posed by the energy crisis.

The first action of direct significance taken by the CPUC in response to escalating wholesale costs was the issuance of D-00-12-067 on December 12, 2000, shortly after FERC had acted to remove the price cap of $250/megawatt for wholesale electricity. In response to escalating prices and the removal of this cap, the CPUC determined that expedited hearings would be necessary to decide how to balance the CPUC’s responsibility both to the IOUs and to California consumers. See PRJN (Doc # 117, Exh # 15):

In this decision, we address the financial difficulties facing Pacific Gas and Electric Company (PG & E) and Southern California Edison Company (Edison) as a result of the severe problems in the wholesale electric market. *** The financial health of California’s utilities and continued financial viability are critical. We are concerned about the specter of potential financial failure that the utilities have raised. These difficulties have been exacerbated by the December 8 and December 15 order of the [FERC] that removed the price cap of $250/megawatt, as requested by the Independent System Operator (ISO). This action has caused prices for electricity obtained through the Power Exchange (PX) and ISO’s real-time energy markets [to rise] to unprecedented levels and has pushed the resources of California’s investor-owned utilities to the breaking point. This problem occurs because PG & E and Edison are charging rates frozen at 1996 levels, as mandated by PubUtilCode § 368(a), but must procure power at market-based rates. The elimination of price caps by FERC on December 8 and the resulting five-fold increase in electricity prices has expanded the crisis to one that involves not only utility solvency but the very liquidity of the system.

Id. at 2.

After conducting emergency hearings on these matters, the CPUC implemented an interim surcharge, or raise in frozen rates, on January 4, 2001, in order D-01-01-018:

[W]e will allow PG & E and Edison each to raise their revenues by increasing the electric bill of each customer by one cent per kilowatt hour (kWh) applied on a usage basis. The surcharge will be applied on an equal cents per kWh basis and will result in an increase of approximately 9% for residential customers, 7% for small business customers, 12% for medium commercial customers, and 15% for large commercial and industrial customers. *** The increase will be a temporary surcharge to improve the ability of the applicants to cover the costs of procuring future energy in wholesale markets that they cannot produce themselves to serve their loads.

PRJN II (Doc # 117, Exh # 14) at 1-2.

On March 27, 2001, after, among other things an independent accounting review of the utilities’ claims of the urgency of their financial problems, the CPUC issued D-01-03-082, which contained two important components. First, determining that “[w]e have come to the bitter moment when the record shows that additional ratepayer money must be provided,” PRJN II (Doc # 117, Exh # 12) at 12, the CPUC made permanent the one-cent increase authorized on an interim basis by D-01-01-018 and authorized an additional increase of three-cents per kWh, subject to certain conditions and to possible refund. See id at 16-18. These increases effected a substantial increase in the cost of electricity for end-users.

Second, the CPUC adopted the accounting “true-up” discussed above, that requires PG & E and SCE to match total operating costs against total operating revenues by transferring the balance in the TRA each month, whether positive or negative, to the TCBA, over the period of the rate freeze. Id at 24. As previously discussed, the accounting true-up requires PG & E and SCE to apply revenues from the sale of capital assets and from selling electricity generated by their own plants against operating costs, before applying any excess to paying down stranded costs.

Resolving PG & E’s preemption claims requires determining the sufficiency of these, and other, actions by the CPUC during the period of the wholesale spikes, in light of the legal rule adopted by the court. Unsurprisingly, the parties present these actions and their effects in a considerably different light. Defendants and TURN contend that the accounting true-up properly applies operating costs against total revenues over the appropriate period (the transition period), that PG & E’s revenues actually “greatly exceeded its operating costs” over this period, DefBr (Doe # 104) at 15, and that, accordingly, PG & E’s financial problems are the result of financial mismanagement and irresponsibility. It has, in fact, long been the position of defendants that by this lawsuit PG & E is improperly attempting to hold ratepayers responsible for irresponsible financial management. The CPUC, for example, noted in its January 2001 decision:

As reported in the monthly TCBA reports, PG & E has received over $9 billion in headroom and other transition cost revenues ***. [Disbursements, from PG & E to the parent company, PG & E Corporation (PG & E Corp) during the transition period were approximately $9.6 billion. Out of this total, PG & E Corp issued dividends (both common and preferred stock) of approximately $1.5 billion. PG & E also repurchased stock in the amount of approximately $2.8 billion and retired approximately $2.8 billion of debt. PG & E recognized that market problems were beginning to occur in June of this year, but decided to declare a third-quarter divided. PG & E did not consider establishing a contingency fund or retaining cash to cushion its risk ***.

