Aguilar v. Atlantic Richfield Co.

92 Cal.Rptr.2d 351 (2000) 78 Cal.App.4th 79

Theresa AGUILAR et al., Plaintiffs and Appellants,
v.
ATLANTIC RICHFIELD COMPANY et al., Defendants and Appellants.

No. D030628.

Court of Appeal, Fourth District, Division One.

January 31, 2000. Review Granted May 17, 2000.

Cohelan & Khoury, Timothy D. Cohelan, Isam C. Khoury, Margaret L. Coates; Daniel J. Mogin and Angela Milea Mogin, San Diego, for Plaintiffs and Appellants.

John J. Sansone, County Counsel (San Diego), Diane Bardsley, Chief Deputy County Counsel, Timothy M. Barry, Deputy County Counsel; Casey Gwinn, City Attorney (San Diego) and Leslie E. Devaney, Assistant City Attorney, as Amici Curiae on behalf of Plaintiffs and Appellants.

Ronald C. Redcay, Richard C. Morse, John J. Kralik, IV, Susan C. Wright, Los Angeles; Post, Kirby, Noonan & Sweat, David J. Noonan, Sandra L. Lackey; Latham & Watkins, James W. Baker, Peter

H. Benzian, San Diego, John J. Lyons, Gregory N. Pimstone, Los Angeles, Julia E. Parry, San Diego; Pillsbury, Madison & Sutro, Robert A. Mittelstaedt, Craig E. Stewart, Caroline N. Mitchell; Paul R. Truebenbach, San Francisco; Hogan & Hartson, Mary Carter Andrues, Kirsten S. Harbers, John Mark Potter, Los Angeles, Andrew J. Kilcarr, Stephen G. Vaskov, Washington, Dist. of Columbia; Elizabeth J. Haeglin; Patrick J. Sullivan, San Diego, Kelly Scoffield, Gregory T. Kenny; Munger, Tolles & Olson, Ronald L. Olson, Bradley S. Phillips, William D. Temko, Los Angeles, Hojoon Hwang, San Francisco; Raymond V. McCord, Los Angeles; Robert E. Fuller, Universal City, Mark D. Litvak; Howrey & Simon, Los Angeles, Charles H. Samel, Washington, Dist. of Columbia, Dale J. Giali, Cheryl O'Connor Murphy, Los Angeles, Mark I. Levy, Alan M. Grimaldi, Washington, Dist. of Columbia; Lawrence R. Jerz, White Plains, NY; Blecher & Collins, Maxwell M. Blecher, Harold R. Collins, Jr., William Hsu, Los Angeles; Marylin Jenkins Milner; Manatt, Phelps & Phillips, Craig J. de Recat, Kevin O'Connell and Dennis Franks, Los Angeles, for Defendants and Appellants.

Neilsen, Merksamer, Parrinello, Mueller & Naylor, John E. Mueller, Andrew M. Wolfe, Mill Valley; McDougal, Love, Eckis & Grindle, and Glenn P. Sabine, El Cajon, as Amici Curiae on behalf of Defendants and Appellants.

McDONALD, J.

The defendants in this class action lawsuit are companies that refine oil and market gasoline to California consumers.^[1] Plaintiffs' antitrust lawsuit asserts that defendants seized the opportunity provided by California's requirement that a cleaner-burning gasoline (CARB gas) be used in California, and agreed with each other to restrict CARB gas refining capacity and production, and fix CARB gas prices. Plaintiffs allege defendants' conduct violated the Cartwright Act (Bus. & Prof.Code, ง 16720 et seq.) and the Unfair Competition Act (Bus. & Prof.Code, ง 17200 et seq., hereinafter the UCA).

After exhaustive discovery proceedings, defendants moved for summary judgment. Defendants relied on (1) the declarations by numerous senior managers of each defendant except Tosco that denied CARB gas refining capacity, production or pricing resulted from any agreement or conspiracy but rather from independent decisions based on each defendant's self-interest (the "denial declarations") and (2) plaintiffs not having evidence to support a reasonable inference that the defendants' CARB gas capacities, production or pricing resulted from an agreement or conspiracy (the "no evidence" contention). The trial court granted summary judgment for defendants, concluding defendants' showing shifted the burden to plaintiffs to demonstrate the existence of a triable issue of material fact, and plaintiffs' counter-showing did not raise a triable issue of the existence of an agreement in violation of the Cartwright Act.

Plaintiffs moved for a new trial. The trial court concluded it had erroneously shifted the burden to plaintiffs to raise a triable issue of material fact and granted plaintiffs' motion for a new trial.

Defendants appeal the order granting a new trial. Plaintiffs cross-appeal the summary judgment for defendants, asserting that if the new trial order is reversed it was error to grant summary judgment. We conclude the trial court erred by granting plaintiffs' motion for a new trial. We also conclude the defendants are entitled to summary judgment because they showed there is no evidence of an agreement among them and plaintiffs' showing of collusion among defendants is based on inferences from circumstantial evidence that are equally explicable by defendants' independent actions as by an illegal agreement.

I

UNDISPUTED FACTS

A. The Parties

Plaintiffs represent the class of consumers in California who purchased CARB gas produced and marketed by defendants at allegedly inflated prices during 1996. Defendants are companies that refine oil in California and market gasoline to California consumers.

B. CARB Gas

In 1991, the California Air Resources Board adopted regulations requiring a cleaner-burning gasoline for sale in California. (13 Cal.Code Regs., งง 2260 et seq.) The regulations required that by June 1, 1996, only CARB gas could be sold at retail in California.

C. Creating CARB Gas Refining Capacity^[2]

Defendants' refineries required modification to produce CARB gas for the California market. These modifications required extensive capital expenditures.

Each defendant modified its California refineries to produce some CARB gas. The capital expenditures and capacities created varied widely among defendants. Chevron spent more than $1.3 billion to create a capacity of 240,000-250,000 barrels of CARB gas per day (BPD). Shell spent more than $1 billion to create a capacity of approximately 100,000 BPD. ARCO spent approximately $450 million to create a capacity of 150,000 BPD. Unocal spent more than $400 million to create a capacity of 100,000 BPD. Ultramar spent approximately $325 million to create a capacity of 60,000 BPD. Exxon spent almost $200 million to create a capacity of 100,000 BPD. Mobil spent $126 million to create a capacity of 88,000 BPD. Texaco spent more than $113 million to create a capacity of 47,000 BPD. Tosco spent $100 million to create a capacity of 50,000 BPD. The refinery modifications converted a significant portion of California's oil refinery capacity from conventional gasoline to CARB gas.^[3]

The resulting capacities did not all equal the volume of each defendant's historic gasoline sales to California consumers. Some defendants, including Texaco, created a capacity equal to the volume of its historic gasoline sales level. Other defendants, including Tosco and Unocal, created capacities less than each anticipated would be necessary for its California customers; each elected to fill its shortfalls by obtaining CARB gas from other refiners through exchange agreements. Other defendants, including Chevron, maximized their capacity.

