In Re Pacific Gas & Electric Company

295 B.R. 635 (2003)

In re PACIFIC GAS & ELECTRIC COMPANY, Debtor.

No. 01-30923DM.

United States Bankruptcy Court, N.D. California.

May 15, 2003.

Benjamin Hoch, Dianne Coffino, Marc Hirschfield, Michael C. Hefter, Robert C. Myers, Law Offices of Dewey Ballantine, New York City, David Agay, Stacy D. Justic, Law Offices of Winston and Strawn, Chicago, IL, Adam M. Cole, Heller, Ehrman, White and McAuliffe, J. Michael Kelly, Martin S. Schenker, Law Offices of Cooley Godward, James L. Lopes, Janet A. Nexon, Jeffrey L. Schaffer, Kimberly A. Bliss, William J. Lafferty, Howard, Rice, Nemerovski, Canady, Jon B. Streeter, Robert A. Van Nest, Law Offices of Keker and Van Nest, San Francisco, CA, Jamie L. Edmonson, Richard Levin, Skadden, Arps, Slate, Meagher and Flom, Los Angeles, CA, John S. Moot, Skadden, Arps, Slate, Meagher and Flom, Washington, DC, for debtor.

Patricia A. Cutler, Stephen L. Johnson, Office of the U.S. Trustee, San Francisco, CA, for Office of the U.S. Trustee.

Lorie A. Ball, Michael I. Sorochinsky, Paul S. Aronzon, Robert J. Moore, Milbank, Tweed, Hadley & McCloy, Los Angeles, CA, for Official Committee of Unsecured Creditors.

MEMORANDUM DECISION ON ESTIMATION OF ANTITRUST CLAIMS

DENNIS MONTALI, Bankruptcy Judge.

I. Introduction

A. Procedural Background

In November 2002, the Northern California Power Agency ("NCPA") and the City of Palo Alto ("Palo Alto") (together, "Objectors") and Debtor, Pacific Gas & Electric Company ("PG & E"), entered into (and the California Public Utilities Commission (the "CPUC") approved) the Amended Stipulation And [Proposed] Order Re Procedures For Estimating Certain Disputed And Unliquidated Claims of the Northern California Power Agency And City of Palo Alto For Feasibility Purposes Only ("Estimation Stipulation"). As set forth in the Estimation Stipulation, Objectors contend that:

the PG & E Plan and the CPUC Plan are not feasible . . . because they both fail to appropriately provide for damages attributable to certain disputed and unliquidated claims (the "Municipal Claims") of NCPA and Palo Alto based on PG & E's alleged breaches of the "Stanislaus Commitments," Section 2 of the Sherman Act and related alleged wrongs, which claims are described in the Opposition Of The City of Palo Alto To Motion of Pacific Gas & Electric Company For A Protective Order . . . the Palo Alto Objection and the NCPA Objection. (Estimation Stipulation at 1.)

The Estimation Stipulation provides a process for estimation of Objectors' Municipal Claims for purposes of determining plan feasibility. It is to serve no other purpose. It does not estimate any claim of NCPA, Palo Alto, or any other party for allowance, distribution, or any other purpose. The sole reason the court has undertaken this analysis is to ascertain what amount of damages, if any, PG & E should include in its forecasts for meeting obligations that "pass through," i.e., are not dealt with, under its proposed Plan Of Reorganization (as amended, the "Plan").

The Estimation Stipulation provided for a three-day estimation trial, with a maximum of five percipient witnesses, and three expert witnesses, per party (NCPA and Palo Alto being one "party" for these purposes), together with such written exhibits (including deposition testimony and declarations) and demonstrative exhibits as each party offered. Because of the abbreviated nature of the estimation trial, the parties also agreed that the witnesses' testimony would be presented in writing and that cross-examination would be by way of deposition testimony taken of the witnesses before trial. Finally, all agreed that the evidence offered by the parties would be received subject to the Court's rulings on written objections the parties were permitted to file.^[1]

Trial was conducted on January 27, 28 and 29, 2003. Proposed findings of fact and conclusions of law were submitted on March 26, 2003, after which the matter was considered submitted for decision.

Although the "Municipal Claims" were defined in the Estimation Stipulation to include a broader range of contingent claims, Objectors chose to limit their evidence and presentation to their alleged claims arising under Section 2 of the Sherman Act (15 U.S.C. § 2) ("Section 2") and related state antitrust and unfair competition claims (together, the "Antitrust Claims").^[2]

Because the Court's estimation of the Antitrust Claims of Objectors bears not only upon the feasibility of the Plan, but also upon the feasibility of the competing plan filed by CPUC (the "CPUC Plan"), CPUC was given a full opportunity to participate in the estimation trial.^[3] CPUC did not designate any witnesses or allow them to be deposed before trial, and did not offer any evidence at the estimation trial.

B. Objectors' Contentions

Objectors' principal contention is that PG & E has attempted illegally to maintain a monopoly in the market for the distribution of electricity to residential and business customers in PG & E's Northern California service territory in violation of Section 2 (and analogous state law doctrines) by failing to provide transmission services over PG & E's transmission facilities in Northern California on just and reasonable terms. A substantial part of PG & E's Northern California transmission system is a "strategic bottleneck" facility, particularly the PG & E lines that transmit electricity into and within the Greater Bay Area ("GBA"). In particular, Objectors contend that under the "Stanislaus Commitments" PG & E is required to provide them with "firm transmission," which Objectors define to mean transmission free from costs associated with congestion. PG & E is required by Section 2 to transmit ("wheel") electricity to Palo Alto and NCPA's other members "on fair and reasonable terms that do not disadvantage them."

The primary exclusionary acts alleged by Objectors include the following: PG & E's alleged reliance upon costly local generation to supplement, and thereby avoid the need to improve, an allegedly deficient transmission system; PG & E's failure to designate the existing interconnection agreements between PG & E and Objectors as "existing transmission contracts" ("ETCs") that might be protected from future market reforms; PG & E's alleged improper termination of those contracts; PG & E's failure to negotiate replacement agreements or an alternative resolution that would ensure that Objectors would not incur congestion charges (including PG & E's refusal to sell Objectors a portion of PG & E's transmission system); and PG & E's divestiture of generation assets without taking steps to ensure that this would not increase Objectors' exposure to increased congestion charges.

Objectors also allege that, by raising its local distribution rivals' costs, PG & E is attempting to place Objectors in an anti-competitive price squeeze and maintain its local distribution monopoly in a manner forbidden by Section 2.

C. Ruling

For the reasons explained below, the court concludes that Objectors have not established that the Antitrust Claims will affect the Plan's feasibility. Therefore, solely for feasibility purposes, the court will estimate the Antitrust Claims as having no value.

II. Estimation Procedures

No complaint asserting the Antitrust Claims has been filed. Thus, the court cannot approach the matter at hand in the traditional way a United States district court would deal with a motion for a judgment on the pleadings (Fed.R.Civ.P. 12(c)), a motion for failure to state a claim upon which relief can be granted (Fed.R.Civ.P. 12(b)(6)), a motion for summary judgment (Fed.R.Civ.P. 56), or any other case-dispositive motion.

Nevertheless, the Estimation Stipulation lets the court engage in the little make believe, viz., to act as if the court were determining Antitrust Claims at a future date after the Plan had become effective, to accept the undisputed facts, to find facts where there are material disputes, to consider the legal principles advanced by Objectors to support their Antitrust Claims, and to consider the defenses tendered by PG & E. Then, unlike the more conventional estimation "for purpose of allowance" (11 U.S.C. § 502(c)), the court is to glean from all before it what PG & E should presume are its liabilities to Objectors on account of the Antitrust Claims in order to determine whether the Plan is feasible under Section 1129(a)(11).^[4] If the Antitrust Claims are too high, then the Plan may not be feasible; if they are too low — as the court has determined — then PG & E (and CPUC) need not worry about the Antitrust Claims for Plan confirmation purposes.^[5]

There are relatively few guidelines for the court. Section 502(c) provides little direction and the cases interpreting that section give the court wide discretion.

Section 502(c) requires the court to estimate "for purpose of allowance" any contingent or unliquidated claim, "the fixing or liquidation of which, as the case may be, would unduly delay the administration of the case[.]" 11 U.S.C. § 502(c)(1). Section 502(c) additionally requires the court to estimate (for purposes of allowance) "any right to payment arising from a right to an equitable remedy for breach of performance." 11 U.S.C. § 502(c)(2).

An estimation under section 502(c) may be for broad or narrow purposes. For example, the court may estimate a claim solely for the purpose of determining a creditor's ability to vote on a plan of reorganization or solely for the purpose of determining feasibility of a plan. See Pizza of Hawaii, Inc. v. Shakey's, Inc. (In re Pizza of Hawaii, Inc.), 761 F.2d 1374, 1382 (9th Cir.1985) (estimation necessary for a determination of plan feasibility); In re Trident Shipworks, Inc., 247 B.R. 513, 514 (Bankr.M.D.Fla.2000) ("the estimation proceeding may be used for the purpose of voting on a Plan of Reorganization, and also to determine the allowed amount for distribution purposes"). Cf. 4 Collier on Bankruptcy ¶ 502.04[3] (15th ed. rev.2003) (§ 502(c) estimation "generally should result in an allowed claim for all purposes in the bankruptcy case.").

This court is required to follow the substantive law governing the nature of the claim (such as following contract law when estimating a breach of contract claim). Bittner v. Borne Chemical Co., Inc., 691 F.2d 134, 135-36 (3d Cir.1982). Otherwise, neither the Bankruptcy Code nor the Bankruptcy Rules set forth a procedure for estimating claims; instead, the court may use "whatever method is best suited to the particular contingencies at issue." Id.; see also In re Ralph Lauren Womenswear, Inc., 197 B.R. 771, 775 (Bankr.S.D.N.Y.1996) ("Neither the Code nor the Rules prescribe any method for estimating a claim, and it is therefore committed to the reasonable discretion of the court, which should employ whatever method is best suited to the circumstances of the case."). "There is no question that the Court has discretion to determine the appropriate method of estimation, especially the purpose of the estimation." Trident Shipworks, 247 B.R. at 514 (further noting that estimation is a core matter).

Estimation of a claim "does not require that a bankruptcy court be clairvoyant." Matter of Federal Press Co., 116 B.R. 650, 653 (Bankr.N.D.Ind.1989) (quoting In re Baldwin-United Corp., 55 B.R. 885, 898 (Bankr.S.D.Ohio 1985)). Instead, this court "only needs to reasonably estimate the probable value of the claim." Federal Press, 116 B.R. at 653. "Such an estimate ``necessarily implies no certainty' and ``is not a finding or fixing of an exact amount. It is merely the court's best estimate for the purpose of permitting the case to go forward. . . ." Id. (quoting Baldwin). In some cases parties have requested courts to estimate claims by assigning a present value to the probability that the claimants would be successful in an action in another court (i.e., allow claim in amount of 40% if only 40% of evidence supports the claim). Bittner, 691 F.2d at 136-37. In Bittner, the court of appeals held that the bankruptcy court did not abuse its discretion by estimating claims according to their ultimate merits and assigning a zero value to those claims where it seemed more probable than not that the claims would ultimately fail in another forum. Id. Myriad other alternatives for estimating claims exist. Federal Press, 116 B.R. at 653.

In this unique procedural setting, the court's determination has several limitations. As already noted, the court is estimating the Antitrust Claims solely for feasibility purposes. Moreover, because post-effective date feasibility inherently depends upon future circumstances the court must to some extent predict those circumstances — an uncertain process. In addition, the court is put in the position of using an abbreviated mini-trial to predict what a future judge or jury might conclude from the evidence presented at a full antitrust trial. That task is made more complicated because Objectors have acknowledged that application of their legal theories to the unique facts of this case go beyond the reported cases.

In other words, the court is forced to make some predictions. Such predictions, however, will be limited.

The court will not attempt more than a very general prediction of future market design. Objectors have presented evidence of pending proposals, including the "Comprehensive Market Design Proposal" referred to as "MD-02" proposed by the California Independent System Operator Corporation ("ISO") and a Notice of Proposed Rulemaking entitled "Remedying Undue Discrimination through Open Access Transmission Service and Standard Electricity Market Design" ("SMD") issued on July 31, 2002 by the Federal Energy Regulatory Commission ("FERC"). The court understands from these proposals the general direction of market reforms, but it would be pointless to try to predict exact contours.

In addition, although other matters pending before FERC could overlap with estimation issues, the court will not predict how FERC might rule in those other proceedings for several reasons. First, those proceedings do not include the Antitrust Claims directly, and only indirectly might affect them by reducing Objectors' damages if Objectors prevail. Second, the parties devoted little of their presentations to those matters, emphasizing the antitrust issues the court confronts here. Third, there is a paradox presented in that PG & E vigorously opposes the Antitrust Claims here while (presumably) vigorously opposing Objectors at FERC. Thus, while PG & E argues that a victory for Objectors at FERC will reduce the Antitrust Claims, the fact is that it seems to be doing all in its power to see that what happens at FERC does not reduce the Antitrust Claims.

As for Objectors' legal theories, on the one hand the basic elements of a Section 2 claim are clear. Objectors must establish that PG & E (1) possessed monopoly power in the relevant market, (2) wilfully acquired or maintained that power through exclusionary conduct, and (3) caused antitrust injury. City of Vernon v. Southern California Edison Company, 955 F.2d 1361, 1365 (9th Cir.1992), cert. denied, 506 U.S. 908, 113 S.Ct. 305, 121 L.Ed.2d 228; Metronet Svcs. Corp. v. U.S. West Communications, 329 F.3d 986 (9th Cir.2003).