D01-01-018, PRJN II (Doc #117, Exh # 14) at 15-16.

Defendants and TURN also assert that if, in fact, PG & E matched its revenues from sources, including headroom revenue from earlier in the rate freeze, gains on sales of generation assets and revenue from the sale of power into the PX, against its operating costs, PG & E would not suffer a deficit over the period of the rate freeze, but would have approximately a $7.7 billion dollar surplus, that could be applied, at least in part, to pay down stranded costs. See Long Decl (Doc # 105) at ¶¶ 84-37. Defendants and TURN contend that this netting of a broad category of revenues over the rate freeze period as directed by the accounting true-up is true to the intentions of California regulators in instituting AB 1890 and does not impermissibly interfere with FERC’s regulatory authority. Defendants and TURN also contend, of course, that the filed rate doctrine does not absolutely require the recovery of wholesale costs in the present context, but this contention has been rejected by the court.

PG & E, on the other hand, argues that the accounting true-up is a mere bookkeeping measure that does nothing, in fact, to remedy the undercollection experienced during the wholesale price spike. PG & E contends that pursuant to federal law only its retail rate revenues may be set against its operating costs and that not even its revenues from electricity sold to itself on the PX may be considered revenue for cost recovery purposes. PG & E further contends that the accounting order improperly attempts retroactively to identify revenues already received and spent for other purposes, including headroom received before the price spike, as offsetting PG & E’s wholesale procurement cost undercollection. PG & E asserts that, properly calculated, it has suffered an undercollection of approximately $6.7 billion from June 2000 going forward and must be able to recover this undercollection prospectively in retail rates.

Resolving this dispute requires consideration of two issues: First, the revenue sources that may be applied within constitutional bounds against PG & E’s operating costs and, second, the appropriate period for considering whether a net un-dercollection has, in fact, occurred.

Taking the latter issue first, PG & E’s theory in this litigation is not easily comprehended. Initially, PG & E’s position was that in order to satisfy the requirements of the filed rate doctrine retail rates must adjust essentially concurrently to cover ongoing wholesale power procurement costs. TURN argued in its opening brief in support of its summary judgment motion that such a theory does not translate well from a cost-of-service regulatory regime to one in which wholesale prices are set by the market and in constant flux:

When wholesale power is purchased at a fixed price and resold at a different fixed price, the spread is uniform across all sales. If, under such fixed-price tariffs, that spread is negative *** that will be true for each kilowatt-hour, for cumulative sales in any period, and cumulatively for all purchases and sales over the life of the disputed tariff. But when, say, the wholesale power is purchased at continuously varying prices, even if the retail price at which the power is resold is fixed, how does one evaluate whether the utility is or is not recovering its wholesale costs in retail rates? *** Could it seriously be argued that each kilowatt-hour purchased at a rate above the retail level is a filed-rate violation, even as other kilowatt-hours are purchased at a cost so low that the spread would more than offset the loss on the high-cost kilowatt hour?

TURN Br. (Doc # 119) at 11.

In response, PG & E agreed in its opposition brief “that such an instantaneous comparison of costs and revenues is not sensible for purposes of determining whether an undercollection exists” and that, instead, this determination should be made on a monthly basis. PG & E OppBr (Doc # 148) at 31. According to this theory, in months in which an undercollection exists, the CPUC must allow for the recovery of that undercollection “in retail rate revenues over a reasonable prospective period.” Id. at 32.

Attempting to express the reason why a month-by-month accounting and rate adjustment is “sensible,” although a concurrent accounting is not, PG & E notes that it is simply borrowing the accounting period established by the CPUC for tracking the recovery of stranded costs during the rate freeze. The fact that the CPUC has required PG & E to submit monthly reports summarizing its TRA balance does not furnish legal cause for settling on monthly adjustment to retail rates as the period of adjustment commanded by the filed rate doctrine. Monthly adjustments may be sensible and practical, but the court does not perceive that they are required as a matter of law.