D. Introduction of CARB Gas

Prices for gasoline significantly increased when CARB gas was introduced in 1996, although defendants' total gasoline production in 1996 exceeded gasoline production in 1995. Defendants operated their refineries at almost full capacity and as efficiently as possible to produce CARB gas consistent with the production of other refined products.^[4]

II

DEFENDANTS' SUMMARY JUDGMENT MOTIONS

Defendants' summary judgment motions were based on denial declarations and the no evidence contention showing.

A. The Capacity Decisions

Plaintiffs alleged defendants agreed to limit each defendant's capacity by coordinating their refinery modifications. A typical evidentiary showing attacking this allegation was submitted by Texaco. Mr. Walz, Texaco's most senior executive with direct responsibility over Texaco's two California refineries during the relevant time periods, averred he was continuously and personally involved in the decision-making process and the decisions about Texaco's CARB gas refinery modifications. He averred that Texaco's decisions were made independently and without agreement or collusion with other defendants, and were based on Texaco's best business judgment of its own economic interests. He explained that to comply with California Air Resources Board regulations, Texaco assessed whether to purchase or produce CARB gas to meet its CARB gas needs. Texaco compared the costs of refinery modifications with the risks that the market price for CARB gas would not provide an adequate return on its investment, and the cost to import CARB gas from outside California with the cost to produce CARB gas in California. Texaco used internal analyses and outside consultants to evaluate the alternatives, and ultimately determined it was in Texaco's best independent economic interest to modify its two California refineries to produce CARB gas.^[5] Texaco made refinery modifications that its studies deemed economically justifiable.

Mobil, ARCO, Chevron, Ultramar, Exxon and Shell made similar evidentiary showings that denied capacity decisions were based on agreement or collusion with other defendants and affirmatively demonstrated each independently followed its own economic self-interests in determining capacity. (See Appendix A.)

Tosco's evidentiary showing asserted that its decision to make CARB gas refinery modifications was driven by its independent business judgment of its own economic interest. In 1991 Tosco began investigating the costs and benefits of converting some or all of its Avon refinery product from conventional gasoline to CARB gas. In 1991 its engineers estimated converting to 100 percent capacity would cost almost $300 million. A 1992 outside study commissioned by Tosco estimated those costs would exceed $400 million, which might be understated by 50 percent. Based on an estimate that CARB gas would yield 9.5 cents per gallon (cpg) more than conventional gasoline, the study concluded that Tosco would lose money by converting Avon to 100 percent capacity. Tosco determined the cost of converting Avon to CARB gas production would be minimized by converting to only 50 percent capacity. Tosco authorized $100 million to convert its Avon refinery to a 50,000 BPD capacity. Tosco estimated it would break even financially if CARB gas brought a 9 cpg premium over conventional gasoline, and would realize a 15 percent internal rate of return were the differential 12 cpg. Other options based on projected lower rates of return at the same price differentials were considered but rejected.^[6]

B. The Production Decisions

Defendants' capacity decisions limited the amount of CARB gas each could produce. Plaintiffs' complaint alleged defendants' refineries were "operating at virtually full capacity," and their expert conceded defendants operated their refineries as efficiently as possible.

Plaintiffs nevertheless alleged defendants agreed to limit production of CARB gas. Defendants denied their production was influenced by any agreement or collusion with other defendants and affirmatively showed each defendant produced the volume of CARB gas that each unilaterally determined was in its best economic interest to produce.

Ultramar's evidentiary showing was typical of defendants' showings. Mr. Hass, Ultramar's Manager for Refinery Economics and Planning, denied that Ultramar's volume of production of CARB gas, conventional gasoline and other refined products was influenced by express or implied agreements or collusion with other defendants. He explained the process by which Ultramar's production decisions are made. Because a refinery's size is finite, a decision to produce CARB gas necessarily limits the production of conventional gasoline or jet fuel. Each month Ultramar unilaterally determines the product mix its refinery will produce during the next 90-day period. To determine product mix, Ultramar uses a computer model that considers variables, including the types and costs of crude oil and other raw materials, the anticipated demand, and prices for and cost to produce each of its products. The model evaluates the product mix most likely to optimize the refinery's efficiency and profitability. Hass also averred that from April through August 1996 Ultramar produced more gasoline than it had during the same period of the preceding year, and produced as much gasoline as it could without endangering the operation of the refinery. The other defendants made substantially similar evidentiary showings. (See Appendix B.)

C. The Pricing Decisions

Plaintiffs alleged defendants agreed to fix CARB gas prices at excessive levels by restricting capacity and CARB gas production and by manipulating a CARB gas benchmark price used on the spot market (the "OPIS" spot price);^[7] the manipulated price was then used to coordinate defendants' pricing decisions.

Defendants' evidentiary showings denied their CARB gas pricing decisions were influenced by any agreement or collusion with other defendants and demonstrated their CARB gas prices were unilaterally determined in each defendant's best economic interests.

Exxon's evidentiary showing was typical of defendants' showings. Mr. Voller, Exxon's Fuel Products Price Coordinator, whose primary job responsibility was to develop appropriate wholesale gas prices, denied that Exxon consulted, agreed or conspired with other defendants to set the price for either CARB gas or conventional gasoline. He explained the extensive and ongoing process, which was unchanged for CARB gas, by which Exxon set its prices for sales to branded Exxon dealers (the DTW price) and distributors or jobbers (the rack price). Exxon's goal is to set prices within a range that allows Exxon dealers and distributors to compete, and its prices are typically below the prices of its competitors. Rack and DTW prices are determined by consulting numerous sources of information, most of which are publicly available. Exxon surveys the street price being charged by Exxon's competitors, obtains historical DTW and rack prices from third party trade publications, including OPIS, the Lundberg Survey (Lundberg) and Platts, and considers market conditions and operational problems experienced by Exxon's competitors, all of which are factored into Voiler's recommendation of the DTW and rack prices charged by Exxon. Similar evidentiary showings were made by the other defendants. (See Appendix C.)

To rebut plaintiffs' allegation that defendants agreed to manipulate the OPIS CARB gas spot price and to coordinate their CARB gas prices based on that manipulated price, defendants first asserted they lacked the ability to manipulate the OPIS spot price. Mr. Berhang, an OPIS official, testified that OPIS gathers information from twenty-five different supply and trading sources to determine the CARB gas spot price, and only two or three defendants are sources for that information.^[8] Second, defendants averred their decisions on DTW and rack prices were not based on OPIS spot prices. (See Appendix C.) Finally, defendants pointed out plaintiffs' expert admitted in his deposition that, although the OPIS spot price is considered, market conditions, including the retail and DTW prices of other defendants, costs of production and desired margins are the primary factors determining CARB gas retail, DTW and rack prices.