On the other hand, Objectors' focus is not on these basic elements but on the "essential facilities" doctrine, which applies in a narrower set of circumstances. The Ninth Circuit Court of Appeals has described it, generally, as imposing liability when one firm, which controls an essential facility, denies a second firm reasonable access to a service the second firm must have to compete with the first. Alaska Airlines, Inc. v. United Airlines, Inc., 948 F.2d 536, 542 (9th Cir.1991), cert. denied, 503 U.S. 977, 112 S.Ct. 1603, 118 L.Ed.2d 316 (1992). To a lesser extent Objectors rely on a version of the "price squeeze" doctrine, which is also applicable in a narrow set of circumstances.

III. Issues

The issues to be considered by the court in order to estimate the Antitrust Claims for Plan feasibility are as follows:

1. Does PG & E exercise monopoly power and are Objectors and PG & E competitors?

2. Does PG & E control an essential facility?

3. Has PG & E illegally refused access to an essential facility?

4. Has PG & E orchestrated an illegal price squeeze?

5. Have Objectors established other grounds for their Antitrust Claims?

6. Does the "filed rate doctrine" bar the Antitrust Claims?

7. Does the "state action doctrine" bar the Antitrust Claims?

8. Does the "Noerr-Pennington doctrine" bar the Antitrust Claims?

9. Has PG & E established valid business justifications for its conduct?

10. Are Objectors' damages calculations too speculative to support a damage claim?

IV. Discussion^[6]

A. Summary

PG & E and Objectors are competitors. The relevant market is the market for local distribution of electricity in PG & E's northern California territory.

There are limited sources to supply that market. Generating capacity near Objectors is expensive, and increasing the amount of local generation is generally impractical. The alternative is to import cheaper power, but Objectors and other potential competitors of PG & E cannot do that because transmission capacity is limited. The transmission infrastructure, owned by PG & E, is an essential facility.

The effects of limited transmission capacity are coming to a head because the market for electricity is changing. Some of the high cost of local generation will be shifted from PG & E's customers (who have paid that cost as part of PG & E's general rate base) to Objectors or, if they prove their claims, to PG & E.

Objectors' essential facilities claim fails because they have not established that PG & E has denied access to its transmission system nor made the costs of such access enough to drive Objectors from the market. Objectors' price squeeze claim fails because they have not shown any differential in prices that would squeeze them out of competition, and at least until a new market structure is determined it is not clear that there will be any regulatory gap at all.

Apart from the essential facilities and price squeeze doctrines, Objectors do not directly argue a monopolization claim. Thus, Objectors have not persuaded the court that in the future a judge or jury in an antitrust case would rule against PG & E as to liability.

If, however, that future a judge or jury were to reach a different conclusion, the court believes they would reject most of PG & E's affirmative defenses. The court gives no weight to PG & E's filed-rate and state-action defenses, and little weight to PG & E's Noerr-Pennington defense. That leaves PG & E's business justifications for its actions.

PG & E's business justifications stand or fall on whether its reliance on expensive local generation, rather than transmission, was justified by what it calls "least-cost planning." Least-cost planning generally means planning designed to result in the least overall cost to customers. If PG & E did engage in least-cost planning then it has justified its level of investment in transmission, even if transmission congestion later results in charges that Objectors must pay. In that event, PG & E would also be justified in terminating the interconnection agreements and taking other steps to assure that it would not have to pay those charges. If, on the other hand, PG & E caused the problem, its attempts to shift the costs to Objectors are not justified.

PG & E has not met its burden of proof on this issue. Its allegations that it engaged in least-cost planning are insufficient to overcome Objectors' evidence that PG & E intentionally cut back on its investments in transmission infrastructure, that PG & E's investment in transmission proved to be inadequate, and that PG & E had the motive to under-invest in transmission.

To the extent the future judge or jury might find liability, the court must consider damages. The court is persuaded that Objectors might have to pay substantial congestion charges. Ordinarily the court would discount that possibility to some present value, but there are simply too many ways in which Objectors's damages might be reduced or eliminated.

Unless and until a market structure unfavorable to Objectors is adopted and fully phased in, the amount of congestion charges is unknown. To an uncertain extent, congestion charges are likely to be offset by "congestion revenue rights" ("CRRs") or similar credits. Any damages from congestion charges would have to be reduced by Objectors' savings from not having paid for upgraded transmission (which they would have been required to pay under, for example, the Stanislaus Commitments). Finally, Objectors' projection of damages nearly a half-century into the future is too speculative.

B. Background

1. Objectors

NCPA is a California joint power agency whose members provide local electric distribution services in their relevant geographic areas. NCPA was formed in 1968, and its present members are the cities of Alameda, Biggs, Gridley, Healdsburg, Lodi, Lompoc, Palo Alto, Redding, Roseville, Santa Clara ("SVP"), and Ukiah, together with the Port of Oakland, the Turlock Irrigation District, and the Truckee Donner Public Utility District. NCPA members pool their resources to obtain electricity from the Western Area Power Administration ("WAPA") and other sources of generation, and to construct and operate generation facilities to supplement their purchases.

Objectors' major source of purchased electricity is that generated by the United States of America at Shasta Dam and other Central Valley Project ("CVP") facilities, and sold to them under long term contracts with WAPA. In 1983-85, NCPA began augmenting its WAPA-purchased sources by constructing two geothermal generating plants in the Geysers area of Sonoma County ("Geysers"). NCPA then constructed a hydroelectric facility (the "Calaveras Project") on the Stanislaus River watershed in the Sierra Nevada consisting of two dams, tunnels, and a power plant at Collierville, California. As a 22.92 percent participatory owner in the Calaveras Project, Palo Alto invested over $137 million in its construction; total construction costs were therefore about $600 million. NCPA also increased its generation capacity through the construction of five gas turbine generation units in Alameda, Roseville and Lodi in 1986, and the construction of a steam-injected gas turbine unit in Lodi in 1996.

In addition to investing over $1 billion in the construction of its Geysers, Calaveras Project, and gas-turbine generation facilities, NCPA's members now obtain additional electricity by means of their participation and investments in another joint power agency known as Transmission Agency of Northern California ("TANC"). In 1993, TANC's 25 members completed construction of a 340-mile long high voltage transmission line (the "COTP line") between Southern Oregon and Tracy, California that allows NCPA's members to import additional electricity purchased from Seattle City Light, Bonneville Power Administration, and other generation facilities in the Pacific Northwest. As a 4.032 percent participatory owner of the COTP line, Palo Alto invested another $17 million in its construction; total construction costs were in excess of $420 million.

2. PG & E

PG & E is an investor-owned utility that is vertically integrated.^[7] It owns and operates generation facilities and an extensive transmission network, and is the provider of local distribution services to over 4.6 million residential and business customers in its Northern California service territory.^[8] Under the Plan filed by PG & E, PG & E's generation business and assets will be placed in an entity referred to as "Gen," and its electric transmission business and assets will be placed in an entity referred to as "E-Trans." Both of these entities will be wholly-owned by PG & E's parent company, PG & E Corporation. The Plan further provides that the PG & E's local distribution business and assets will be placed in a separate entity referred to as "Disco" or "Reorganized PG & E." This local distribution entity will then be spun-off from PG & E Corporation by means of a distribution of its new capital stock to the shareholders of PG & E Corporation. Under the CPUC Plan these three discrete business units of PG & E will not be disaggregated, but instead would remain under the ownership and control of PG & E Corporation.

3. Palo Alto's Power

Palo Alto municipalized^[9] its electric distribution in 1900, constructed a power plant, and, over time, supplemented the plant's output by purchases from PG & E. In 1948, Palo Alto had no choice but to purchase all of its needed electricity from PG & E. In 1964, when the CVP was completed, Palo Alto dropped PG & E in favor of purchasing all its needs from the United States Bureau of Reclamation and, later, WAPA, under long-term contracts.

Approximately 80% of the energy needed for Palo Alto's current load is purchased from the federal government through WAPA. The power is cheap: around 2001 the price for Palo Alto was $22.21 per MWh. Other Objectors also rely on significant quantities of cheap power purchased from the federal government. This is due to a federal policy, at least historically, of favoring sales to Muni's over sales to investor-owned utilities.

By the mid-1980's, Palo Alto's needs for electricity were beginning to approach its maximum allotment under its contract with WAPA. Palo Alto therefore joined with other NCPA members in the construction of the Calaveras Project, which gave it access to additional electricity. Palo Alto purchases the largest portion of its total electricity needs from WAPA.

4. Transmission

Pursuant to the Stanislaus Commitments (described in detail below), PG & E provides transmission services to Objectors, linking their sources of purchased and generated electricity with their municipally owned local distribution networks. Objectors are dependent upon PG & E to transmit such electricity economically to it under those obligations.

As a part of Geysers, NCPA constructed transmission lines from that facility to interconnection points with PG & E's transmission system at Lakeview and Fulton, California, whence the electricity generated at the Geysers is transmitted by PG & E to points of interconnection with the local distribution facilities of NCPA's various members. As part of the Calaveras Project, Objectors constructed transmission lines from the Collierville generation plant to an interconnection point with PG & E's transmission system at Bellota, California, whence the electricity generated at Collierville is transmitted by PG & E to points of interconnection with the local distribution facilities of NCPA's various members. PG & E's transmission system likewise interconnects with Objectors' gas turbine generation facilities in Roseville, Lodi, and Alameda. WAPA's transmission lines, which tap the generation plants of the CVP, terminate and interconnect with PG & E's transmission system at Tracy, California, as does TANC's COTP line, which taps generation sources in the Pacific Northwest.

PG & E transports WAPA and COTP-delivered electricity west from Tracy, over the Altamont Pass and across San Francisco Bay, to a location in Palo Alto known as the Colorado substation. At this substation, PG & E's transmission system interconnects with Palo Alto's local distribution network, which Palo Alto uses to transmit the electricity to each of its citizen customers.

Palo Alto obtains the remainder of the electricity it needs from the Calaveras Project. This electricity is first transmitted over NCPA's 40-mile transmission line from Collierville to a point of interconnection with PG & E's transmission system at PG & E's Bellota substation, east of Stockton, California. From Bellota this NCPA-generated electricity is transmitted by PG & E west to Palo Alto's Colorado substation.

There are no transmission lines running west from Tracy to Palo Alto, or west from Bellota to Palo Alto, other than those owned and operated by PG & E. Palo Alto is completely dependent upon PG & E to transmit all of its electricity to Palo Alto's Colorado substation for distribution from that point. Other members of NCPA are likewise dependent upon PG & E to transmit the electricity they need, over at least a portion of PG & E's transmission lines, from the points of generation or interconnection to PG & E's system.

For a number of years, Palo Alto has been asking PG & E to allow Palo Alto to finance an upgrade of the PG & E transmission line from PG & E's Ravenswood substation to Palo Alto's Colorado substation so that Palo Alto might enjoy significant economic benefits attributable to that upgrade. NCPA has made overtures to PG & E regarding a possible sale by PG & E of a load ratio share of its transmission system at a negotiated fair value. It has also suggested obtaining firm transmission rights and to structure such a transaction so that PG & E will not suffer averse tax consequences. Those overtures have been rejected by PG & E.

5. Congestion

Transmission congestion arises when there is insufficient capacity in the transmission system to allow the generation resources with the lowest operating costs to serve demand throughout the grid. Transmission congestion is found in virtually every transmission system as a result of legitimate economic planning decisions. It is generally uneconomic to build sufficient transmission to handle every load even at its peak — by definition much of the transmission capacity would be unused except when the load peaks.

It might be possible to build or upgrade local generating facilities to run efficiently even at peak loads, but again that may be uneconomic because most of this increased capital investment would be unused except when the load peaks. Therefore, the lowest overall cost to the utility's customers is often served by tolerating some level of transmission congestion (supplemented by some level of expensive, but brief, local generation). This is an example of legitimate least-cost-planning.^[10]

PG & E alleges it has engaged in least-cost planning. On some occasions, PG & E's transmission lines into and within the GBA are congested, meaning that these lines do not have the capacity to transmit all the lower cost power requirements of NCPA's GBA members plus all the lower cost power requirements of PG & E's own retail customers in the GBA. PG & E has chosen to address this congestion by operating its less efficient gas-fired generation plants within the GBA at outputs higher than normal, thereby incurring the incrementally higher fuel costs and other expenses associated with these plants, rather than constructing additional transmission capacity to solve the problem or curtailing its own deliveries of electricity to its own GBA retail customers. Focusing on the San Francisco peninsula, PG & E claims it was justified in relying on generation at old and somewhat inefficient local generating plants, rather than upgrading the overall transmission capacity into the GBA and, in addition, upgrading the transmission capacity within the GBA (across San Francisco Bay and up the peninsula).

Objectors disagree. They believe that, at least prior to deregulation, PG & E intentionally built too little transmission capacity so it could maintain a monopoly on local distribution.

The parties' disagreements about congestion are complicated by their use of slightly different terminology. For purposes of this Memorandum Decision "net congestion costs" or "charges" will mean the difference in price between lower cost, remote power and more expensive (but geographically nearer) power. The court recognizes that some net congestion costs are inevitable in an efficient system. The court distinguishes these costs from the "congestion charges" or "costs" to be levied on Objectors or others under MD-02 or whatever market system is eventually adopted, which costs may or may not bear any relation to actual net congestion costs.