Any period other than the period of the rate freeze itself — the relevant period urged by defendants and TURN — picked in the abstract would appear to be arbitrary. Nonetheless, state regulators cannot turn a blind eye to the risk of illiquidity to which an overlong accounting period may expose PG & E and thereby do through the selection of an accounting pe riod what the filed rate doctrine does not permit.

What case law there is on this issue also suggests that a monthly accounting and mandatory retail rate adjustment are not required by the filed rate doctrine and that some substantial deference must be accorded to state accounting and adjustment procedures. In Kentucky West Virginia Gas Co. v. Pennsylvania Public Utility Com., 862 F.2d 69 (3d Cir.1988), the Third Circuit considered a challenge based on the filed rate doctrine to a method administered by the Pennsylvania Public Utility Commission (PPUC) for setting retail natural gas rates. Plaintiff Equitable, a retail distributor of natural gas, purchased natural gas wholesale, pursuant to FERC-filed tariffs, from several wholesaler suppliers including its affiliate, Kentucky West, through long-term contracts. The contract between Equitable and Kentucky West required Equitable to pay a minimum bill obligation, pursuant to Kentucky West’s FERC-approved rate schedule, reflecting Equitable’s contract entitlement.

Pursuant to state legislation, the PPUC considered a proposal for a retail rate change by conducting a forward-looking and backward-looking review, considering both the utility’s future cost estimates and its actual cost performance in the previous rate period. As part of its backward-looking analysis, in setting the new rate the PPUC could either refund to consumers a past overcollection or allow the utility to recover a past undercollection. In the challenged order, as part of its forward-looking analysis, the PPUC set retail rates based on the assumption that FERC would approve reductions in Equitable’s contract entitlement with Kentucky West in the upcoming period, which would lead to a corresponding decrease in Equitable’s FERC-approved minimum bill obligation. Accordingly, the PPUC excluded $865,000 of anticipated wholesale procurement costs in calculating the new retail rates. Equitable filed suit, charging that this failure to pass on its minimum bill obligation immediately violated the filed rate doctrine. The court of appeals overruled the district court, denied Equitable’s challenge and concluded that “[i]f a state wants to use such an adjustment mechanism the preemptive force of FERC jurisdiction does not prevent it, so long as the full FERC-approved filed rate is ultimately passed-through to the utility’s retail customers.” Id. at 74.

This holding reflects an appropriate measure of deference to the state regulatory scheme chosen for adjusting retail rates to meet filed rate obligations. Pursuant to this deference, within a period of a fixed-rate, deficits in sub-periods may be applied against surpluses in sub-periods and, indeed, pursuant to the Third Circuit’s analysis, the entire period of a fixed-rate may result in an undercollection, as long as that undercollection is recovered in the subsequent period. In light of these quite sensible considerations, PG & E’s contention that an undercollection in a single month is an immediate violation of the Supremacy Clause, requiring a future recovery of that undercollection without respect to surpluses in surrounding months, appears untenable.

Yet deference to a state’s system for accounting costs and revenues and adjusting retail rates is not without limit. Clearly, in light of the above, a refusal to allow for the recovery of an undercollection over the course of the transition period in future retail rates would be a violation of the filed rate doctrine, but, under certain circumstances, a state’s retail adjustments or lack thereof during a fixed-rate period in which a utility does not suffer a net under-collection may also impermissibly intrude on federal regulatory authority.

The instant circumstance may be one in which a state’s actions during a period of a fixed-rate may amount to a constitutional violation whether or not PG & E suffered a total undercollection under the rate freeze. For one thing, the period pressed by defendants and TURN as the appropriate accounting period is quite long: approximately four years. For another, the price spike occurred toward the end of this period and the accounting order, requiring the application of total revenues against total costs, occurred after the initial price spike. Accordingly, it is unclear whether the entire rate freeze is the most appropriate period over which to consider whether undercollection has resulted or whether some sub-period is more appropriate.