D. The Exchange Agreement Decisions

Plaintiffs alleged that a major component of defendants' conspiracy to limit capacity and CARB gas production and fix CARB gas prices was a web of two-party exchange agreements between defendants, the purpose and effect of which were to prevent excess CARB gas from entering markets not controlled by defendants. Defendants' evidentiary showings denied the exchange agreements were arrangements to restrict the availability of CARB gas and demonstrated that each defendant had independent business reasons for entering into exchange agreements.^[9]

Shell's presentation was typical of the evidentiary showings supporting defendants' positions on exchange agreements. Mr. Murphy, an exchange trader responsible for negotiating exchange agreements for Shell's Supply & Trading group, averred that his negotiations for exchange agreements are conducted to advance Shell's own independent business interests, and are not used as a pretext to pass confidential information to Shell's competitors or to further a conspiracy to fix prices. He also explained the independent business reasons for each of Shell's exchange agreements. Shell has refineries in Northern California and Anacortes, Washington, but none in Southern or Central California. Shell meets its Southern and Central California marketing needs through five long-term exchange agreements. Under the Shell-ARCO exchange agreement, Shell delivers 30,000 BPD to ARCO in Northern California and ARCO delivers 30,000 BPD to Shell in Southern California, where Shell has substantial marketing needs but no refinery. Shell estimated this agreement saves the company between $8 and $14 million per year in shipping costs and $4 million in terminalling costs. Under the Shell-Chevron exchange agreement Shell delivers 20,000 barrels per day of conventional gasoline to Chevron at Anacortes and Chevron delivers 12,000 BPD to Shell in Northern California and 8,000 BPD to Shell in Southern California, with a price differential paid to Chevron for the higher cost of producing CARB gas rather than conventional gasoline.^[10] The agreement saves Shell millions of dollars it would otherwise spend for marine shipping from its Anacortes refinery to Southern California.^[11] Under the Shell-Texaco exchange agreement, Shell exchanged approximately 40,000 BPD from its Northern California facility with Texaco for CARB gas from Texaco's Bakersfield facility. Texaco has one of only two refineries located in the Bakersfield area. Shell's alternatives to meet its marketing needs in that area were shipping product to Bakersfield by truck or obtaining product by exchange. Truck shipping is prohibitively expensive and the exchange permitted Shell to supply product to its Bakersfield retail outlets at competitive prices.^[12]

Mr. Stewart, Mobil's products trader responsible for negotiating exchange agreements, averred Mobil's exchange agreements are not used as a device to exchange price or production information with competitors. Mobil has a single exchange agreement: Mobil provides 30,000 BPD premium unleaded to Texaco in Los Angeles in exchange for three grades of CARB gas provided to Mobil's branded wholesale distributor at Texaco's Bakersfield facility. That agreement, coupled with a term purchase agreement under which Mobil purchases CARB gas from Texaco that is delivered to the Santa Fe pipeline for transport to Fresno, supplies Mobil's marketing needs in the Bakersfield and Fresno areas. In 1996 Mobil also had three term sales agreements under which Mobil sold 12,000 BPD to three independent gasoline resellers without restriction as to whom they could resell the CARB gas.

Ms. Barnes, Ultramar's General Manager of Product Supply and Distribution, responsible for balancing product supply and demand for Ultramar and a participant in the negotiation of Ultramar's exchange agreements, averred the exchange agreements were independently negotiated in Ultramar's independent business judgment. Ultramar has a single refinery in Los Angeles, and its production exceeds the amount needed for Ultramar's branded retail sales in Southern California. Ultramar sells part of its excess CARB gas to independent, unbranded dealers and large, third party wholesalers,^[13] and exchanges part with Exxon, Tosco and Shell. These exchange agreements provide Ultramar with product in Northern California to supply Ultramar's Northern California branded and independent customers. Because there is no pipeline from Southern to Northern California, Ultramar's exchange agreements allow it to be competitive in Northern California; without these exchange agreements Ultramar would be forced to withdraw from supplying its Northern California branded and independent customers.

The other defendants provided similar explanations for their exchange agreements. (See Appendix D.)

E. The Information-Gathering Decisions

A final aspect of plaintiffs' antitrust claim alleged that defendants agreed to restrict capacity and CARB gas production and to fix CARB gas prices by using common consultants as a conduit for exchanging confidential information. Plaintiffs contended that each defendant used consultants, including PACE, Wright Killen & Co., and PIRA Group (PIRA), under the guise of assisting refinery modification planning, as clearinghouses for transmitting to competitors confidential information about capacity plans. This shared information facilitated defendants' agreement to limit CARB gas supply and permit CARB gas prices to be maintained at supra-competitive prices.

Defendants' summary judgment motions responded to this claim in disparate ways. Tosco demonstrated that it did not use the consultants that plaintiffs' discovery responses listed as transmitting confidential information on industry-wide projections for capacity. Other defendants showed that they used one or more of the consultants to assist in evaluating the economic feasibility of their particular refinery modifications. However, they executed confidentiality agreements with the consultants prohibiting disclosure of confidential information to third parties. For example, Mr. Miller, Shell's Manager for Refinery Planning responsible for developing the planned capital expenditures for Shell's refinery modifications, explained Shell had internally projected that costs to convert its Martinez and Anacortes facilities to 100 percent capacity would exceed $750 million. Shell examined whether the projected price for CARB gas justified that investment. It estimated the likely supply and demand, using publicly available information, and evaluated whether Shell could satisfy its own marketing needs with lower capital expenditures. Shell used its internal studies and commissioned PACE to identify the most economic way for Shell to meet the new CARB gas requirements. Shell provided PACE with confidential information regarding its refinery but required PACE to keep the information confidential. The PACE study focused on Shell's refineries and modification costs, and Shell did not ask PACE to estimate the capacities of any other refiners. PACE recommended Shell convert its refineries to 70 percent capacity, but Shell chose not to follow PACE's recommendations. PACE kept confidential the information provided by Shell and did not use the Shell data in any other studies conducted by PACE. Similar evidentiary presentations were made by other defendants. (See Appendix E.)

F. Summary

Defendants argued they established that no agreement or conspiracy existed, and plaintiffs' discovery responses showed there was no evidence of an agreement or conspiracy; therefore, summary judgment for defendants was appropriate.