6. The Stanislaus Commitments

By letter dated April 30, 1976, PG & E submitted to the United States Department of Justice ("DOJ") a "statement of commitments" in connection with PG & E's efforts to license the Stanislaus Nuclear Project. Those commitments, which later became part of the license conditions for PG & E's Diablo Canyon Nuclear Power Plant Units 1 and 2 ("Diablo Canyon"), have become known as the "Stanislaus Commitments." The Stanislaus Commitments were entered into after DOJ concluded that an anti-competitive situation existed in Northern California. DOJ dropped its antitrust investigation of PG & E in return for PG & E's agreement to include those commitments as part of its federal license for the operation of Diablo Canyon. The Stanislaus Commitments were designed to address certain antitrust concerns of DOJ and to provide for open and non-discriminatory access to PG & E's transmission system by "neighboring entities," as that term is defined in the Stanislaus Commitments.

Paragraph VII(A) of the Stanislaus Commitments refers to "transmission services" and provides, in relevant part, that PG & E "shall transmit power pursuant to interconnection agreements, with provisions which are appropriate to the requested transaction . . . such service shall be provided (1) between two or among more than two Neighboring Entities. . . ." PG & E must wheel all the electricity required by Objectors to meet the demands of the customers served by them at all times, i.e., "Firm Power," as defined in Definition G of the commitments. PG & E has repeatedly acknowledged its obligation to provide transmission services to Objectors under these commitments, although it has also pointed out some qualifications to that obligation.

Paragraph VII(B) of the Stanislaus Commitments provides, in relevant part, that PG & E "shall include in its planning and construction programs such increases in transmission capacity or such additional transmission facilities as may be required for the transactions referred to in paragraph A . . . provided any Neighboring Entity . . . gives [PG & E] sufficient advance notice . . . and provided further that the entity requesting transmission services compensates [PG & E] for the Costs incurred as a result of the request." (Emphasis added.) It further provides that PG & E shall provide such transmission "pursuant to interconnection agreements which . . . are consistent with these license conditions." However, Paragraph VII(C) of the Stanislaus Commitments further provides (in relevant part) that PG & E shall not be required to construct additional transmission facilities if "construction of such facilities is inconsistent with Good Utility Practice [discussed later in this Memorandum Decision]. . . ." Finally, paragraph VII(D) of the Stanislaus Commitments provides, in relevant part, that "[r]ate schedules and agreements for transmission services . . . shall be filed by [PG & E] with the regulatory agency having jurisdiction over such rates and agreements."^[11]

NCPA has asserted that it is a "Neighboring Entity" and has standing to enforce the Stanislaus Commitments as a third-party beneficiary. Interconnection agreements govern the relationship between PG & E and wholesale transmission customers connected to its transmission system. The Stanislaus Commitments do not set forth any specific terms or conditions constituting a transmission contract between NCPA or SVP and PG & E. As set forth in the Offering Circular for NCPA's 1981 Series A Public Power Revenue Bonds:

[w]hile the Stanislaus Commitments provide in general terms, that PG & E will transmit power from the Project lines to NCPA Member participants . . . these Commitments do not, of themselves, create contractual relations or set out the obligations of PG & E in the detail necessary for a complete analysis of costs. Thus, some sort of further agreement with PG & E or order of the [Nuclear Regulatory Commission] or FERC would, in all likelihood, be necessary before the relationship between NCPA and PG & E can be considered to be stable or assured in detail.

Beginning in 1983, PG & E provided transmission and scheduling services to NCPA and SVP pursuant to separate interconnection agreements that set forth the terms and conditions upon which service would be provided (the "1983 IAs" or individually "IA"). The 1983 IAs were filed with and approved by FERC. Consistent with the Stanislaus Commitments, PG & E acknowledged in writing that "It was intended that the IA be consistent with the ``Stanislaus Commitments' which were made by PG & E as part of the licensing process of the Stanislaus Nuclear Project in 1976." In connection with a federal court action brought by the Nuclear Regulatory Commission ("NRC") against PG & E to enforce the Stanislaus Commitments, PG & E and NCPA entered into an additional agreement (the "1991 Settlement Agreement") under which PG & E agreed that its obligations under the Stanislaus Commitments (as set forth in Attachment 1 to the 1991 Settlement Agreement) "shall extend for so long as the Commitments are included in any federal license held by PG & E, but in any event shall not be extinguished prior to January 1, 2050." Attachment 1 to the 1991 Settlement Agreement sets forth the parties' rights and obligations in the event of termination of the 1983 IA or any successor IA.

Under the 1991 Settlement Agreement the Stanislaus Commitments became contractual obligations of PG & E owed directly to Objectors as parties to that agreement, rather than simply as third-party beneficiaries of the letter agreement between the DOJ and PG & E.

In 1991, PG & E and NCPA also entered into an amended interconnection agreement (the "1991 IA"). It provided that PG & E was entitled to seek an increase in transmission rates from FERC pursuant to Section 205 of the Federal Power Act. In particular, Section 8.2 of the 1991 IA provided that after January 1, 1998, PG & E could unilaterally apply to FERC for a change in rates, which was defined to include "all rates, charges, classifications, rate principles, rate methodology, accounting principles and practice."

In a sense, all of these contracts — the 1983 IAs, the 1991 IA, and the Stanislaus Commitments — are not critical to this estimation proceeding because, although the parties disagree whether PG & E has breached them, they give rise to contractual obligations rather than create the Antitrust Claims. Nevertheless, Objectors claim that PG & E's disregard for these agreements is part of its illegal, anti-competitive conduct.

In addition, the parties' dispute whether future congestion charges will be costs that Objectors must pay under the Stanislaus Commitments or would have had to pay under the IAs (before PG & E terminated the 1991 IA). For now the court simply notes that the Stanislaus Commitments state: "``Costs' means all capital expenditures, administrative, general, operation and maintenance expenses, taxes, depreciation and costs of capital, including a fair and reasonable return of [PG & E's] investment, which are properly allocable to the particular service or transaction as determined by the regulatory authority having jurisdiction over the particular service or transaction." The definitions of "Costs" in the 1983 IAs and 1991 IA are not materially different from the definition of Costs in the Stanislaus Commitments.

In accordance with the provisions of the Stanislaus Commitments, and the implementing provisions of the 1983 IAs and 1991 IA, Objectors provided PG & E with annual and other periodic forecasts of their needs for firm transmission services, and paid PG & E its defined costs of providing such services in two ways: (a) by means of a transmission access charge per megawatt of electricity transmitted by PG & E, representing each NCPA member's aliquot share of PG & E's defined costs of providing transmission generally; and (b) by means of discrete payments (or self-funding) in those instances in which transmission facilities were necessary for the specific but peculiar needs of NCPA, as distinguished from the needs of all customers. Examples of the latter were the costs paid by NCPA to PG & E to interconnect the Geysers with PG & E's transmission system, and NCPA's construction of the 40-mile transmission line from its Collierville generation facility.

PG & E did not, however, construct all such additional facilities as were specified in Section VII-B of the Stanislaus Commitments. Instead, as already noted, PG & E relied on local generating plants in what it alleges was a legitimate exercise of least-cost planning. PG & E claims:

Prior to CAISO operations, PG & E had a practice of least cost planning of transmission and generation where strategically located generation was used to support the reliability of the transmission system. The costs associated with this method of least cost planning (i.e., the use of generation to support transmission system reliability) were recovered in incrementally high fuel costs for out-of-merit order dispatch of generation when needed for transmission system support. The incrementally higher fuel costs were recovered from all entities that purchased power from PG & E, not only PG & E's retail load. [Emphasis added.]

Put differently, Objectors have not paid net congestion costs. Rather, those costs have been included in PG & E's rate base and paid by all of PG & E's customers.

The Stanislaus Commitments and the IA's contemplate passing along to Objectors the costs of upgrading transmission infrastructure but the parties disagree whether they contemplate passing along net congestion costs where PG & E has elected to rely on local generation rather than upgrading transmission. This is part of the parties' contractual disagreement over the term "Costs," which the court does not address. The essential fact is that Objectors historically have not paid net congestion costs.

7. Deregulation

In the mid-1990s, state and federal authorities took steps to restructure the electric industry in an effort to open the wholesale and retail electric markets to greater competition. CPUC set forth its proposed restructuring of the California markets in its Preferred Policy Decision No. 95-12-063 (1995) as modified by Dec. No. 96-01-009 (1996), Rulemaking No. 94-04-031 (1994), Investigation No. 94-04-032 (1994); 1996 Cal. PUC Lexis 28, 166 P.U.R.4th 1, 1996 WL 47921 ("Preferred Policy Decision"). In 1996, the California Legislature enacted Assembly Bill 1890 ("AB 1890"), which restructured the California electric industry by unbundling transmission, generation and distribution services. AB 1890 generally codifies the market structure proposed in the Preferred Policy Decision.

AB 1890 required investor-owned utilities to transfer operational control of their transmission facilities to the newly created ISO, an independent, non-profit entity charged with managing the transmission grid. Under the new structure, PG & E would act as a "Scheduling Coordinator" for pre-existing customers, such as Objectors, who were not in contractual privity with ISO. As Scheduling Coordinator, PG & E would be responsible for submitting and adjusting energy forecasts for Objectors. Under deregulation, PG & E would also act as a Transmission Owner ("TO") pursuant to a TO Tariff it would file with FERC, under which PG & E would receive payments from ISO in exchange for use of PG & E's transmission facilities.

As noted above, Objectors historically have not paid net congestion charges. In fact, prior to creation of ISO and implementation of the ISO Tariff net congestion costs were not separately calculated nor were they charged to PG & E's wholesale transmission customers. Instead, PG & E was allowed to recover these costs as part of wholesale and retail energy rates. This is changing under deregulation.

In 1996, FERC issued Order 888, 61 Fed.Reg. 21,540, 1996 WL 239633 (May 10, 1996) ("Order 888"), which required integrated utilities to: (1) file open-access transmission tariffs assuring non-discriminatory access to the grid; (2) unbundle generation and transmission to allow greater transparency of rates; and (3) consider the creation of an independent system operator. In 1997, FERC approved the ISO Tariff for California and effective March 31, 1998 ISO created a number of new categories of charges. These included new charges to Scheduling Coordinators (such as PG & E) for ancillary services, reliability services, imbalance energy and grid management. In addition, the ISO Tariff created a new category of congestion charges to reflect the costs associated with serving load during periods when the transmission system was constrained. By this change, net congestion costs, previously absorbed primarily by PG & E's retail customers (and to a significantly lesser extent by wholesale energy customers), became unbundled as congestion costs that were charged to PG & E. If PG & E does not pass those costs along to its customers it must absorb them.

The current ISO Tariff includes two categories of congestion charges: inter-zonal and intra-zonal congestion. Congestion that occurs between the three large contiguous geographic congestion zones within California is called inter-zonal congestion. Inter-zonal congestion charges are imposed when a Scheduling Coordinator transmits power across a congested inter-zonal interface. Intra-zonal congestion refers to congestion within a zone, and the regulations and charges for such congestion are still evolving.

As a result of the imposition of the ISO Tariff, PG & E became the Scheduling Coordinator for Objectors and counter-parties to other ETCs, and incurred certain ISO charges associated with serving that ETC load. These changes raised concerns within PG & E that PG & E's role as "middleman" under the IAs with Objectors would cause PG & E to incur charges without a means to obtain reimbursement from Objectors. As a result of these concerns, PG & E sought termination of the existing IAs and their replacement with agreements under which Objectors would receive service directly from ISO. After termination of the IAs, NCPA and SVP became their own Scheduling Coordinators and became subject to the costs imposed under the ISO Tariff, including congestion charges.

8. Termination Of Interconnection Agreement

On July 21, 1997, PG & E gave notice to NCPA of its intent to terminate the 1991 IA effective July 31, 2000, in accordance with the notice provisions of the agreement. The effective date of termination was later extended to March 31, 2002.^[12] PG & E and Objectors thereafter engaged in extensive negotiations in an effort to agree upon a new structure that would allow Objectors to obtain transmission service directly from ISO.

With the assistance of FERC staff they reached agreement as to virtually all unresolved operational issues after engaging in extensive negotiations in the period between May though July 2002. As a result, Objectors and PG & E entered into a settlement agreement and replacement interconnection agreements, and Objectors entered into separate agreements with ISO under which Objectors obtained services directly from ISO. In Comments filed with FERC in support of the settlement, NCPA advised FERC that "the settlement package effectively resolves many of the operational issues associated with moving forward into a new relationship with [ISO], and transitioning away from a primary relationship with PG & E, a transition that both NCPA and PG & E prefer." NCPA further requested that FERC undertake to resolve the remaining "basic dispute" between PG & E and Objectors, "the issue of who is responsible for congestion costs."

In the August 30 Order, FERC approved the settlement agreement and exercised its jurisdiction to determine the remaining issue of "transmission service rights, and the right to be exempted from congestion charges under the Stanislaus Commitments. . . ." FERC appointed an administrative law judge to conduct proceedings to determine this issue and a schedule for discovery and hearing has been established.