Kentucky West Virginia Gas Co is again instructive. Besides referring to the question of an “ultimate[ ]” undercollection, the Third Circuit also noted, in a footnote, that a delay in passing through FERC-mandat-ed costs “for an unreasonable period of time” could also amount to a constitutional violation. Id. at 75 n. 11. PG & E in its opening brief in support of summary judgment similarly invoked a reasonableness standard, referring to the obligation of the CPUC to provide “meaningful rate relief.” PG & E Br (Doc # 111) at 9. The court is indeed convinced that under the circumstances of this case, the relevant period for determining the existence of an undercol-lection may not be determined in the abstract, nor assumed to be the entire transition period, but must be some meaningful period, reasonable under the circumstances, which may be the transition period or some shorter period within it. An inherent feature of the filed rate doctrine is that it must rigidly be applied; but the effect of its application is not clear in the abstract and must rather be gleaned from issues of fact and considerations of the case at hand.

Making the determination whether the entire transition period or some part of it is the reasonable period for determining the existence of an undercollection may involve considering such factual issues as the understanding of the parties at the beginning of the transition period and their reasonable expectations, the reasonableness of PG & E’s use of surplus revenues during the first years of the transition period and the availability of this and other revenue during the energy crisis. Based on such considerations, the entire transition period may well be too long, but any individual month will likely be too short. The court cannot make this determination as a matter of law and undisputed fact on the record before it.

PG & E, relying upon Board of Public Utility Comm’rs v. New York Tel. Co., 271 U.S. 23, 46 S.Ct. 363, 70 L.Ed. 808 (1926), claims that the CPUC may not retroactively apply past surpluses against present deficits to satisfy its obligations under the filed rate doctrine, but that present deficits may only be recovered prospectively. There is some validity to this claim. But the right to prospective recovery of past undercollection does not solve the dilemma posed by PG & E’s preemption claims. The court has determined — and on this point defendants and TURN are incorrect — that if PG & E has suffered an undercollection it must be able to recover that undercollection prospectively. But it must first be determined whether PG & E has, in fact, actually experienced an under-collection over a meaningful period of time. Because the wholesale price spike occurred toward the end of a fixed-rate period and because the court has determined that a concurrent undercollection is not, by itself, a filed rate doctrine violation, some past surpluses may be relevant in determining whether an undercollection has, in fact, occurred. The question is which past surpluses are considered in this determina tion and what revenue sources may be considered in determining the existence of past surpluses.

The court now turns to this issue: the revenue sources that may properly be considered in determining the existence of an undercollection.

Before the accounting true-up, PG & E may have had a stronger claim for entitlement to judgment on its preemption claims as a matter of law and undisputed fact. Pursuant to the CPUC’s most recent accounting decision before the true-up, headroom in the TRA was transferred on a monthly basis to the TCBA, while under-collection was not. Accordingly, during the first two years of the rate freeze, PG & E collected billions of dollars of surplus in the TRA, which was transferred to the TCBA on a monthly basis and applied against transition costs. In the TCBA, meanwhile, revenues from power plant sales, sales of electricity on the PX and proceeds from rate reduction bonds were also being applied to pay down stranded costs. When wholesale prices spiked and began regularly to exceed frozen retail rates, liabilities began to mount in the TRA, although stranded costs were still being paid down in the TCBA in large part from PG & E’s electricity sales.

Notwithstanding rising undercollection, the CPUC determined that liabilities in the TRA would not be able to be collected at the end of the transition period. Accordingly, absent the accounting true-up, PG & E’s filed rate claim would be of at least more superficial clarity, as the revenues marked by California for the recovery of operating costs, retail rates, would be insufficient over a fixed-rate period, thereby giving the appearance that wholesale costs had been trapped.

The accounting true-up in some ways, however, exposes the superficiality of such a conclusion. In the accounting true-up decision, the CPUC addressed the apparent anomaly of its current system for accounting revenues and liabilities, whereby PG & E appeared both to be suffering tremendous liabilities and applying tremendous surpluses to pay down stranded costs. The CPUC determined that its current regulatory accounting mechanisms overstated power purchase liabilities, because those liabilities were not set against revenues from, among other things, power sales. See D-01-03-082 (RJN II, Exh # 12) at 24. Determining that such accounting was flawed, the CPUC, as noted, required operating costs to be set against total revenues, before surpluses could be used to pay down stranded costs.