III

PLAINTIFFS' OPPOSITION TO SUMMARY JUDGMENT

A. Overview

Plaintiffs opposed defendants' summary judgment motions on numerous grounds. Plaintiffs asserted that defendants' showing did not negate an essential element of plaintiffs' claim, and plaintiffs therefore had no burden to provide evidence raising a triable issue of material fact to defeat defendants' motions. Plaintiffs alternatively argued that summary judgment was improper because a jury could infer from defendants' planning activities preceding the introduction of CARB gas, agreements that pooled supplies, and use of a common cost of goods to fix CARB gas prices that defendants agreed to: (1) coordinate their capacity decisions to limit CARB gas supply and maintain CARB gas prices at supra-competitive levels; (2) exchange production information to control the availability of CARB gas; and (3) pool and control supplies of CARB gas and exchange confidential information to fix CARB gas prices.

Plaintiffs offered no direct evidence that defendants expressly or tacitly agreed or conspired to limit capacity or CARB gas production, or fix CARB gas prices. However, plaintiffs asserted the agreement among defendants can be inferred from circumstantial evidence. Plaintiffs argued several categories of evidence they submitted in opposition to defendants' summary judgment motions support the inference of an express or tacit agreement or conspiracy.

B. The Common Motivation

An industry survey prepared by PIRA concluded in early 1996 that CARB gas supply appeared to be "tight in California" and that the companies involved in CARB gas production "share a motivation to recoup costs and not undermine the market." Similarly, an industry analyst from the American Petroleum Institute opined at a late 1995 trade conference that unless the United States industry reduced capacity, there would never be any substantial increase in refining margins.

Internal documents from several defendants also acknowledged excess CARB gas supply could reduce prices and hurt profitability.^[14] Several defendants were apparently concerned that California's small refiners exemption from California Air Resources Board regulations might permit them to produce conventional gasoline at lower costs and thereby undermine the CARB gas market premium. Some defendants filed suit to overturn the small refiners exemption.

C. Exchanging Competitive Information

Plaintiffs argued each defendant signaled the others on its planned capacity and CARB gas production with the tacit understanding that other defendants' choices of capacity and CARB gas production would be guided by this information. These signals were allegedly given and confidential information was allegedly exchanged through several channels.

1. Direct Channels for "Signals": Negotiations for Exchange Agreements

Plaintiffs argued that defendants, by entering into two-party exchange agreements, disclosed to each other confidential competitive information on capacity and CARB gas production. With this information each defendant coordinated its own capacity and CARB gas production to limit the supply of CARB gas. While defendants were making decisions on their respective capacities, many of the defendants negotiated exchange agreements. For example, ARCO's Northern California retail outlets required 50,000 BPD but ARCO had no Northern California refinery. ARCO had previously satisfied these needs through a 1986 exchange agreement with Tosco that was to expire in 1996. ARCO and Tosco recognized Tosco's capacity conversion would be affected by whether they agreed on an extension of their exchange agreement.^[15]

ARCO also explored exchange agreements with Chevron and Shell.^[16] The ARCO/Chevron negotiations included discussions between senior company officials about pricing the CARB gas Chevron would supply to ARCO. ARCO told Chevron that an ARCO/Chevron exchange agreement "could change some of [ARCO's] investment decisions or change investment decisions of others on supplying CARB gasoline." By March 1994, Chevron began discussing internally whether to change its capacity strategy, because the discussions with ARCO showed that Chevron's "excess CARB [gas] may provide a significant opportunity for Chevron [and] highlighted Chevron's excess CARB [gas] capacity...."^[17] Additional meetings between ARCO and Chevron officials did not result in a contract.^[18] Instead, Chevron and Tosco eventually agreed to exchange Chevron's excess CARB gas for Tosco's conventional gasoline,^[19] and ARCO and Shell reached a separate agreement. (See pt. H.D., ante.)

Plaintiffs asserted that this web of two-party exchange agreements provided defendants' senior management an opportunity to exchange confidential capacity information and to coordinate their capital expenditures and capacity decisions.

2. Media Channels for "Signals"

In September 1994 Chevron and Tosco discussed Tosco obtaining CARB gas from Chevron's excess capacity,^[20] and ARCO considered the risks of choosing between an exchange agreement or waiting until the supply and demand picture cleared. Exxon, a primary supplier to the unbranded market, examined its strategy to maximize profits from its own CARB gas production.

In the fall of 1994, the industry press reported Tosco's President O'Malley projected the production of CARB gas by all refiners would be about 800,000 BPD, and consumers would use 940,000 BPD, resulting in a shortage in 1996 of 140,000 BPD. Because of the projected shortfall, he warned retailers that "if I were a California retailer and didn't have a widely-recognized brand or locked up supply, I'd be concerned about the future."^[21]

Plaintiffs argued that O'Malley's signal was integrated into Exxon's plans. A few days after O'Malley's comments were reported, Exxon's internal documents show discussions within Exxon that "changing [West Coast] market fundamentals may provide unique opportunities: [ถ] -California projected to be short CARB gasoline by 100 kBD [100,000 BPD], [creating a] potential premium for secure supplies."^[22] The memo also projected that the San Francisco Bay Area was shifting from "100 kBD long conventional to [approximately] 10 kBD long CARB," and therefore proposed a business strategy that would "grow the Branded Distributor [] business...; [and] proactively prevent [Exxon] oversupplying [San Francisco Bay Area] pipeline and rack markets by [] [p]ursuing pipeline and rack sales to end users versus brokers and resellers [and] [e]xecuting [San Francisco Bay Area] export sales when end-users sales not available."^[23] By May 1995 Exxon's West Coast strategy was to place its additional volume to "branded `anyone,'" to avoid placing CARB gas into the San Francisco Bay Area spot market and to shift those sales into the branded business.

3. Common Consultant Channels for Exchanging Information and Coordinating Decisions

Plaintiffs asserted that common consultants acted as "pollinators" to distribute among defendants information that assisted them in coordinating their anti-competitive conduct.

Shortly after the adoption of the CARB gas regulations, PACE held a January 1992 seminar entitled "Refining Operations-After CARB," attended by several defendants. Among the topics discussed were how to obtain CARB gas and the future of capital spending, including the subtopic of "Increasing Supply Impact On Prices." However, plaintiffs do not suggest any of the information distributed during this seminar was confidential competitive information.

In late 1993 PACE solicited subscriptions from defendants to its multi-client study entitled "California Refining: 1996 and Beyond." PACE advertised the study as providing defendants with valuable information about the economics of producing and selling CARB gas.^[24] PACE contacted various defendants in conducting its study and obtained their views on how various technical issues would be handled within their respective companies,^[25] and their opinions on how their competitors might act. PACE also examined environmental impact reports filed by major refiners, described by PACE as a "fertile source of competitive intelligence," to summarize the proposed modifications of the various refineries. PACE published its August 1994 report that surveyed the challenges, risks and market projections for refinery operations after the introduction of CARB gas and provided PACE's view of the competitive positions of the various defendants. Even assuming PACE's conclusions were influenced by confidential competitive information,^[26] plaintiffs do not cite any portion of that report in which confidential competitive information for any defendant was disclosed to the subscribers of that report.^[27]

Plaintiffs alleged that other consultants performed similar studies that provided a defendant's capacity and CARB gas production information to other defendants. (See Appendix E.)