Objectors and PG & E are currently litigating before FERC the issue of whether the Stanislaus Commitments exempt Objectors from congestion charges. Objectors' damage claim is therefore also dependent to some extent upon the assumption that they will not prevail at FERC and that FERC will determine that Objectors are subject to congestion charges. A favorable recovery by Objectors on these contract-based claims (which the parties have not included as part of the Estimation Stipulation and are not before this court) will reduce any liability of PG & E on the Antitrust Claims.

Although Objectors are now technically subject to the ISO congestion charges imposed by the current ISO Tariff, those changes have proven to be small. In fact, Objectors have not identified any congestion charges they have paid to ISO or will pay in the future under the existing ISO Tariff. Objectors' Antitrust Claims are based on an assumption that in the near future ISO will implement a new type of congestion charge, described below.

9. Sale Of Generation Units

In 1995, PG & E owned and operated eight fossil generation plants: Humboldt Bay, Morro Bay, Moss Landing, Oakland, Contra Costa, Pittsburg, Potrero and Hunters Point.

In its Preferred Policy Decision, CPUC stated "that, at a minimum, it was necessary to disaggregate the vertically integrated electric utility by separating the elements of generation, transmission and distribution" and affirmed its proposal that "the utilities transfer the operational control of all transmission facilities to an [Independent System Operator]."

In addressing the issue of "Concentration of Generating Facility Ownership or Control," CPUC observed that:

market power problems almost certainly will require the existing investor-owned utilities to divest themselves of a substantial portion of their generating assets, particularly their fossil generating plants located within their service territory. Therefore, we will require PG & E and SCE [Southern California Edison] to file within 90 days of the effective date of this order a plan to voluntarily divest themselves through a spinoff or outright sale to a nonaffiliated entity of at least 50% of their fossil generating assets.

Preferred Policy Decision, 1996 Cal. PUC Lexis 28 at Part 2, *34 (footnote omitted).

The Preferred Policy Decision further states:

[t]o provide an incentive for the utilities to voluntarily divest these assets, we will tie the utility's allowed rate of return on the equity component of the non-nuclear and non-hydroelectric equity component of its transition cost CTC balancing accounts. We will grant an increase in the rate of return for the equity component of up to 10 basis points for each 10% of fossil generating capacity divested.

Preferred Policy Decision, 1996 Cal. PUC Lexis 28 at Part 2, *35.

CPUC ordered a plan for "voluntary" divestiture of 50% of fossil generation assets; in addition, CPUC provided substantial economic incentives for PG & E to divest its remaining fossil generation plants, including tying the permissible rate of return on PG & E's equity to the amount of generation capacity divested.

Ultimately, PG & E divested all of its fossil generation plants with the exceptions of Humboldt Bay and Hunters Point. Facilities at Morro Bay, Moss Landing and Oakland were sold by auction in the fall of 1997 ("Wave One"), and the sale of remaining facilities was approved in 1998 ("Wave Two").

Although PG & E's divestitures in Wave Two exceeded CPUC's requirement of "voluntary" divestiture of 50% of fossil generation, the Wave Two divestitures were a voluntary business decision because of the economic risks of a reduced rate of return on equity if PG & E chose to hold the remaining generation assets. In addition, PG & E was required by CPUC to market value its generating assets by December 31, 2001, by appraisal, sale or other divestiture, and the auction process met this requirement. PG & E's divestitures of its fossil generation facilities were approved by CPUC.

PG & E divested its Bay Area power plants without making any arrangements that would have enabled PG & E to continue to provide congestion-free transmission service to NCPA/PA under the Stanislaus Commitments, such as entering into "vesting contracts" that would have given PG & E an option to purchase power from the divested plant at a guaranteed price.

PG & E had been using its own gas-fired generation plants in the GBA to provide "cover" electricity to Objectors during times when congestion in PG & E's transmission system prevented PG & E from transmitting all of Objectors' electricity. PG & E's decision to sell those plants necessarily meant that if PG & E were to continue providing cover power, it would have to purchase that power at market rates from the new owners of those plants, rather than providing such cover power to Objectors at the incremental fuel costs previously borne by PG & E.

10. Current Regulatory Situation

On May 1, 2002, ISO filed its MD-02 proposal for a new market design. On July 31, 2002, FERC issued its SMD. Both proposals use a new pricing model for transmission known as Locational Marginal Pricing ("LMP"). ISO currently measures congestion charges based on the transmission of power across three large geographic zones. Under LMP, as proposed by both MD-02 and SMD, congestion charges would be measured using smaller zones, perhaps as many as several thousand "nodes" in the transmission grid. This will create "price signals" for the cost of additional increments of power at each location. In theory, those price signals will act as an incentive to more efficient use of the transmission system and ensure that customers demanding energy over congested lines bear the costs associated with that consumption.

There has not been a determination whether congestion will be charged on such a disaggregated basis or whether ISO will aggregate these individual nodes in some way — what ISO refers to as the level of "granularity." Originally, from the summary in MD-02, it was clear that ISO's intent had been to move to a finer level of granularity as soon as technically feasible:

. . . ISO proposes to require loads to be scheduled and settled initially at a level of geographic granularity at least as fine as today's demand zones. The requirement would shift to the finer load group level as soon as technically feasible, with allowance for loads to select the nodal level or a custom aggregation.

ISO has modified this approach. Its January 10, 2003 status report to FERC proposes that, after calculating individual load nodes, ISO would (at least initially) aggregate load into four relatively large geographical areas as opposed to a greater number of smaller areas.

Under this proposal, the level of aggregation would be similar to the current level of congestion aggregation, so that the costs of congestion to Objectors would be spread over a large customer base in northern California. As noted, under the present three-zone system, Objectors do not incur any significant congestion charges. If ISO's latest proposal were adopted, and if it were not phased-out, then Objectors' damage claims would be virtually eliminated.

The method of calculating congestion charges also has not been determined. Under the current system, the congestion costs associated with re-dispatched generation are charged only for the incremental power obtained from the geographically closer, more expensive generation source. Both SMD and MD-02, however, contemplate the use of higher "excess congestion rents" as an incentive to more efficient use of the transmission system and to ensure that customers demanding energy over congested lines bear the costs associated with that consumption.

Under SMD and MD-02, congestion charges would be assessed upon all of the power flowing across a congested line based upon the higher marginal costs of re-dispatched power. In other words, the gross amount of excess congestion rents imposed by SMD and MD-02 could be far greater than the actual net congestion cost of obtaining out-of-merit local generation.^[13]

Excess congestion rents are, however, only part of the process of calculating the final, net congestion costs under SMD and MD-02. Both the SMD and MD-02 provide that the costs of excess congestion rents would be offset through the allocation of CRRs, formerly known as firm transmission rights ("FTRs"). CRRs act as a "hedge" against congestion risks by providing a credit against congestion charges; essentially, a megawatt of CRRs charged in the day-ahead market would fully cover the congestion charges for a megawatt of power along a congested transmission path.

MD-02's Introduction to FTRs showed that, at least when that document was prepared, ISO and FERC intended to phase-out pre-existing rights for ETCs:

FERC's recent Options Paper on the Standard Market Design expresses clear concern about incompatibilities between ETCs and the LMP approach, and supports the objective of eventually treating all grid users according to a common Open Access Transmission Tariff.

Nevertheless, MD-02 would initially allocate CRRs to historic users of out-of-merit generation, such as Objectors. The May 1, 2002 version of MD-02 allocated CRRs to: (1) ETC holders who voluntarily convert to a CRR system; (2) load-serving entities based on historic use; and (3) buyers who purchase the balance of transmission capacity through an ISO-conducted auction. Similarly, the most recent drafts of the SMD and MD-02 indicate that Objectors would be entitled to receive CRRs based upon their rights under certain existing contracts as ETCs and based upon their historic use of the system as load-serving entities.^[14] In addition, under the current ISO proposal revenues from the initial auctions of CRRs would themselves go to historic users of the system. Once again, if these proposals are adopted, and if they are not phased out, then Objectors' damage claims would be virtually eliminated.

Despite the foregoing initial protections, the court is convinced that Objectors bear a substantial risk the protections will be phased out. The latest proposal by ISO rejects any allocation of CRRs longer than three years. It also suggests that issues involving granularity and CRRs will be revisited in future. The court concludes that, unless LMP is effectively abandoned, Objectors are likely to incur some significant level of congestion charges in future. Ultimately, however, the amount of congestion charges and CRRs is unknown.

C. Analysis Of Legal Issues

As noted above, Objectors rely principally on the essential facilities doctrine, and to a lesser extent on the price squeeze doctrine or a variant thereof. Before turning to these doctrines, the court will consider whether PG & E is a monopolist, whether Objectors and PG & E are competitors, and whether PG & E controls an essential facility.

1. PG & E exercises monopoly power, and competes with Objectors, in a defined relevant market.

"Monopoly power, commonly referred to as market power, is defined as ``the power to control prices and exclude competition.'" Metronet, 329 F.3d at 1003 (citations omitted). The Ninth Circuit has instructed that in determining whether monopoly power exists,

The key question is whether existing competitors and immediate potential entrants have sufficient capacity to take business away from the incumbent monopolist and thereby constrain the incumbent's ability to raise prices above competitive levels.

Metronet, 329 F.3d at 1006 (citations omitted). In particular, to establish that PG & E has monopoly power Objectors must:

"(1) define the relevant market, (2) show that [PG & E] owns a dominant share of that market, and (3) show that there are significant barriers to entry and . . . that existing competitors lack the capacity to increase their output in the short run."

Id. at 1003 (quoting Rebel Oil, Inc. v. Atl. Richfield Co., 51 F.3d 1421, 1434 (9th Cir.1995)).

The court agrees with Objectors that the relevant market for purposes of Section 2 is the market for the distribution of electricity to residential and business customers in PG & E's Northern California service territory. The court excludes SMUD's service territory, but would reach the same conclusions if that territory were included. Under this definition, the market is discreet because generation and transmission are not substitutes for local distribution, and there are no other close substitutes for the local distribution of electricity. Defining the market in this manner also makes sense because PG & E and Objectors are direct competitors in that market, as discussed below.

PG & E owns a dominant share of the market. As of December 31, 2001, PG & E provided local distribution of electricity to over 4.6 million customers in Northern California, and all other entities providing such service (excluding SMUD) served approximately 440,000 customers. PG & E's share of the relevant market therefore exceeded 90 percent. Even if SMUD's approximately 522,000 customers were included in the relevant market, PG & E's share of the relevant market would exceed 80 percent. By either measure, PG & E has a dominant share of the relevant local distribution market.

Objectors have also offered persuasive evidence that there are significant barriers to entry and that existing competitors lack the capacity to increase their output in the short run (or, for that matter, the long run). Metronet, 329 F.3d at 1003. First, as long as transmission capacity is constrained there is no way for a competitor to offer more imports or different sources of imported electric power. There is simply too much congestion in the transmission lines to do so, particularly into the GBA and (of particular concern in Palo Alto's geographic area) across San Francisco Bay and up the peninsula. In addition, as discussed further in the court's essential facility analysis, it would be impractical if not impossible for Palo Alto (or anyone else) to duplicate PG & E's transmission lines.

Second, Objectors have offered persuasive evidence that it is too expensive and impractical to build local generating plants in Palo Alto, or in other parts of the GBA that would relieve congestion. PG & E has suggested no other means by which existing or potential competitors could compete in the local market for distribution in the "short run," as Objectors must show, or even in the "long run." Metronet, 329 F.3d at 1003. Therefore, PG & E exercises monopoly power in the relevant market.^[15]

Both PG & E's monopoly power and its direct competition with Objectors is illustrated by a simple example of a business deciding whether to locate within the geographic boundaries of Palo Alto (where Palo Alto is generally the sole distributor) or next to Palo Alto in areas where PG & E is generally the sole distributor. If Palo Alto can obtain transmission of cheap power from WAPA then it can offer cheap power to that business (the "New Customer"). All other things being equal, the New Customer might be more likely to locate in Palo Alto than in a neighboring city, where power is more expensive.^[16]

One consequence of the New Customer locating in Palo Alto might be to increase Palo Alto's tax base. In addition, to the extent Palo Alto does not pass along all the savings from cheap WAPA power to the New Customer, it can collect the remaining profit margin, which will be larger than PG & E can maintain, because PG & E does not have as high a percentage of cheap sources of power.

If, on the other hand, Palo Alto cannot obtain sufficient power from WAPA then it cannot offer the New Customer firm cheap power (unless Palo Alto reduces its profit margins on sales to other customers). All other things being equal, that might persuade the New Customer to locate next door to Palo Alto in territory where PG & E has a monopoly on local distribution. That would increase PG & E's revenues from local distribution (and perhaps other services) and decrease Palo Alto's revenues from local distribution.

Objectors are competitors with PG & E in another important respect. Retail customers in PG & E's Northern California service territory (and who do not reside in the service territory of SMUD) have only two choices for the provision of local distribution services. One is to obtain those services from PG & E, and the other is self-provision by means of municipalization. Municipalization has been a threat to PG & E for a long time, in that it reduces the size and scope of PG & E's activities.

Muni's provide competition and a competitive threat to PG & E's monopoly position in the relevant market for local distribution. They provide important "benchmarking" or "yardstick" competition to PG & E, as a comparison of their rates and service quality to those of PG & E are matters that voters may consider in deciding upon "municipalization" measures. Such was the case most recently in the City and County of San Francisco, where proponents of the municipalization measure on the November, 2002 ballot drew attention to the rates of five California Muni's — Palo Alto, Alameda, Santa Clara, SMUD, and the Los Angeles DWP — as showing that "Public Power is Cheaper, Much Cheaper." PG & E responded by arguing such things as "Takeover is costly" and "an idea whose time has passed"; "Takeover means more government bureaucracy"; and "rates may be higher and service may be lower with a municipal utility."