PG & E contends that this accounting adjustment is preempted by federal law, but the court is not convinced that the filed rate doctrine extends to invalidate state adjustments to the accounting of surpluses and liabilities. Although federal law may provide some substantial relief for PG & E, it does not appropriately function to give PG & E a windfall by trapping state regulators in an accounting system improvidently designed, particularly in light of events apparently unforeseen by either party. The CPUC developed the accounting system and made changes to it pursuant to a statutory mandate. Indeed, the accounts themselves were created by a CPUC decision, D-97-10-057, and originally interacted in the manner currently challenged by PG & E. See Res E-3527, DRJN Exh G. To the extent PG & E believes that the adoption of the true-up was a dereliction of the CPUC’s statutory authority, this would seem to be a matter of state law and PG & E has, in fact, brought such a challenge in state court.

The federal preemption challenge to this order, on the other hand, involves the question whether it is permissible pursuant to the filed rate doctrine for a state commission to consider revenue from non- retail sources when setting retail rates. Was the CPUC, in other words, preempted from considering PG & E’s revenue from, among other things, sales of electricity into the PX when determining that the frozen retail rates, with limited modifications, were sufficient to cover operating costs?

Quoting selectively from some seminal filed rate cases, PG & E argues that the filed rate doctrine forbids state regulators from looking to sources of revenue other than retail when determining whether existing rates are sufficient to cover operating costs. PG & E in particular relies on language from Mississippi Power & Light, in which the Supreme Court stated that the state commission must treat the utility’s “FERC-mandated payments *** as reasonably incurred operating expenses for the purpose of setting [the utility’s] retail rates” and that “States may not bar regulated utilities from passing through to retail consumers FERC-mandated wholesale rates.” MP & L, 487 U.S'. at 370, 372, 108 S.Ct. 2428. But such language does not reflect an essential requirement of the filed rate doctrine that state commissions must be blind to other sources of revenue for regulated utilities besides retail revenue when setting retail rates. In the regulatory context in which Mississippi Power & Light and Nantahala Power & Light arose, of course, retail revenues were, essentially, the only revenues relevant to a state utility commission’s determination of appropriate retail rates. But the filed rate doctrine itself in this context only prohibits FERC-mandated costs from being trapped; it does not require that retail rates be excessively high when utilities are receiving revenue from other sources. Retail rate setting based on a regulated utility’s “overall financial structure” may be reasonable. Narragansett Elec. Co. v. Burke, 119 R.I. 559, 381 A.2d 1358, 1363 (R.I.1977). See also Nantahala, 476 U.S. at 967-68, 106 S.Ct. 2349 (noting that “an increase in FERC-approved wholesale rates need not lead to an increase in retail rates *** if costs other than those resulting from the purchase of FERC-regulated power or gas were to decrease” and citing Na'iTagansett).

Moreover, in considering challenges to state rate making orders under the Takings Clause, the Supreme Court has considered “[t]he overall impact of ratemaking orders,” when making the determination whether such orders are constitutionally objectionable. Duquesne Light Co. v. Barasch, 488 U.S. 299, 312, 109 S.Ct. 609, 102 L.Ed.2d 646 (1989). See also id. at 314, 109 S.Ct. 609 (“It is not theory, but the impact of the rate order which counts”)(quoting Federal Power Com. v. Hope Natural Gas Co., 320 U.S. 591, 602, 64 S.Ct. 281, 88 L.Ed. 333 (1944)); Verizon Communs, Inc. v. FCC, — U.S.-, 122 S.Ct. 1646, 1680, 152 L.Ed.2d 701 (2002) (noting that in Duquesne, “we held as usual that a rate setting methodology would normally be judged only by the ‘overall impact of the rate orders.’ ”).

The conclusion that the filed rate doctrine does not categorically prohibit state utility commissions from considering other revenue sources seems particularly appropriate in the regulatory context involved here, in which much of the non-retail rate revenue received by PG & E was directly tied to the existence of the retail rate freeze. Revenue from the rate reduction bonds, for example, which were issued to finance the 10% rate reduction for residential and small commercial customers, although perhaps not strictly retail rate revenue would seem to be appropriately considered as offsetting operating costs. See D-01-03-082 (RJN II, Exh # 12) at 38.