D. Use of Exchange Agreements to Control and Coordinate Production and to Elevate and Fix Prices

Plaintiffs argued there was no pro-competitive justification for the exchange agreements, the only purpose of which was to "preserve market discipline." Plaintiffs' premise was that the only legitimate reason for an exchange agreement is to provide product for a geographic area in which the exchange parties lack capacity.^[28] From this premise, plaintiffs asserted the exchange agreements permit an inference of coordinated anti-competitive conduct because the exchange agreements were geographically unorthodox.^[29] Plaintiffs argued the real purpose of exchange agreements was to pool capacity, coordinate production and allocate CARB gas among the defendants, thereby limiting the supply available to unbranded independents.^[30] For example, plaintiffs pointed out that Chevron and Exxon produced more CARB gas than necessary to meet their branded marketing needs and marketed the excess through exchange agreements rather than through sales to independent retailers.^[31] Plaintiffs also pointed out that, even among those defendants that sold to the unbranded market, some limited the volume an independent retailer could purchase.^[32]

Plaintiffs also argued that defendants used exchange agreements to fix the CARB gas price. Plaintiffs pointed out that in two exchange agreements in which CARB gas was exchanged for conventional gasoline, the party that delivered conventional gasoline and received higher cost CARB gas was required to pay a differential on a per-gallon basis, a portion of which was fixed and a portion of which was based on the OPIS spot market price.^[33] Chevron believed the "fixed CARB/conventional differential [of the Chevron/Tosco agreement] may help establish, more or less, the market differential."^[34] Plaintiffs asserted that by agreeing to use the OPIS spot price to fix the price of CARB gas in their exchange agreements, defendants effectively agreed to use the OPIS spot price, which they could manipulate, as a common pricing mechanism.^[35]

E. Use of Spot Market to Manipulate and Fix Prices

Plaintiffs' theory was that defendants' control of the CARB gas supply obtained through exchange agreements permitted them to manipulate and use the OPIS spot price to coordinate their wholesale CARB gas prices and collusively set rack and DTW prices.

Plaintiffs' theory begins with the OPIS spot price, which is determined daily by OPIS based on the prevailing market price paid that day for bulk purchases of CARB gas on the spot market. OPIS sets the spot price even though at times there are no actual spot sales to report.^[36] In the absence of sales, OPIS makes a subjective assessment of the market to report the spot price. OPIS gathers information from twenty-five different supply and trading sources, including two or three of the defendants, to determine the CARB gas spot price.

Defendants control the supply of CARB gas in California and the spot market is a small fraction of the total CARB gas sales in California. Exxon recognized that the spot market was "very thin." Among Exxon's goals was excluding CARB gas from the San Francisco Bay Area spot market and developing an exchange strategy to minimize spot market impact.

Plaintiffs argued that in addition to defendants' alleged manipulation and use of the OPIS spot price to control prices, defendants obtained instantaneous information from Lundberg on the current wholesale prices charged by their competitors; this information allowed defendants to coordinate prices. Rack and DTW prices are collected on a daily basis from jobbers and retailers,^[37] and Lundberg publishes the information to its subscribers, including defendants.

The OPIS spot price and the wholesale prices reported by other services are used by defendants to determine product mix and by some but not all defendants to set their rack and DTW prices. (See Discussion, pts. II.B. & LLC, ante.)

F. Plaintiffs' Experts

Plaintiffs produced testimonies from experts Dr. Noll and Dr. Bloom.

Noll believed gas prices have been above competitive levels for all but three months since CARB gas was introduced. Noll opined that defendants engaged in anti-competitive conduct by coordinating their capacity decisions.^[38] Noll also opined that, once the capacity decisions had been made, defendants knew the product mix other defendants would select because they knew each other's refinery capacities and used similar programs to make their product mix decisions. Noll's principal thesis appears to be that by coordinating their refinery capacity, entering into exchange agreements and disseminating pricing information, defendants erected an institutional structure necessarily resulting in anti-competitive and collusive conduct.^[39]

Noll identified several price collusion methods employed by defendants. He explained that because of the thin market for spot transactions, a trader who buys and sells CARB gas on the spot market knows that his buy-sell price will have "spill-over" effect on other contract prices to which his company, as well as other defendants and third persons, are parties. Noll also criticized the exchange agreements as creating a horizontal merger and coordination of supply among defendants. By linking the price in exchange contracts to the spot price, which defendants control through ostensibly separate spot transactions, defendants can fix the entire market price for CARB gas through a few spot trades.^[40]

Noll apparently opined that defendants, motivated to charge artificially high prices to the public, have the incentive and ability to buy and sell CARB gas at artificially high prices to increase the cost of CARB gas in all "linked" transactions. Defendants then use this artificially elevated price to determine the DTW and rack prices charged to retailers, who then charge these inflated prices to consumers. Noll's theory at bottom asserts that, because of the institutional structures within which the trading and pricing activities operate, it is functionally impossible for defendants to act independently even though they may believe they are acting independently. First, prices at all levels are influenced by the prices that are disseminated through the reporting services and, on receiving this information, defendants use that information to set their prices. Second, because prices for major portions of the supply are under the exchange agreement structure influenced by spot transactions, a relatively small spot transaction necessarily fixes the price on ostensibly separate transactions. Finally, some prices for large transactions are based on a blend of a fixed and a floating differential, thereby locking in the cost of those goods and influencing the DTW and rack prices of those goods, which prices in turn influence an unrelated competitor's DTW and rack pricing.

It appears to be Noll's theory that the key collusive acts of defendants were to limit capacity and enter exchange agreements using a price floating with the market. A defendant who uses market prices of competitors to set its own spot, rack or DTW prices is incapable of independent conduct because the oligopolistic nature of the industry combines with California's isolation from imports to eliminate any possibility for independent conduct.^[41]

Bloom opined that defendants exchanged and used information in an anti-competive competitive manner. Bloom believed defendants used trade association meetings to signal competitors on pricing and production.^[42] Bloom also opined defendants exchanged information through sharing common consultants, but did not identify any confidential information transmitted through these consultants.

Bloom's principal position was that defendants disseminated pricing information through OPIS, Lundberg and other media, and used that information in anti-competitive behavior. He claimed the OPIS spot price signaled future prices and permitted coordinated pricing. He also opined there was no justification for disseminating defendants' DTW and rack prices to OPIS, Lundberg and other media services other than to facilitate collusive pricing.