PG & E has been concerned with the threat of a municipal "takeover" in San Francisco; it hired consultants and used dedicated teams of both company employees and PG & E retirees to promote the message that sticking with PG & E was better than having "the bureaucrats at City Hall running your electricity system."

Additional evidence of competition is that the public policy of California recognizes and encourages competition between actual and potential Muni's and PG & E. In AB 1890, the California Legislature included a provision that positively encourages and promotes such direct competition. This proviso, codified as Public Utilities Code § 9601(c), specifically grants reciprocal rights to a Muni to compete to serve customers served by PG & E, and to PG & E to compete to serve customers served by that Muni. Pursuant to this statute, Palo Alto and PG & E entered into a written Reciprocity Agreement, dated July 17, 2000, agreeing to the billing procedures and other details for "electric power sales made by [Palo Alto] to customers in PG & E's service territory," and reciprocal "electric power sales made by PG & E to customers in [Palo Alto's] service territory." The temporary suspension of such arrangements by CPUC during the "power crisis" of 2001 does not in any way diminish the long-term public policy of direct competition between the Muni's and PG & E.

Finally, in considering whether Objectors and PG & E are competitors for purposes of Section 2, the Second Circuit's analysis in City of Groton v. Connecticut Light & Power Co., 662 F.2d 921 (2nd Cir.1981), is compelling here, and settles the question for this court. There the district court found that the Muni's before it were not in competition with the defendant utility, Connecticut Light & Power Co. ("CL & P"). The court of appeals rejected that finding, as follows:

The district court expressly found that the municipalities were not in competition with CL & P. Though the court made this finding only in reference to the price-squeeze claims, [City of Groton v. Connecticut Light & Power Co.,] 497 F.Supp. [1040] at 1055-56 [(D.Conn.1980)], its opinion clearly indicates that it thought the municipalities were purchasing power solely as customers, not as competitors. It is inherently difficult to define competition in the electric-power industry; the best definition, we believe, at least for purposes of this case, is one by the Federal Energy Regulatory Commission. In Connecticut Light & Power Co., 31 Pub. U. Rep. 4th 315, 320-22[, 0079 WL 167679] (Aug. 20, 1979), the Commission obtained guidance from two cases. The first, United States v. El Paso Natural Gas Co., 376 U.S. 651, 659-61, 84 S.Ct. 1044, 1048-49, 12 L.Ed.2d 12 (1964), states:

This is not a field where merchants are in a continuous daily struggle to hold old customers and to win new ones over from their rivals. . . . the competition then is for the new increments of demand that may emerge with an expanding population and with an expanding industrial or household use of gas. . . . The presence of two or more suppliers gives buyers a choice. (Emphasis omitted.)

The second case, Borough of Ellwood City v. Pennsylvania Power Co., 462 F.Supp. 1343, 1346 (W.D.Pa.1979) states:

For practical purposes, competition between Penn Power and plaintiffs can be seen most strongly in the service of industrial and commercial customers having the option to locate in either the service area of Penn Power or that of plaintiffs. These customers do have a choice of suppliers when making their initial decision to locate their operations. . . . Plaintiffs and Penn Power also compete, at least theoretically and on a long term basis, for service areas. If plaintiffs were to become unable to serve their customers profitably, Penn Power would logically be in the best position to assume plaintiffs' present service.

The Commission thus viewed the essential characteristic of competition in the electric-power industry as being "that there are or could be alternate suppliers of the same product in the same geographic area," 31 Pub. U. Rep. 4th at 321, and further held as to the utilities involved here that "it is sufficient if it is demonstrated that a wholesale customer and the filing utility are in geographic proximity and that the wholesale customer is or could be an alternative supplier of electricity to some of the customers presently served by the company or that the company could be an alternate supplier for customers presently served by the wholesale customer." Id. The Commission also noted that the utility and the wholesale customer "could be alternate suppliers to new customers who may choose to locate in the relevant geographic area." Id. The Commission divided competition into three categories: competition for individual customers, including large industrial or commercial loads; franchise competition, for the right to serve all of the customers in a given territory, usually for a specific period of time (see Otter Tail Power Co.); and fringe area competition, for customers on the fringes of the present service areas of the rival utilities. See Conway Corp. v. FPC, 510 F.2d 1264, 1268 (D.C.Cir.1975), aff'd, 426 U.S. 271, 96 S.Ct. 1999, 48 L.Ed.2d 626 (1976); Meeks, [Concentration in the Electric Power Industry: The Impact of Antitrust Policy, 72 Colum.L.Rev. 64 (1972)], at 81-100. It is true that the Commission's decision that these parties were competitors was solely a determination that there was a prima facie case of a "price squeeze," whereas the district judge has, after hearing all the evidence, made findings concerning the absence of competition. Nevertheless, under the Commission's definition of competition, which we find persuasive both on its face and in the light of the Commission's expertise with respect to the electric industry, and which we here adopt, the district court's general findings of no competition cannot stand.

City of Groton, 662 F.2d at 930.^[17]

In sum, PG & E has monopoly power in the relevant market; Objectors and other Muni's are competitors of PG & E in that market; and would-be Muni's are potential competitors of PG & E in that market. These actual and potential competitors, and the threat of municipalization, are part of the competitive process in the relevant market. The court is convinced that Objectors and other existing and potential local distributors do not have sufficient capacity to take business away from PG & E and thereby constrain PG & E's ability to raise prices above competitive levels.

2. PG & E controls an essential facility.

In assessing the feasibility of PG & E's Plan, the court must consider what will happen if that Plan is confirmed and becomes effective, meaning PG & E will be disaggregated. Therefore, one might think that the court should consider whether E-Trans, as the future owner of the transmission system, will own and control an essential facility. The parties have not approached the issue this way, and nor will the court. The reason is that Objectors base their damages claims on acts or omissions that have already occurred or are now occurring, while PG & E is a vertically integrated utility. Therefore, the court will consider whether PG & E, not E-Trans, controls an essential facility.

One principal characteristic of an essential facility is that it truly must be essential:

[A]s the word "essential" indicates, a plaintiff must show more than inconvenience, or even some economic loss; he must show that an alternative to the facility is not feasible.

Alaska Airlines, 948 F.2d at 544 (9th Cir.1991) (quoting Twin Laboratories, Inc. v. Weider Health & Fitness, 900 F.2d 566, 570 (2d Cir.1990)).

PG & E contends that its transmission system is not an "essential" or "bottleneck" facility because the Muni's could construct local generation facilities to eliminate their dependence upon PG & E's system. That is not feasible.

California will not allow a competing electric transmission system to be built. It is not reasonable to assume PG & E's transmission system can be duplicated. In particular, that is true through those portions of PG & E's transmission system that are used to wheel power from points of interconnection (such as Lakeview, Bellota, and Tracy) to the respective local distribution systems of NCPA's members. Such transmission is essential for Objectors to obtain power feasibly.

Even if California were to allow a competing transmission system to be built, that would not be a feasible alternative to using PG & E's transmission system. Objectors have evaluated the cost and feasibility of constructing their own transmission system and have determined that, both economically and politically, that alternative is impossible. This is because of significant environmental and feasibility problems, including the possibility that the line might have to be "submarined" beneath federally protected wildlife marshlands in the southern part of San Francisco Bay.

As for building more local generation, there are considerable hurdles of siting and other matters. Even if those hurdles could be overcome, it would cost Objectors far more to substitute their own local generation than it would to continue to obtain transmission from PG & E, even with significant congestion charges added to those costs.^[18]

Therefore, neither the construction of new, duplicative transmission lines by Objectors nor the construction of new local generation plants is a practical and economically feasible alternative to Objectors' use of PG & E's transmission system. Objectors must have use of that transmission system to continue to obtain the low-cost electricity that Objectors now purchase from WAPA and other sources, and that Objectors generate at Geysers and the Calaveras Project facilities. In other words, to maintain competition in the market for distribution it is essential that Objectors be able to use PG & E's transmission system.

PG & E argues, however, that it no longer controls the transmission system. ISO does. Although PG & E mostly raises this argument to show that it could not possibly deny access to the transmission system (which the court will address below), the degree of PG & E's control is critical to determining whether a facility is "essential" in the first place:

A facility that is controlled by a single firm will be considered "essential" only if control of the facility carries with it the power to eliminate competition in the downstream market.

Alaska Airlines, 948 F.2d at 544 (emphasis in original, footnote omitted).^[19]

The short answer is that although ISO operates the transmission system PG & E controls all the aspects cited by Objectors: how much to invest in transmission (i.e., PG & E's reliance on costly local generation to supplement an allegedly deficient transmission system), whether to designate the Stanislaus Commitments and IAs as ETCs, whether to terminate the IAs, and so on. In other words, PG & E has control.

In addition, as the existence of the Stanislaus Commitments attests, PG & E's control of its transmission system gives it sufficient power (if not held in check) to put competitors at a disadvantage and discourage them from remaining in the business — to eliminate competition. See Metronet, 329 F.3d at 1012 (discouraging plaintiff from staying in relevant market was sufficient to state essential facilities claim).

In sum, PG & E's transmission system is an "essential" or "bottleneck" facility within the meaning of Section 2 jurisprudence.

3. Objectors have not shown that PG & E has refused access to an essential facility in violation of Section 2.

An essential facility claim requires the plaintiff to prove (1) that the defendant was a monopolist in control of an essential facility (as Objectors have done here); (2) that plaintiff, as a competitor, could not reasonably or practically duplicate the facility (again, as shown here); (3) that defendant has refused plaintiff access to the facility; and (4) that it is feasible for defendant to provide such access. City of Anaheim v. So. Cal. Edison Co., 955 F.2d 1373, 1380 (9th Cir.1992); Metronet, 329 F.3d at 1010.^[20]

As determined above, the essential facility is PG & E's transmission system. To find a violation of Section 2 under Otter Tail Power Co. v. United States, 410 U.S. 366, 93 S.Ct. 1022, 35 L.Ed.2d 359 (1973) and its progeny, Objectors would have to show a denial of access to that system by PG & E. They have not done so, for indeed, PG & E wheels all the power necessary to keep all of the local transmission and distribution systems of Objectors (and other members of NCPA) operating.

In addition, as PG & E points out, ISO controls the operation of its transmission system. Objectors have not alleged that ISO has discriminated in providing access to that system.

Instead of a denial of access, Objectors paint a picture of economic burdens in the future if and when they must shoulder the congestion charges now handled by PG & E and absorbed throughout its entire rate base. In doing so they run squarely up against Alaska Airlines, 948 F.2d 536, and are not saved by the recent Ninth Circuit decision in Metronet, 329 F.3d 986.

Alaska Airlines describes the "essential facilities" doctrine as imposing liability when one firm, which controls an essential facility, denies a second firm reasonable access to a service the second firm must have to compete with the first. 948 F.2d at 542. It then labels Otter Tail an extreme case: ". . . this refusal did more than merely impose some handicap on potential competitors; it eliminated all possibility of competition in the downstream market." Id. at 543.

Objectors' argument depends, therefore, on whether they can fit through the narrow window left open by Alaska Airlines: "whether at some level, charging a price may be the same as an outright refusal to deal." Id. at 545, n. 13. Metronet answers that question.

In Metronet, the plaintiff was required by the defendant to accept what was described as "per location pricing" for telephone systems (line access and calling features) it bought and resold to small businesses. It sued on three counts under Section 2, including denial of access to an essential facility.

In discussing denial of access, the Metronet court cited decisions from other circuits standing for the proposition that absolute denial of access need not be shown, as unreasonable terms and conditions of access, such as in rates charged, may result in a practical denial of access. Metronet, 329 F.3d at 1012. Then, citing Alaska Airlines, the court shaped the contours of such unreasonable terms, conditions and rates: "However, providing access at a fee that is not so high as to drive away competition does not amount to a denial of access." Id., citing Alaska Airlines, 948 F.2d at 545-46. Applying that standard to the facts of the case before it, the court required the plaintiff to show more than a decrease in profitability. It stated that a decrease in profitability must be significant enough to discourage plaintiff from staying in business: "In other words, [plaintiff] must show that per location pricing made the [phone system] resale business unprofitable, or squeezed the profit margin to the point where the game was no longer worth the candle." Metronet, 329 F.3d at 1012.

Here the court recognizes that Muni's are not for-profit enterprises, but the message is the same. Under Alaska Airlines and Metronet, in order to make a case for the economic equivalent of a denial of an essential facility there must be a showing of such significant harm that would make Objectors' operation of their municipal electricity distribution systems "no longer worth the candle." Objectors have not met that heavy burden.

Objectors' damages are too speculative (as discussed below) to show that they will be driven out of the market for local distribution. In addition, Objectors' existing access to cheap WAPA power gives them substantial potential profit margins, and because they have not disclosed the economics of their operations the court does not know whether, even with substantial congestion charges, they would be so damaged economically that they will be effectively denied access.