At least some revenue from sales into the PX, too, could likely be considered by the CPUC without offending the filed'rate doctrine. The purchase of wholesale elec tricity on the PX by PG & E, pursuant to AB 1890, undoubtedly added to PG & E’s operating costs as compared to PG & E’s previous operation as a vertically integrated utility; but so too did the sale of electricity on the PX add to PG & E’s revenues. Although again not strictly retail revenues, it would not seem necessarily to offend filed rate doctrine principles to factor into a retail rate calculation the revenue gained from such sales. Indeed, while AJB 1890 was being instituted, PG & E itself recognized that costs of the purchase of electricity would be “offset” by revenues gained from sale on the PX. See PG & E Phase II filing, in DRJN (Doc # 106, Exh E) at 8-9.

Yet the court, again, does not believe it appropriate to make the determination what revenue sources available to PG & E may be considered in determining whether an undercollection exists as a matter of law, nor is summary judgment on this issue appropriate on the record before the court. The rule adopted by the court to apply to this dispute would find defendants in violation of the filed rate doctrine if they kept retail rates excessively low and thereby trapped wholesale costs. Pursuant to this rule, if retail rates alone were insufficient to cover costs, but were sufficient in combination with other revenue, properly considered, then wholesale costs would not necessarily be trapped and there would be no violation of the filed rate doctrine. Determining whether, as defendants and TURN claim, this has occurred, however, will require concrete analysis of factual circumstances.

Determining whether gain from sales of power plants, for example, may be set against operating costs may involve consideration of historical factors, such as the allocation of investment and risk of investment in these plants between shareholders and ratepayers. See Democratic Cent. Com of DC v. Washington MAT Comm’n, 485 F.2d 786 (D.C.Cir.1973). Determining whether and if so what portion of electricity sales and headroom recovered during the beginning of the transition period, or some shorter period, may be applied against operating costs will require consideration of factors such as the availability of this revenue to PG & E during the relevant period. Just as PG & E’s claim that only retail revenue may be considered must be rejected as overly broad, so too must defendants’ and TURN’S claim that any and all revenue accounted for over the course of the rate period may be applied against operating costs, whether or not such revenue was actually available to sustain operations when the wholesale price spike hit and when the accounting order was entered, be rejected as excessively formalistic. If revenue was simply not available to sustain PG & E’s operations, and this unavailability was reasonable and not due to financial mismanagement, the CPUC’s reliance upon the existence of that revenue “on the books” is insufficient to meet its obligations under the filed rate doctrine.

Ultimately, broad legal principles will not resolve whether the CPUC kept retail rates excessively low in the face of escalating wholesale costs. Issues of fact, considerations of reasonableness and proper allocation of risk must be considered.

D

Finally, the court must consider defendants’ claim to the right to conduct a review of the prudence of PG & E’s wholesale purchases, under the so-called “Pike County exception,” which receives it name from a Pennsylvania state court decision: Pike County Light & Power Co. v. Pennsylvania Pub. Util. Comm’n, 77 Pa.Cmwlth. 268, 465 A.2d 735 (1983). Under the Pike County exception, state commissions are not prevented from conducting a review of a utility’s purchasing options and may disallow wholesale costs, when setting or adjusting retail rates, that were imprudently incurred. “The Supreme Court has never squarely decided the question of whether imprudence is an escape hatch from [a] Commission’s otherwise existing obligation to respect FERC’s authority to determine the just and reasonable rate. But the Court has twice said that it would assume arguendo that such escape hatch existed; the Third Circuit has so held; and FERC has concurred, citing prior cases of its own.” Patch V, 167 F.3d at 35, citing MP & L, 487 U.S. at 373-74, 108 S.Ct. 2428; Nantahala, 476 U.S. at 972, 106 S.Ct. 2349; Kentucky W Va. Gas Co. v. Pennsylvania Pub. Util. Comm’n, 837 F.2d 600, 609 (3d Cir.1988); Palisades Generating Co., 48 FERC 61,144, 1989 WL 262105 at 61,574 and n 10 (1989). The First Circuit must also be added to this list.

As noted, PG & E advances two preemption claims: one concerning events before and one after August 3, 2000. Initially, PG & E was obligated to make all wholesale purchases on the PX through a variety of submarkets, including markets for power for immediate use (real-time markets), hour-ahead markets and day-ahead markets. After 1998, the ISO became an additional market. In July 1999, the CPUC created a new block-forward market operated by CalPX Trading Services (CTS) which allowed for the purchase of electricity for extended periods in an attempt to improve price stability. See Res E-3618 (PRJN II, Exh #22). In March 2000, the CPUC expanded the block-forward market. See Res E-3658 (PRJN II, Exh # 20).