G. The Bottom Line: Prices for CARB Gas Go Up

Plaintiffs cited evidence that defendants had predicted there would be a price increase when CARB gas was introduced, and the predictions were correct.

Numerous defendants had anticipated that tight supplies would produce price increases and price spikes.^[43] ARCO's November 1995 CARB gas roll-out chart contained plans to "begin price increases to recover costs" when the April 1996 roll-out began. Defendants used public relations campaigns to assuage public outcry over price increases.^[44]

Plaintiffs asserted there was no satisfactory explanation for the price increases. An April 1996 press release from the California Air Resources Board chairman stated that, although an extra 5-8 cpg hike for CARB gas was attributable to regulations, "[t]he rest of these outrageous price hikes reflect events ... that are unrelated to California regulations. After listening to the oil industry's testimony, I still have not heard anything that fully explains these price increases."^[45]

IV

THE RULINGS

A. The Order Granting Summary Judgment

The trial court initially concluded defendants' evidentiary showing was sufficient to prima facie negate the necessary antitrust element of agreement, and shifted to plaintiffs the burden to produce evidence raising a triable issue of material fact on the element of an agreement. The trial court then examined plaintiffs' showing and concluded plaintiffs had not shown the existence of a triable issue of material fact. The trial court evaluated plaintiffs' evidence of defendants' motivation, use of common consultants, exchange agreements, use of spot markets, dissemination, and use of price and other public information, and concluded there was no direct evidence of any agreement to restrain trade or fix prices.

In part, the trial court found that declarations of officials of some defendants denying that they or the defendants based any of their capacity, production, pricing or marketing decisions on agreements with, or confidential information obtained from, competitors was sufficient to shift the burden of proof to plaintiffs under Biljac Associates v. First Interstate Bank (1990) 218 Cal.App.3d 1410, 267 Cal.Rptr. 819 (Biljac). The trial court also concluded plaintiffs' evidence did not permit a reasonable inference of the requisite agreement.

Based on these conclusions, the trial court granted defendants' summary judgment motions.

B. The New Trial Order

Plaintiffs moved for a new trial under Code of Civil Procedure section 657^[46] asserting error in law, irregularity in the proceedings, accident or surprise, newly discovered evidence and insufficiency of the evidence. Defendants opposed the motion.

The trial court, reassessing its interpretation of and focusing exclusively on Biljac, concluded it had made an error of law and granted plaintiffs' motion for new trial.^[47] However, the trial court's accompanying statement of reasons shows it merely reconsidered the sufficiency of defendants' denial declarations denying any agreement among them to fix prices, exchange confidential pricing, production and investment information, or to restrict California CARB gas supplies. The trial court, after further briefing, argument and review of the defendants' denial declarations, determined that, standing alone, they were inadequate to shift the burden of proof to plaintiffs. The trial court's reasons were as follows: "In granting summary judgment to defendants, the court relied on the declarations of various of defendants' employees, who denied the existence of any agreement, the purpose of which was to fix prices and/or supply." In granting defendants' motion for summary judgment, the trial court had relied on Biljac for the proposition that these declarations were sufficient under section 437c, subdivision (o)(2) to shift the burden of proof to plaintiffs to raise a triable issue of material fact. Although the trial court characterized its grant of the motion for new trial as based on an error in law, because the declarants submitting the denial declarations were not the ultimate decision-makers, in fact its statement reflected its belief that its initial decision regarding the evidentiary strength of the declarants' professed denials of conspiracy was incorrect. It appears from the trial court's statement that its new trial order encompassed only the issue of whether bare denials of collusive activity contained in declarations by persons who were not shown to be executives having the ultimate authority within each defendant to authorize or engage in meaningful conduct at the corporate level will, without more, suffice to shift the burden of proof to plaintiffs. Here, the trial court concluded the statements by these declarants that they did not conspire and their conclusory unsupported opinions that the defendants who employed them did not conspire (see Biljac, supra, at p. 1424, 267 Cal.Rptr. 819) could not satisfy the defendants' burden of negating conspiracy.

We need not quarrel with the trial court's findings of the inadequacy of those portions of defendants' denial declarations merely denying that an individual employee conspired and dismissing the speculative opinions of the defendant employer's conduct. We do not find it necessary to disagree with its conclusion that standing alone, they would not shift the burden of proof to plaintiffs. Rather, it is our task to review the correctness of what is in effect an order denying summary judgment. (See Malo v. Willis (1981) 126 Cal. App.3d 543, 546, fn. 2, 178 Cal.Rptr. 774.) In doing so, we independently consider defendants' evidence in its entirety, including factual portions of the denial declarations, a small portion of which contain denials of collusive action, as well as the evidence submitted by plaintiffs. Each defendant's denial declaration comprehensively recites relevant facts that are within the declarants' personal knowledge and are verified by competent accompanying documentation.

V

ANALYSIS OF NEW TRIAL ORDER

The appellate court ordinarily defers to the trial court's grant of a motion for new trial and will affirm absent an abuse of discretion. (Sandco American, Inc. v. Notrica (1990) 216 Cal.App.3d 1495, 1506, 265 Cal.Rptr. 587.) However, the grant of a motion for new trial on the ground solely of legal error is reviewed de novo to determine whether the legal error occurred. (Stoddard v. Rheem (1961) 192 Cal.App.2d 49, 13 Cal.Rptr. 496.)

Plaintiffs argue that a new trial order must be affirmed if any of the grounds stated in the motion for new trial justified the order. (ง 657.) Plaintiffs' motion for new trial asserted in addition to error in law the following grounds: irregularity in the proceedings, accident or surprise, and newly discovered evidence.^[48] On appeal, plaintiffs peremptorily claim there was irregularity or surprise because the trial court in granting summary judgment for defendants did not rule on specific evidentiary objections or advise plaintiffs it had determined the burden had been shifted. First, the record does not support plaintiffs' claim that they were unaware of the burden shifting; to the contrary, the court advised at the outset of oral argument that it intended the primary focus of the summary judgment hearing to be on plaintiffs' showing. More importantly, plaintiffs cite no authority to support their argument that a trial court, when ruling on a motion for summary judgment, is required to rule on specific evidentiary objections or advise that the burden had been shifted. Furthermore, as stated in Neal v. Montgomery Elevator Co. (1992) 7 Cal.App.4th 1194, 1199-1200, 9 Cal.Rptr.2d 497: "Although generally a new trial order will be affirmed if it should have been granted on any ground stated in the notice of intention [citation], where `a trial court in granting a new trial based its order exclusively upon an erroneous concept of legal principles applicable to the cause, its order will be reversed [citation].' [Quoting Conner v. Southern Pacific Co. (1952) 38 Cal.2d 633, 637, 241 P.2d 535.]")