For all of these reasons, Objectors have not established the third element of their essential facilities claim. They have not shown that PG & E denied them access to its essential transmission system.^[21]

4. PG & E has not engaged in an illegal "price squeeze."

Reduced to its simplest terms, a price squeeze for Section 2 purposes occurs when a monopolist "games" two regulatory systems in such a manner as to damage its competitor. As noted by the court in City of Mishawaka:

The term "price squeeze" is used in this context refers to a situation where the monopolist charges its wholesale customer a wholesale rate high enough to impede that customer's competition with the monopolist in the retail market.

616 F.2d at 979, n. 4.

As a general matter wholesale rates under FERC control go into effect automatically without approval while retail rates must await CPUC approval. While under certain circumstances there can be a short-term delay of a new wholesale rate at FERC, there is sometimes a much more cumbersome process at the state level with no certainty as to time limits. This gives rise to a great possibility for abuse:

Behind the rate applications there are differing regulatory procedures, differing tests and standards to be applied, and differing accounting principles to be used in the computations. At best a utility may find itself in a legal and practical maze, but for price squeezing the dual system also offers an obvious, ready made illegal opportunity with a legitimate gloss.

City of Mishawaka, 616 F.2d at 983-84.

A hypothetical illustrates what can happen. Suppose a supplier of energy files a wholesale rate with FERC that goes into effect immediately and which is higher than the price at which a downstream competitor can sell the same power to its retail customers. As long as the wholesale rate is higher than the retail rate, the squeeze is on. Only sometime later, when the retail seller obtains an increased retail rate from the state agency, or obtains a reduction from FERC in the wholesale rate, leading to a refund, is the squeeze released. In the meantime, the retail competitor has been harmed significantly.

The problem for Objectors in the present case is that to squeeze the retailer requires two tongs and one of them is missing. Of course Objectors' rates are what they are, and to change them may be time consuming and cumbersome. But FERC, while approving the termination of the IAs, has not taken any action to establish a rate that PG & E could use to squeeze Objectors.

Objectors generally allege that the price squeeze doctrine can be applied more broadly, to what they call the alleged regulatory lacuna in this case. They have not, however, developed or proved that theory sufficiently.

In the Ninth Circuit the vice of the price squeeze has been said to be that it can cause severe damage to competitors by unjustifiably raising their cost of doing business. City of Anaheim, 955 F.2d at 1376. Contrasting such conduct in ordinary commercial transactions with similar behavior by a public utility, the court stated:

Those concerns are attenuated in the electrical industry whose rates are regulated at both the wholesale and retail levels. Nevertheless, because the regulatory systems do not work in perfect harmony, it is possible for a utility to manipulate its filings and requests in a manner that causes a, at least temporary, squeeze which might be just as effective as one perpetrated by an unregulated actor.

Id. at 1377 (citing John E. Lopatka, The Electric Utility Price Squeeze As An Antitrust Cause Of Action, 31 UCLA L.Rev. 563 (1984)).

The court went on to note that although the price squeeze theory had not been applied in this circuit to the electrical industry, it had been applied elsewhere. After discussing various approaches, the court adopted the approach taken in City of Mishawaka, requiring something more than general intent to establish a violation of the Sherman Act and requiring the trial court to discern from a consideration of all of the evidence of the utility's activities, not only a general intent but a specific utility intent to serve its monopolistic purposes at municipal expense. Id. (citing City of Mishawaka, 616 F.2d at 985). More particularly, the court stated:

We agree with the district court and with Mishawaka II that the requirement of a specific intent is an appropriate way to erect a dike which is sufficient to prevent an untoward invasion of the land of legal monopolies by the sea of antitrust law. Of course, in so holding we emphasis that the specific intent need not be proved by direct admissions of wrongdoing. Rather, the actions of the utility, taken as a whole, can and should be considered.

Id. at 1378.

The court made this analysis after first suggesting that it is not proper to focus on specific individual acts of the monopolist without refusing to consider their overall combined effect. It continued:

At the same time, if all we are shown is a number of perfectly legal acts, it becomes much more difficult to find overall wrongdoing. Similarly, a finding of some slight wrongdoing in certain areas need not by itself add up to a violation. We are not dealing with a mathematical equation. We are dealing with what has been called the "synergistic effect" of the mixture of the elements. City of Groton v. Connecticut Light & Power Co., 662 F.2d 921, 929 (2d Cir.1981). Thus, while our discussion will speak to the specific claims, we emphasize that we have also ruminated upon the effect of combining those claims, but the result of that rumination makes no difference in our ultimate conclusion.

955 F.2d at 1375.

Having considered all of the challenged conduct of PG & E as a whole, and having already rejected Objectors' essential facilities theory, the court cannot, consistent with City of Anaheim, apply Objectors' tenuous price squeeze theory and find a violation of Section 2.

5. Objectors have not established other grounds for their Antitrust Claims.

Apart from the essential facilities and price squeeze doctrines, Objectors do not directly argue a monopolization claim. Objectors claim that PG & E denied them access to its transmission system and caused a price squeeze, by acts or omissions that they expect to result in substantial congestion charges. They do not otherwise argue that such acts or omissions constituted "exclusionary conduct" whereby PG & E "wilfully acquired or maintained" its market power. Metronet, 329 F.3d at 1002. Absent such argument, the court believes it would be unfair and unwise to address such a claim.^[22] For the same reason the court has not focused specifically on any of Objectors' purported claims under Cal. Bus. & Prof.Code section 17200 or other state laws.

6. PG & E may not avoid the Antitrust Claims under the filed rate doctrine.

Much of the filed rate doctrine^[23] involves issues of federal preemption, but inasmuch as FERC's scheme of regulation permits parties such as Objectors to enjoy the benefits of ETCs, no issues of federal preemption are present in this case. Further, the filed rate doctrine is inapplicable in suits between competitors. Cost Management, 99 F.3d at 948; City of Groton, 662 F.2d at 929; City of Kirkwood v. Union Elec. Co., 671 F.2d 1173, 1179 (8th Cir.1982), cert. denied, 459 U.S. 1170, 103 S.Ct. 814, 74 L.Ed.2d 1013 (1983). Cf. Barnes v. Arden Mayfair, Inc., 759 F.2d 676 (9th Cir.1985).^[24]

This exception might more properly be called the "multi-jurisdiction" exception. It arises when the relevant rates are not the subject of exclusive regulation by a single regulatory agency. In that setting, a party may be able to place its rivals at a competitive disadvantage because of a gap between regulatory agency jurisdictions, neither agency having the authority to remedy the situation by regulation of all relevant rates in harmony with one another.

Such is the case here, as the FERC's regulatory authority is limited to wholesale transmission rates, be those of PG & E, the ISO, or both. The FERC lacks authority to regulate PG & E's bundled retail rates for electricity, an undifferentiated component of which is PG & E's costs of transmission. CPUC for its part has no authority to regulate wholesale transmission rates, inasmuch as its rate regulation jurisdiction is confined to that of PG & E's bundled retail rates for electricity to its 4.6 million customers in the relevant market.

Therefore, PG & E cannot claim that any federal approval of its rates shields it from liability for any illegal anti-competitive actions. The filed rate doctrine is inapplicable.

7. The State Action defense is not available to PG & E.

PG & E contends that its actions are immunized from liability by the "state action" doctrine, because its actions were undertaken under a "clearly articulated and affirmatively expressed" policy of the State of California, and that such policy is "actively supervised" by that State. This doctrine derives from Parker v. Brown, 317 U.S. 341, 63 S.Ct. 307, 87 L.Ed. 315 (1943), where the Supreme Court ruled that the Sherman Antitrust Act did not restrain actions by governmental officials from carrying out directives of state legislatures. The Court noted that states cannot by legislative act create antitrust immunity (317 U.S. at 351, 63 S.Ct. 307), but that state acts themselves are not unlawful.

PG & E would have the court apply that doctrine to its actions vis-a-vis Objectors on the theory that its conduct is closely supervised by the Sate of California. It relies on California Retail Liquor Dealers Ass'n v. Midcal Aluminum, Inc., 445 U.S. 97, 100 S.Ct. 937, 63 L.Ed.2d 233 (1980). There the Court explained that the state action immunity is available where the challenged restraint is "clearly articulated and affirmatively expressed as State policy" and that policy is "actively supervised by the State itself." Midcal, 445 U.S. at 105, 100 S.Ct. 937 (citing City of Lafayette v. Louisiana Power & Light Co., 435 U.S. 389, 98 S.Ct. 1123, 55 L.Ed.2d 364 (1978)).

While the Preferred Policy Decision, AB 1890, and other aspects of deregulation were — though perhaps are not now — clearly articulated policies of the State of California, the court does not agree with PG & E that the second necessary element of Midcal — active supervision by the state itself — is present. In addition, the policy to which Midcal refers is a state policy expressly authorizing and compelling a party to take action that would otherwise violate the federal antitrust laws. Id., 445 U.S. at 105-06, 100 S.Ct. 937. PG & E has not proven the existence of any policy of the State of California prohibiting competition between PG & E and the Muni's, or any policy expressly authorizing and compelling PG & E to take action aimed at suppressing actual and potential competition with the Muni's, or suppressing or eliminating the competitive process in the relevant market. California's policy, expressed in Cal. Public Utilities Code § 9601(c), is one of promoting competition between the Muni's and PG & E.

The distinction was explained by the Ninth Circuit in Cost Management, 99 F.3d at 942, where the court noted that the State of Washington had displaced competition in the market with a regulatory structure, but examined the relevant question of whether the regulatory structure specifically authorized the alleged unlawful conduct.

Among other things, it also suffices to say that PG & E has not proven that the State of California both compelled and actively supervised (a) PG & E's business decision to sell all of its generating plants in the GBA without seeking vesting contracts from the new owners;^[25] (b) PG & E's business decision to terminate the 1991 IA; (c) PG & E's business decision to refuse to designate the Stanislaus Commitments and the 1991 IA as ETCs; or (d) PG & E's business decision to refuse to consider selling a load-ratio portion of its transmission network to Objectors, particularly in light of PG & E's recognition that such a sale might result in greater competition between the Muni's and PG & E.

Likewise, PG & E has not proven that the State of California compelled PG & E's business decisions regarding its level of investment in transmission, which Objectors allege is responsible for creating the very congestion problems that PG & E is now attempting to use (they claim) to impose congestion charges upon them. CPUC itself pointed out that congestion was greater in PG & E's service territory than in those of the State's two other investor-owned, vertically integrated utilities. In a report by CPUC's Energy Division, in response to a legislative mandate,^[26] dated March 12, 2001 and entitled "Relieving Transmission Constraints, an Overview in Response to AB 970" (the "AB 970 Report"), CPUC recommends 31 projects to reduce congestion and states:

All but 5 of the 31 recommended projects are to reduce or remove the normal overload, stability, RMR, contingency, and economic transmission constraints in PG & E's territory. Unlike SCE and SDG & E, PG & E's capital investment strategy has been to build local generation rather than transmission. PG & E cut back its infrastructure investments during the 1990s and made limited investments in redundant distribution-related facilities. Therefore, many of PG & E's current projects were discussed internally years ago, but not built.

Although PG & E took issue with CPUC's findings, and defended its own business decisions regarding transmission, PG & E did not state or suggest that CPUC had compelled it to act in this way.

8. The Noerr-Pennington defense might be available to PG & E on some issues, but not most.

PG & E asserts that its conduct otherwise in violation of Section 2 is immunized from liability by reason of the "Noerr-Pennington" doctrine.^[27] That doctrine shields from the Sherman Act a constitutionally protected right of petition by way of a concerted effort to influence public officials, regardless of intent or purpose. See City of Mishawaka v. American Electric Power Co., 616 F.2d 976, 981 (7th Cir.1980), cert. denied, 449 U.S. 1096, 101 S.Ct. 892, 66 L.Ed.2d 824 (1981), reh. denied, 450 U.S. 960, 101 S.Ct. 1421, 67 L.Ed.2d 385.^[28]

Objectors point out that this doctrine protects advocacy, not anticompetitive behavior. Objectors allege that PG & E's core dispute is with the public policy of Congress granting Muni's a preference to federally generated electricity, and that what is at issue is not any petition to Congress to change that policy but PG & E's "self-help" to narrow or eliminate the advantages of that federal preference, and to undermine Objectors' considerable investments in generation facilities and the COPT transmission line.

Objectors cite the following passage from Mishawaka:

It appears to us that the municipalities are not barred by Noerr-Pennington in the particular circumstances. Were we to view it otherwise, the federal and state regulatory processes would provide the utility with a method of effectively advancing its illegal monopolistic purposes while maintaining an outward appearance of total innocence and shielded from the Sherman Act.

Mishawaka, 616 F.2d at 982.

Objectors argue that the circumstances here are analogous to those of Mishawaka and that here, as there, the defendant is attempting to use the gap between federal and state regulatory jurisdiction to fasten an anti-competitive result upon its municipal competitors.

PG & E maintains that its actions, or at least some of them, are advocacy protected by Noerr-Pennington. The court recognizes that, for example, not designating the IAs as ETCs could be seen as advocacy to the applicable regulators not to put PG & E in the position of having to pay future congestion costs. On the other hand, that omission could also be seen as an attempt to evade the congestion costs that PG & E itself (allegedly) created.

As discussed below, in connection with PG & E's business justification defense, it is unclear whether PG & E actually is responsible for excessive congestion. Therefore the court cannot entirely determine whether Noerr-Pennington will apply.