In an August 3, 2000, decision, the CPUC authorized PG & E and SCE to enter into bilateral power contracts directly with wholesale electricity suppliers outside the PX and ISO. See D-00-08-023 (RJN II, Exh # 16). PG & E concedes that the CPUC may conduct a limited-reasonableness review into the prudence of costs incurred under PG & E’s bilateral contracts. See PG & E Br (Doc # 111) at 24. But PG & E contends that the CPUC may not conduct a prudence review of PG & E’s bidding strategies within the sub-markets of the PX and ISO. PG & E makes, essentially, an estoppel claim, asserting that such a review is not permitted because the CPUC repeatedly stated that all purchases on these markets were “deemed reasonable.” PG & E also contends that such a review is not permitted by the filed rate doctrine, because determining the prudence of PG & E’s bidding strategies would require theorizing about what market rates might have been under hypothetical alternative bidding scenarios.

At this point, however, this dispute is merely academic. Despite the claim now to the right to conduct a prudence review, the fact is that the CPUC did not disallow an increase in retail rates to recover an acknowledged undercollection because it had determined that PG & E’s wholesale costs were, in part, imprudently incurred. Defendants appear to be arguing that the court should decline to resolve PG & E’s preemption claims because of the possibility that the CPUC may decide to conduct a prudence review and that review may lead to the conclusion that some of PG & E’s costs were imprudently incurred. But this contingency does not relieve the court of its responsibility to consider PG & E’s preemption claims. Moreover, a prudence review will likely not result in the CPUC determining that a prospective increase in retail rates is necessary. In this litigation, defendants have taken the position that PG & E may not recover in future retail rates costs incurred during the rate freeze. If it is determined that this position is erroneous and PG & E is entitled to recovery of past losses in future retail rates, then the court’s determination of the bounds of a permissible prudence review may become meaningful. Currently, however, defendants’ invocation of the Pike County exception is merely an attempt to hedge their bet, and it gives the court no great pause to turn down this attempt.

IV

In sum, judgment is not appropriate on PG & E’s preemption claims as a matter of law. Substantial factual issues remain in dispute and in need of development. Accordingly, all dispositive motions must be denied.

• TURN’S motion to intervene (Doc ## 18, 79) is GRANTED. TURN is permitted to intervene as of right and, in the alternative, permissively.

• TURN’S motion to dismiss (Doc ## 20, 80) is DENIED.

• Defendants’ motion to dismiss (Doc ## 24, 77) is DENIED.

• Defendants’ motion for summary judgment (Doc # 104) is DENIED.

• PG & E’s motion for partial summary judgment (Doc # 111) is DENIED.

• TURN’S motion for summary judgment (Doc # 119) is DENIED.

• Defendants’ motion for leave to file declaration under seal (Doc # 140) is GRANTED. The clerk is directed to file the lodged document under seal.

• Defendants’ motion for leave to file statement of recent decision (Doc # 189) is GRANTED.

A case management conference in this matter is hereby set for August 16, 2002, at 10:00 am.

IT IS SO ORDERED.

1 . Richard A Bilas, a former defendant, has resigned from the California Public Utilities Commission, and Michael R Peevey has been appointed to fill the vacancy. Pursuant to FRCP 25(d)(1), Peevey is substituted by operation of law as a defendant in this action.

2 . The court's discussion of the historical origins of the filed rate doctrine is indebted in particular to the following articles: Joseph D Kearney, Thomas W Merrill, The Great Transformation of Regulated Industries Law, 98 Co-lumLRev 1323 (1998); Wayne Johnson, The Negotiated Rates Act of 1993: Congress Curtails Undercharge Litigation in Bankruptcy by Amending the Filed Rate Doctrine, 68 Am-BankrLJ 319 (1994); Howard R Rubin, Reiter i> Cooper and Unreasonable Rates: Are Reports of the Filed Rate Doctrine’s Death Greatly Ex-agerrated?, 42 Duke LJ 905 (1993); Rene Sa-casas, The Filed Tariff Doctrine: Casualty or Survivor of Deregulation?, 29 DuqLRev 1 (1990).