Apart from the rule that a new trial order based on legal error will be reviewed de novo on appeal to determine whether there was legal error, because the grant of the new trial motion following the grant of summary judgment in this case is in effect a denial of the motion for summary judgment, we conclude the new trial order is reviewed in the same manner as the denial of a motion for summary judgment.

Plaintiffs' motion for new trial following the grant of summary judgment for the defendants was essentially a motion for reconsideration of the summary judgment; it focused on the same factual presentation and legal arguments made by the plaintiffs in opposition to defendants' motion for summary judgment. Ordinarily a trial court's order, in contrast to a judgment, is subject to reconsideration under section 1008. However, after a judgment is entered, the trial court no longer retains power to change the judgment in response to a motion for reconsideration. The trial court may, however, correct a judicial error in the summary judgment in response to a motion for new trial. (Passavanti v. Williams (1990) 225 Cal.App.3d 1602, 1606, 275 Cal.Rptr. 887; Ramon v. Aerospace Corp. (1996) 50 Cal.App.4th 1233, 1238, 58 Cal.Rptr.2d 217; 6 Witkin, Cal. Procedure (4th ed. 1997) Proceedings Without Trial, ง 49(b), pp. 445-446.) As stated in Green v. Del-Camp Investments, Inc. (1961) 193 Cal.App.2d 479 at page 481, 14 Cal.Rptr. 420:

"It may seem semantically anomalous to speak of a `new trial' following summary judgment, which is a determination that there shall be no trial at all. In the light, however, of the rule that a motion for new trial lies after judgment on the pleadings (Carny [Carney] v. Simmonds [(1957)] 49 Cal.2d 84 [315 P.2d 305]), the procedure appears proper."

There are few reported opinions discussing the extent and standard of appellate review of an order granting new trial following grant of summary judgment. However, the same principles apply to appellate review of an order granting new trial following a grant of judgment on the pleadings. (See Green v. Del-Camp Investments, Inc., supra, 193 Cal.App.2d 479, 481, 14 Cal.Rptr. 420.) We conclude from our review of the available authority that when a new trial has been granted following summary judgment, the new trial motion is in the nature of a motion for reconsideration and, unless the new trial motion raises post summary judgment issues of fact like diligence in obtaining newly discovered evidence (Scott v. Farrar (1983) 139 Cal.App.3d 462, 188 Cal.Rptr. 823), the issue on appeal of the new trial order is whether the summary judgment was properly granted or denied. If summary judgment was properly granted, the grant of a new trial was in error; if summary judgment was not properly granted, the grant of a new trial was correct. (See Green v. Del-Camp Investments, Inc., supra, at p. 483, 14 Cal.Rptr. 420 ["[t]hus the motion for summary judgment was properly granted and the motion for new trial should have been denied"]; Carney v. Simmonds, supra, 49 Cal.2d 84, 98, 315 P.2d 305 [judgment on the pleadings improperly granted and motion for new trial therefore properly granted]; Ramirez v. USAA Casualty Ins. Co. (1991) 234 Cal. App.3d 391, 397, 285 Cal.Rptr. 757 [summary judgment treated as judgment on the pleadings granted in error and therefore motion for new trial properly granted: "[t]he trial court had no discretion to grant a new trial unless its original ruling [judgment on the pleadings], as a matter of law, was erroneous"]; Olson v. County of Sacramento (1969) 274 Cal.App.2d 316, 79 Cal. Rptr. 140 [judgment on the pleadings granted in error and motion for new trial therefore should have been granted]; Malo v. Willis, supra, 126 Cal.App.3d 543, 178 Cal.Rptr. 774 [motion for reconsideration of grant of summary judgment treated as motion for new trial; initial summary judgment erroneously granted and motion for new trial therefore properly granted]; Cf. Scott v. Farrar, supra, 139 Cal.App.3d 462, 188 Cal.Rptr. 823 [original summary judgment properly granted, but newly discovered evidence justified grant of motion for new trial and vacation of summary judgment].)

Here, we perceive no post summary judgment factual issue and therefore, in reviewing the order granting the new trial, we focus on whether the summary judgment in favor of defendants was correct. If our conclusion on that issue is in the affirmative, the order granting plaintiffs' motion for a new trial was in error and not subject to the trial court's discretion. In reviewing an order granting or denying summary judgment, we conduct a de novo review of the issues raised by the supporting and opposing papers. (Buss v. Superior Court (1997) 16 Cal.4th 35, 60, 65 Cal.Rptr.2d 366, 939 P.2d 766.)

VI

APPLICABLE SUMMARY JUDGMENT PRINCIPLES

A. General Principles

We begin with a review of the principles of motions for summary judgment. The purpose of summary judgment is to penetrate pleadings to ascertain, by means of affidavits, the presence or absence of triable issues of material fact. (Molko v. Holy Spirit Assn. (1988) 46 Cal.3d 1092, 1107, 252 Cal.Rptr. 122, 762 P.2d 46.) The trial judge determines whether triable issues exist by examining the affidavits and evidence, including any reasonable inferences that may be drawn from the facts. (People v. Rath Packing Co. (1974) 44 Cal.App.3d 56, 61-64, 118 Cal.Rptr. 438.) In examining the affidavits, those of the moving party are strictly construed and those of the opposing party liberally construed. Doubts as to the propriety of granting the motion are resolved in favor of the party resisting the motion. (Stationers Corp. v. Dun & Bradstreet, Inc. (1965) 62 Cal.2d 412, 417, 42 Cal.Rptr. 449, 398 P.2d 785; Branco v. Kearny Moto Park, Inc. (1995) 37 Cal.App.4th 184, 189, 43 Cal.Rptr.2d 392.)

When the motion for summary judgment is supported by affidavits sufficient to sustain the motion, the burden shifts to the party opposing the motion to show that triable issues of material fact exist. (Chern v. Bank of America (1976) 15 Cal.3d 866, 873, 127 Cal.Rptr. 110, 544 P.2d 1310.) A party cannot avoid summary judgment based on mere speculation and conjecture (Pena v. W.H. Douthitt Steel & Supply Co. (1986) 179 Cal.App.3d 924, 931, 225 Cal.Rptr. 76), and must produce admissible evidence raising a triable issue of material fact. (Craig Corp. v. County of Los Angeles (1975) 51 Cal. App.3d 909, 915, 124 Cal.Rptr. 621.)