Nevertheless, it is clear that most of PG & E's alleged wrongdoing cannot be characterized as an exercise of the constitutionally protected right to petition the government. For example, PG & E's decision how much to invest in transmission was in no way a petition to government; and the fact that PG & E also applied to CPUC for approval of its rates does not shield any under-investment in transmission from potential liability.

The court concludes that some of PG & E's alleged violations of Section 2 could be immunized by the Noerr-Pennington doctrine. Most of them, however, cannot.

9. PG & E might not be able to prove valid business justifications for its conduct.

One of PG & E's principal lines of defense to the Antitrust Claims, and theories upon which it would have the court estimate those claims as minimal or non-existent, is that its challenged conduct is fully protected as a product and outgrowth of its business judgment. See Metronet, 329 F.3d at 1007 (once a prima facie case of exclusionary conduct is shown, burden shifts to defendant to offer procompetitive justification). PG & E has offered business justifications for each of the acts and omissions about which Objectors complain.

a. Transmission Capacity

As a business justification for the lack of greater capacity in its transmission lines, PG & E offers some evidence that it engaged in legitimate least-cost planning. Objectors rely chiefly on the AB 970 Report as evidence that PG & E under-invested in transmission. The court concludes that PG & E has not carried its burden to show that its level of investment in transmission was justified by legitimate business considerations.

(i) PG & E's Evidence

PG & E has shown the following. The transmission planning process involves identifying future transmission needs based upon review of peak demand forecasts; running computer simulations to test the transmission system under a variety of normal and emergency situations; identifying system deficiencies; and developing alternative measures to address the identified deficiencies. The standards for identifying and addressing system deficiencies are provided in industry standards and in the ISO's Grid Planning Standards.

Annually PG & E develops a Five Year Plan identifying transmission problems and describing recommended solutions and alternatives. The format, requirements and ultimate approval of PG & E's Five Year Plans and specific transmission projects are overseen and regulated, to some extent, by the ISO and CPUC.

PG & E historically relies upon a combination of transmission and generation in order to meet its obligation to serve its load. The use of out-of-merit dispatch as a form of least cost planning is a generally accepted practice in the electric utility industry.

Least cost planning and, specifically, the use of out-of-merit dispatch are forms of Good Utility Practice in the management of transmission systems; that is, as defined in the ISO Tariff, they are among the practices, methods, and acts engaged in or approved by a significant portion of the electric utility industry during the relevant time period, or any of the practices, methods, and acts which, in the exercise of reasonable judgment in light of the facts known at the time the decision was made, could have been expected to accomplish the desired result at a reasonable cost consistent with good business practices, reliability, safety, and expedition.

The Stanislaus Commitments provide that PG & E's obligation to include appropriate increases in transmission capacity is subject to the further requirement that "any Neighboring Entity or Neighboring Distribution System [including Objectors] give[ ] Applicant sufficient advance notice as may be necessary to accommodate its requirements from a regulatory and technical standpoint and provided further that the entity requesting transmission services compensates [PG & E] for the Costs incurred as a result of the request."

Further, Objectors had the right to seek approval from the ISO for a particular upgrade or to build an upgrade themselves, subject to provisions of the ISO Tariff that provide for competitive solicitations for construction of facilities that the ISO determined to be necessary to ensure reliability of transmission service.

Objectors have not identified any instance in which (1) they gave notice of a need for increased or additional transmission facilities pursuant to paragraph VII(B) of the Stanislaus Commitments and (2) approved undertaking the proposed project (including the obligation to pay the costs of such upgrade as appropriate) but (3) PG & E failed or refused to proceed with the proposed project.^[29]

(ii) Objectors' Evidence

In support of their claims that PG & E under-invested in transmission, Objectors cite CPUC's AB 970 Report. As noted above, that report recommends 31 upgrades to transmission and mentions the difference between other vertically-integrated utilities' approach and PG & E's lack of investment in transmission.

The report also notes that "over the last three years" (i.e., after deregulation began) PG & E has "doubled its transmission investment." The report devotes much attention to transmission constraints into what it calls the "Bay Area" (which the court will treat as roughly equivalent to the GBA, and which is where NCPA members Palo Alto, Alameda, SVP and the Port of Oakland are located). The report states:

Constraints on imports into the Bay Area contributed to the rolling blackouts there on June 14, 2000, when voltage dropped precipitously at a major Bay Area substation. Ten projects for 2001, including new and upgraded transmission lines, transformers, and capacitors and other equipment, will improve service to and within the Bay Area, and partly relieve the constraints in the Bay Area.[^[30]]

In addition, two new generating plants are scheduled to begin operation in the Bay Area during the summer, adding a total of 545 MW. [Emphasis added; footnotes omitted (identifying specific transmission and generation projects).]

The AB 970 Report identifies five categories of transmission constraints, and in each category there is inadequate transmission capacity for areas where Objectors are located (principally the Bay Area). For example, the report states:

. . . . The most severe RMR ["Reliability-Must-Run"] constraints are in PG & E's territory because it has historically relied heavily on local generation, rather than on strong transmission links. PG & E has plans to further reduce RMR constraints and have [sic] incorporated these plans into its annual electric grid expansion plan.

In addition, several of the projects would clearly generate immediate cost savings. According to the AB 970 Report, among the five projects that PG & E identified as RMR is one involving an upgrade for "under $15 million," affecting the Bay Area, that would eliminate an RMR unit costing "$25 million per year." Two more of the projects identified by PG & E would relieve reliance on RMR generation in the Lodi area, including an upgrade that would cost between $1 million and $5 million to eliminate RMR costs "in the range of $1 million per year." The remaining two projects involved either "[n]ot enough information received from the utility" or "[n]o Project Justification" from the utility.

Objectors' expert Dr. Robert B. Wilson adds:

An indication of the magnitude of the deficiency in transmission investments is the fact that between 1990 and 2001, the miles of high-voltage transmission lines in California expanded by only 8.4% while peak demand grew by 25%. [Footnote omitted.]

Although PG & E takes issue with the AB 970 Report, and points out that it was not the result of an adjudicative process, the report is nevertheless the official report of CPUC as the agency charged by specific legislative mandate with identifying those actions necessary to reduce or remove constraints on the state's existing electrical transmission and distribution system. As such, the report is sufficient evidence to overcome PG & E's proffered business justification that it engaged in legitimate least-cost planning. In addition, the court notes that the report's conclusions about transmission inadequacies are supported by California's history, after PG & E's under-investment in transmission, of price spikes and rolling blackouts.

For the foregoing reasons, the court is persuaded that a future judge or jury in an antitrust case would most likely find that: (1) PG & E had substantial control over its level of investment in transmission and intentionally cut back on such investments in the 1990s, choosing instead to rely on local generation, (2) PG & E's investment in transmission proved to be inadequate, (3) PG & E had the motive to under-invest in transmission (as is clear from the court's discussion of competition between PG & E and Muni's), and (4) in view of these factors, PG & E's explanation of least-cost planning shows only what might be justified, not that PG & E's low level of investment in transmission was in fact justified.

The court emphasizes, however, that the burden of proof for the business justification defense is on PG & E. The court is not called upon to decide whether Objectors' evidence would satisfy their own burden of proof to establish intentional under-investment as an element of some antitrust theory, because Objectors have advanced no such theory.

The court now turns to PG & E's business justifications for its other alleged wrongs. If one assumes, contrary to the foregoing analysis, that PG & E was justified in creating a transmission system with significant congestion, then PG & E's other business justifications are largely persuasive.

b. Terminating the Interconnection Agreements, and not designating the IAs and the Stanislaus Commitments as Existing Transmission Contracts.

PG & E's termination of the IAs was justified (ignoring, for present purposes, any under-investment in transmission). In those circumstances, PG & E was also justified in its decision not to designate the IAs and the Stanislaus Commitments as ETCs.

As noted above, PG & E had turned over operational control of its transmission system to ISO; its retail customers absorbed the vast majority of out-of-merit dispatch costs associated with providing services to Objectors; and upon implementation of the ISO Tariff PG & E acted as Scheduling Coordinator for Objectors and was concerned that its role as "middleman" would subject it to charges with no means of recovering those charges under the IAs. Therefore, PG & E had incentives to terminate the IAs for the protection of its customers and itself.

PG & E had the contractual right to terminate the IAs: Paragraph 9.4 of PG & E's IAs with NCPA and SVP provides that either party may terminate the agreement upon three years' written notice. PG & E therefore gave notice that it would terminate the existing IAs and sought FERC approval of a new arrangement under which Objectors would take service directly from the ISO and become their own Scheduling Coordinator. PG & E has presented evidence that it engaged in negotiations with Objectors in an attempt to work out a consensual resolution, and the court is not persuaded that PG & E did so in anything other than good faith.

PG & E was encouraged to terminate the IAs, and not to designate the IAs or the Stanislaus Commitments as ETCs, by the Preferred Policy Decision and by the agreement under which ISO operated the transmission system. The Preferred Policy Decision disfavored the continuation of individually negotiated transmission agreements and sought to impose costs under a principle of "cost causation" by providing for collection of revenues for "congestion costs arising from the redispatch of the system in the face of transmission constraints." The Preferred Policy Decision also notes that "[c]ost of service regulation [of the type provided under the old IAs] is no longer compatible with the changing electric industry and is in need of reform."

Section 5.1.7 of the Transmission Control Agreement between PG & E and the ISO, which was filed in March 1997 and implemented the transfer of operational control over the transmission grid to the ISO, requires "participating TOs [including PG & E] whose transmission lines and associated facilities are subject to Encumbrances [on the ISO Controlled Grid]" to make "all reasonable efforts to remove or relax Encumbrances in order to permit the operational protocols to be amended in such manner as the ISO may reasonably require." This was another incentive for PG & E to remove encumbrances on the transmission system, including the IAs with Objectors.

That is not to say that PG & E could ignore valid contracts. The Stanislaus Commitments, unlike the IAs, did not allow PG & E to give notice of termination. Therefore, PG & E might remain liable in contract if its actions caused a breach of the Stanislaus Commitments. For antitrust purposes, however, PG & E has established a business justification for not designating the Stanislaus Commitments as ETCs (assuming, again, that PG & E was had not under-invested in transmission). In addition, under the same assumption, PG & E has established a business justification for terminating the IAs.

c. Selling generating facilities without vesting contracts.

PG & E had a legitimate business justification for not seeking to sell its generation facilities subject to "vesting contracts" (ignoring, again, the effects of any under-investment in transmission). Under the terms of the Preferred Policy Decision, PG & E was strongly discouraged (if not prohibited) from doing so.

In the Preferred Policy Decision, CPUC explained that "both the transparency and reliability of the pricing signals will be seriously compromised unless the jurisdictional utilities are obligated to bid their generation units into the [Power] Exchange and procure the electric energy needed to supply their full service customers from it." Bidding all generation into the Power Exchange is inconsistent with vesting contracts to supply a portion of power directly to Objectors. In addition, pursuant to the creation of the California Power Exchange to function as a "clearinghouse by providing a transparent market for generation" through open bidding procedures, CPUC ordered that:

for the five year transition period during which they seek recovery of their stranded generation assets and power purchase liabilities, our investor owned utilities should be required to bid all of their generation into the Power Exchange and satisfy their need for electric energy on behalf of their full service customers with purchases made from the Exchange. [Emphasis added.]

As the emphasized language suggests, however, PG & E's business justification does not quite establish its defense under the State Action Doctrine because saying that utilities "should be" required to take certain actions is not the same as an absolute rule that they must do so. For example, CPUC made an exception when it stated:

Fairness dictates honoring existing QF contracts and other existing wholesale power purchase agreements as we move toward a more competitive market.

Preferred Policy Decision, 1996 Cal. PUC Lexis 28 at part 2, p. *74.

PG & E has not shown that CPUC could not have been persuaded that fairness would dictate making another exception for vesting contracts.

In sum, the court is convinced that (ignoring any responsibility for under-investment in transmission) PG & E has shown a business justification for not selling its generating facilities subject to vesting contracts.

d. Not selling Objectors a portion of PG & E's transmission system.

The court is persuaded by PG & E's argument that any business has a legitimate interest in maintaining its existence, and that it was justified in not acceding to Objectors' demands to sell some of its transmission facilities. Although one antitrust remedy might be to compel such a sale, unless and until such remedy is ordered the court doubts there are circumstances in which PG & E would not be justified in refusing to sell some of its transmission system.

10. Objectors' damages are far too speculative to support a damage claim for feasibility purposes.

PG & E points out that Objectors have yet to pay any congestion charges. As for future charges, PG & E argues that damages are likely to be minimal because, under current market design proposals, (1) LMP may be implemented using four relatively large geographical areas rather than thousands of nodes (thus spreading congestion costs over many customers), and (2) Objectors are likely to receive substantial CRRs based on historic usage, largely offsetting congestion charges.

While this is true, it ignores the likelihood that LMP will be phased-in and CRRs will be phased-out. This historically has been the intention of ISO and FERC, as reflected in the SMD, MD-02 and even the latest proposals.

In addition, there are limits to what FERC and CPUC can do unless transmission is upgraded: they can allocate who pays for the increased costs from lack of transmission, but they cannot solve the underlying problem of lack of transmission capacity. Sooner or later it seems very likely that the regulators will refuse to make PG & E's customers pay for congestion that is not in their geographic area. In other words, regardless whether the current system is a subsidy to Objectors (as PG & E argues), or a subsidy to PG & E (as Objectors argue), or something else, it is likely to end.

Nevertheless, there is some evidence in the AB 970 Report (and elsewhere) that PG & E is already upgrading at least some parts of its transmission system. In addition, if PG & E disaggregates (as its plan provides) then there has been no showing why E-Trans would not have every incentive to upgrade transmission to appropriate levels (although, as a close affiliate of Disco, it conceivably would still have incentives not to). In other words, by the time LMP is fully phased-in and CRRs are phased-out the transmission system might have been partially or even fully upgraded.

All of these changes in the regulatory landscape and the transmission system itself make damages highly uncertain. In some circumstances the damages could be large, but they could also be minimal or non-existent.

Objectors' evidence provides no basis to overcome this inherent uncertainty. Their damage claims are based on the assumption that an LMP system will be fully implemented, either without CRRs or, if CRRs are included, without any allocated to Objectors. CRRs have been omitted entirely from Objectors' damage analysis.^[31]

Moreover, the court cannot ignore the fact that Objectors would normally be obligated to pay for their share of upgrades to the transmission system. Any damages they might suffer would have to be reduced to account for their savings in not having had to make these payments. Los Angeles Memorial Coliseum Com'n v. Nat'l Football League, 791 F.2d 1356, 1370-76 (9th Cir.1986) (setoff applied in antitrust case), cert. denied, 484 U.S. 826, 108 S.Ct. 92, 98 L.Ed.2d 53 (1987). The court cannot tell how much savings may be involved, so this makes damages even more uncertain.

Finally, Objectors' projection of damages nearly a half-century into the future is too speculative. Although this is not a trial to establish liability, and proof of damages for feasibility purposes may be less stringent, a projection that far into the future is unwarranted. Not only could the transmission system be upgraded decades before 2050, but new technologies could evolve. Also, events in the regulatory and physical worlds are bound to supersede any damages estimate that far into the future.

For all of these reasons, even if Objectors had established liability, they have not established any meaningful measure of damages. This is an additional reason for the court's estimation of the Antitrust Claims at zero.

V. Estimation

The court has carefully considered all of the issues discussed in the foregoing section of this Memorandum Decision. It has resolved some of those issues in favor of PG & E and some in favor of Objectors. But the estimation process is not a mathematical tally of pluses and minuses for each side in order to see which side wins. Rather, it is an analysis of all of the factors, and as noted in Part II, supra, the court's best estimate of the outcome, albeit in this somewhat artificial setting. The more outcome-determinative factors must be weighed more heavily in favor of the party for whom they support. Less important factors weigh less favorably for their proponent.

Based on the factors presented the court believes that Objectors have not made a case for denial of an essential facility, price squeeze, or more generalized anti-competitive behavior by PG & E. The case might very well end there. Nevertheless, since there is a possibility that the court has undervalued the strength of Objectors' case, it cannot ignore those defenses asserted by PG & E that appear to have merit, nor can it ignore the serious doubts it has about Objectors' damages proof. This leads to the conclusion that, even if the court were called upon to estimate Objectors' Antitrust Claims for voting and/or allowance and distribution purposes — which it has not been asked to do — the estimation would be de minimis, if not zero. Adding the fact that the estimation is for an even more attenuated and remote purpose, in accordance with the Estimation Stipulation, the court is compelled to exercise its discretion and make an estimation of the Antitrust Claims as having no value for purposes of Plan feasibility.

VI. Disposition

The court is concurrently issuing an order estimating Objectors' Antitrust Claims as having no value for purposes of feasibility of the Plan.

NOTES

[1] By separate order issued concurrently with this Memorandum Decision the court is setting forth its rulings on those objections.

[2] All other chapter, section and rule references are to the Bankruptcy Code, 11 U.S.C. §§ 101-1330 and to the Federal Rules of Bankruptcy Procedure, Rules XXXX-XXXX, unless otherwise indicated.

[3] Throughout this Memorandum Decision, except where the context clearly indicates otherwise, the term "Plan" refers both to PG & E's Plan and the CPUC Plan, including later amendments. Among other things, the CPUC Plan has been amended to be a joint plan of CPUC and the Official Committee of Unsecured Creditors.

[4] "Confirmation of the plan is not likely to be followed by the liquidation, or the need for further financial reorganization, of the debtor or any successor to the debtor under the plan, unless such liquidation or reorganization is proposed in the plan." 11 U.S.C. § 1129(a)(11).

[5] PG & E still needs to worry about those claims in the future because, as noted, the court's estimation has no other effect beyond Plan feasibility, the Antitrust Claims have not actually been tried, and nothing in the court's discussion could prevent a United States district court in the future from reaching an entirely different result than that reached by this court in this estimation.

[6] The following discussion constitutes the court's findings of fact and conclusions of law. Fed. R. Bankr.P. 7052(a).

[7] The only other vertically-integrated electric utility in PG & E's Northern California service territory is the Sacramento Municipal Electricity District ("SMUD"), a publicly-owned entity that serves approximately 522,000 customers in the greater Sacramento area.

[8] The business of electricity consists of activities in three adjacent markets: (a) the generation of electricity at power plants in which turbine-generators are powered by various energy sources, including water ("hydro"), natural gas or other fuels, nuclear reactors or steam produced by the heat of subterranean magma ("geothermal"); (b) the long distance transmission of electricity, by means of high-voltage transmission lines and associated equipment, from power plants to local communities; and (c) the local (or "retail") distribution of electricity to individual customers in each community.

[9] The process by which voters decide to have public agencies — municipalities, irrigation districts, and rural electrification districts and the like (collectively, "Muni's") — provide local electrical distribution services to their citizen customers is commonly called "municipalization." Typically, the municipalization process involves voter approval of a ballot measure, followed by the acquisition of appropriate local distribution facilities by construction, condemnation or both, and the subsequent operation of those facilities by the Muni.

[10] Put differently, least-cost planning involves a departure from the protocol of "merit dispatch" (that is, transmitting power from the cheapest available source), and instead involves obtaining power from more costly, local generation sources ("out of merit dispatch" or "generation re-dispatch"), thereby relieving congestion on long-distance transmission lines. While this practice raises the short-term costs of providing power locally when needed, out-of-merit dispatch costs can represent a total lower-cost solution than building more costly transmission facilities. This economic trade-off can lower costs for transmission customers (such as Objectors) who would otherwise have been required to share the costs of building additional transmission facilities.

[11] By its Order Accepting Settlement Agreement and Interconnection Agreements, 100 F.E.R.C. ¶ 61,233, 2002 WL 31986988 (Aug. 30, 2002) ("August 30 Order"), FERC has determined that the transmission portions of the Stanislaus Commitments are within its jurisdiction and has required PG & E to file certain sections of them with FERC.

[12] PG & E filed written notice of intent to terminate the SVP IA on November 15, 2001.

[13] Substantially all of the damages projected by Objectors' damage expert are based on this new form of congestion costs, which, therefore, assumes a change in ISO and FERC policy.

[14] Objectors concede that if they are allocated sufficient CRRs, they will be substantially insulated from any costs associated with excess congestion rents, thereby substantially mitigating their damage claims.

[15] The existence of a few Muni's does not show that there are insignificant barriers to entry into the market for local distribution. PG & E grew to acquire the monopoly power it has today, and PG & E has not shown that the same factors which allowed Muni's to develop historically would be true today. See Metronet, 329 F.3d at 1005 ("[t]he fact that entry has occurred does not necessarily preclude the existence of ``significant entry barriers.'") (citation omitted).

[16] Customers in PG & E's Northern California service territory do not obtain local distribution services from any of California's other three vertically-integrated electric utilities — Southern California Edison Company, The Department of Water and Power of the City of Los Angeles, or San Diego Gas and Electric Company.

[17] The disputes between CL & P and the Muni's in City of Groton were more involved than the refusal to wheel, a denial of an essential facility and price squeeze presented to this court in the estimation proceedings, but there were similar theories advanced by the plaintiffs there.

[18] By "significant" congestion charges the court means significant in relation to the costs of increasing local generation capacity, not the several billion dollars in damages claimed by Objectors.

[19] In footnote 11 the Ninth Circuit alludes to a second condition that probably must be satisfied, viz., that the power to eliminate competition must be "at least relatively permanent." Id. at n. 11 (citations omitted). The court considers PG & E's power over its transmission system sufficiently "permanent," notwithstanding ISO's role, for the reasons discussed below.

[20] The Ninth Circuit has pointed out "that the second element is effectively part of the definition of what is an essential facility in the first place," and that "the fourth element basically raises the familiar question of whether there is a legitimate business justification for the refusal to provide the facility. . . ." City of Anaheim, 955 F.2d at 1380.

[21] Objectors arguably have not established the fourth element of their essential facilities claim: that it would be feasible for PG & E to provide the access they demand. Objectors have demanded "firm" transmission, but by definition PG & E could only provide such transmission at the expense of its other customers. CPUC generally disapproves of such favored treatment, as shown by its suspension of the Reciprocity Agreement between PG & E and Palo Alto. Therefore, it is not clear that PG & E would be permitted to provide Objectors with greater access than it did, at least under any new agreement. That begs the question whether PG & E could have assured such firm transmission under an existing agreement, such as by designating the IAs as ETCs. The parties have not addressed this issue as part of their essential facilities analysis, and the court does not decide it.

[22] The parties have argued whether PG & E has shown a business justification for its conduct, which may be the other side of the coin. The court will address that issue below, after considering PG & E's other affirmative defenses.

[23] See Cost Management Svcs., Inc. v. Washington Natural Gas Co., 99 F.3d 937, 943 n. 7 (9th Cir.1996) (discussing use of term "filed rate doctrine").

[24] Even though the court has rejected Objectors' price squeeze theory, the filed rate doctrine would be unavailable to counter that theory of antitrust liability. City of Kirkwood, 671 F.2d 1173, 1179.

[25] The court recognizes that CPUC disfavored vesting (or any other mechanisms that would inhibit the transition to a more free-market approach). Nevertheless, PG & E has not shown that CPUC either had a flat rule against vesting or, in any particular instance, compelled PG & E to sell its generating facilities without vesting, regardless of any effects that might have had on Objectors. The court will return to this issue in connection with PG & E's business justification defense.

[26] On September 7, 2000, California AB 970 was filed with the Secretary of State. Section 7 of AB 970, Section 399.15, states:

. . . within 180 days of the effective date of this section, [CPUC], in consultation with the Independent System Operator, shall take all of the following actions . . . :

(a)(1) Identify and undertake those actions necessary to reduce or remove constraints on the state's existing electrical transmission and distribution system, including [reinforcement of existing transmission capacity and other specific actions]. The commission shall, in consultation with the Independent System Operator, give first priority to those geographical regions where congestion reduces or impedes electrical transmission and supply.

2000 Cal. Legis. Serv. Ch. 329 (A.B.970) (West).

[27] The doctrine comes from Eastern Railroad Presidents Conference v. Noerr Motor Freight, Inc., 365 U.S. 127, 81 S.Ct. 523, 5 L.Ed.2d 464 (1961), reh. denied, 365 U.S. 875, 81 S.Ct. 899, 5 L.Ed.2d 864, and United Mine Workers v. Pennington, 381 U.S. 657, 85 S.Ct. 1585, 14 L.Ed.2d 626 (1965).

[28] The court is persuaded by the reasoning of Norcen Energy Resource Limited v. Pacific Gas and Electric Company, 1994 WL 519461 (N.D.Cal.1994), that if Objectors had established PG & E's liability under the price squeeze doctrine that would not moot the Noerr-Pennington defense. See also City of Columbia v. Omni Outdoor Advertising, Inc., 499 U.S. 365, 111 S.Ct. 1344, 113 L.Ed.2d 382 (1991) (discussed in Norcen).

[29] Objectors claim that PG & E blocked their attempts to arrange (and pay for) upgraded transmission by exaggerating the costs of transmission upgrades, interposing numerous delays in providing relevant data, and otherwise creating procedural road-blocks. The court makes no determination on this issue. The court does recognize, however, that Objectors had less incentive to press any demands for an upgrade in transmission while PG & E was absorbing net congestion charges.

[30] The AB 970 Report does not determine whether the remaining constraints in the Bay Area would be better served by upgrading transmission or local generation (or some other alternative such as conservation). The report notes that current planning studies (including the report itself) do not explicitly weigh the costs and benefits of additional transmission improvements, and "with a few exceptions, such decisions are not well documented." In the absence of such data the AB 970 Report apparently focused on projects that PG & E had already begun or proposed — in other words, PG & E itself appears to have determined that transmission, rather than local generation, is the preferred alternative for most (if not all) of the projects discussed in the AB 970 Report.

[31] On January 23, 2003, PG & E filed a motion in limine to exclude the expert testimony of Objectors' damages expert, Dr. Michael C. Keeley. On January 27, 2003, Objectors filed a preliminary opposition to that motion (supplemented later, arguing among other things that any analysis under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), should be deferred until after all the evidence has been considered because this is a non-jury trial.) After hearing testimony from Dr. Keeley and PG & E's expert, Dr. Roy Shanker, the court deferred that issue for resolution in this Memorandum Decision.

The court is persuaded by Objectors' opposition and by the portion of their reply brief related to damages that Dr. Keeley's analysis is based upon sufficient facts or data, the product of sufficiently reliable principles and methods, and has been applied with sufficient reliability that it should not be excluded from evidence. Fed.R.Evid. 702. That being said, the testimony has proven to have little probative value to aid Objectors' cause.