Under the summary judgment statute, as revised by the legislative amendments adopted in 1992 and 1993 (see Union Bank v. Superior Court (1995) 31 Cal.App.4th 573, 579-592, 37 Cal.Rptr.2d 653 [legislative amendments to California's summary judgment statute designed to adopt federal burden shifting provisions as interpreted by Celotex Corp. v. Catrett (1986) 477 U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265 to permit defendant to satisfy its burden by pointing to plaintiffs absence of evidence]), a defendant moving for summary judgment can satisfy his burden of showing a prima facie entitlement to summary judgment in at least two ways. The court in Brantley v. Pisaro (1996) 42 Cal.App.4th 1591,1598, 50 Cal.Rptr.2d 431, stated:

"[A] moving defendant now has two means by which to shift the burden of proof under subdivision (o)(2) of section 437c to the plaintiff to produce evidence creating a triable issue of fact. The defendant may rely upon factually insufficient discovery responses by the plaintiff to show that the plaintiff cannot establish an essential element of the cause of action sued upon. [Citation.] Alternatively, the defendant may utilize the tried and true technique of negating (`disproving') an essential element of the plaintiffs cause of action. [Citation.]"

To evaluate whether a moving party is entitled to summary judgment, the trial court must apply a three-step analysis. First, it must identify the issues framed by the pleadings because it is these allegations to which the motion must be directed. Second, it must determine whether the moving party's showing has satisfied the movant's burden of proof and would, if unrebutted, justify a judgment in movant's favor. Finally, if the movant's showing would prima facie justify a judgment in the movant's favor, it must determine whether the opposition demonstrates that a triable, material factual issue exists on the particular issue. (Zuckerman v. Pacific Savings Bank (1986) 187 Cal. App.3d 1394, 1400-1401, 232 Cal.Rptr. 458.)

A moving defendant who chooses the "tried and true technique" of disproving an essential element of the plaintiffs claim must produce evidence that prima facie demonstrates an essential element of the plaintiffs claim cannot be established.^[49] (Allyson v. Department of Transportation, supra, 53 Cal.App.4th at pp. 1317-1318, 62 Cal.Rptr.2d 490; see also Chevron U.S.A, Inc. v. Superior Court (1992) 4 Cal.App.4th 544, 552, 5 Cal.Rptr.2d 674 [defendant's burden to produce sufficient evidence to make out prima facie case of nonliability]; Ahrens v. Superior Court (1988) 197 Cal.App.3d 1134, 1150, 243 Cal.Rptr. 420 [same].) In an antitrust action, that showing can be satisfied by declarations from the alleged conspirators denying the requisite agreement for concerted action or by evidence demonstrating no agreement or conspiracy existed. (Villa v. McFerren (1995) 35 Cal. App.4th 733, 748, 41 Cal.Rptr.2d 719.) It can also be satisfied by demonstrating that under applicable legal principles the plaintiffs' claim cannot be established. (Exxon Corp. v. Superior Court (1997) 51 Cal. App.4th 1672, 1682, 60 Cal.Rptr.2d 195.)

Under the "no evidence" approach, a defendant may satisfy its burden by demonstrating that plaintiff lacks the facts necessary to prove his case. (Leslie G. v. Perry & Associates (1996) 43 Cal. App.4th 472, 482, 50 Cal.Rptr.2d 785; 216 Sutter Bay Associates v. County of Sutter (1997) 58 Cal.App.4th 860, 875, 68 Cal. Rptr.2d 492.) However, the no evidence approach requires more than merely suggesting the possibility that the plaintiff has no facts to support his case; it requires the movant to show plaintiff not only does not have, but cannot reasonably expect to obtain, evidence supporting a prima facie case. (Hagen v. Hickenbottom (1995) 41 Cal.App.4th 168, 186-188, 48 Cal.Rptr.2d 197.) The burden shifts to plaintiff only when, based on the totality of the defendant's submissions, including the plaintiffs discovery responses, and plaintiffs submissions (Villa v. McFerren, supra, 35 Cal. App.4th at p. 751, 41 Cal.Rptr.2d 719), the defendant has shown the plaintiff has no evidence to support his or her claims and will not be able to locate or produce evidence supporting those claims. (Scheiding v. Dinwiddie Construction Co., supra, 69 Cal.App.4th at pp. 74-84, 81 Cal.Rptr.2d 360.)

After the moving defendant has made a prima facie showing under either approach, the burden shifts to the plaintiff to produce competent evidence demonstrating that, on the issues as framed by the pleadings to which defendants' motion is directed, a triable issue of material fact exists:^[50] (FSR Brokerage, Inc. v. Superior Court (1995) 35 Cal. App.4th 69, 73-74, 41 Cal.Rptr.2d 404.) In assessing whether summary judgment is appropriate, the facts submitted by the party opposing summary judgment, together with the reasonable and permissible inferences that may be drawn therefrom, must be accepted as true.^[51] (Sada v. Robert F. Kennedy Medical Center (1997) 56 Cal.App.4th 138, 148, 65 Cal.Rptr.2d 112.)

B. Special Considerations Applicable To Antitrust Actions

The rules generally applicable to summary judgment motions have been refined in antitrust cases.^[52]

In antitrust actions brought under the Cartwright Act, we refer to interpretations of its federal counterpart, the Sherman Antitrust Act (15 U.S.C. ง 1 et seq.), for guidance because the federal and state antitrust laws are parallel in most significant aspects.^[53] (Biljac, supra, 218 Cal. App.3d at pp. 1420-1421, 267 Cal.Rptr. 819.) Plaintiffs' Cartwright Act claim is based on the alleged concerted action by defendants, and it was this element the trial court initially concluded defendants rebutted. Accordingly, to raise a triable issue of material fact, plaintiffs had to offer admissible evidence raising a reasonable inference of agreement or conspiracy.

Both California and federal decisions urge caution in granting a defendant's motion for summary judgment "`in complex antitrust litigation where motive and intent play leading roles, the proof is largely in the hands of the alleged conspirators, and hostile witnesses thicken the plot.'" (Corwin v. Los Angeles Newspaper Service Bureau, Inc. (1971) 4 Cal.3d 842, 852, 94 Cal.Rptr. 785, 484 P.2d 953, quoting Poller v. Columbia Broadcasting (1962) 368 U.S. 464, 473, 82 S.Ct. 486, 7 L.Ed.2d 458.) Because direct evidence of the requisite agreement is rarely available, it may and often must be inferred from circumstantial evidence (Alfred M. Lewis, Inc. v. Warehousemen etc. Local No. 5-2 (1958) 163 Cal.App.2d 771, 778-779, 330 P.2d 53), including the nature of the acts done, the relationship and interests of the parties, and the individual circumstances of each case. (Saxer v. Philip Morris, Inc. (1975) 54 Cal.App.3d 7, 20, 126 Cal. Rptr. 327.)

However, caution is not the same as prohibition, and summary judgment remains available to defendants in an antitrust lawsuit. (Sherman v. Mertz Enterprises (1974) 42 Cal.App.3d 769, 775, 117 Cal.Rptr. 188 ["`this rule of caution should not be allowed to sap the summary judgment procedure of its effectiveness'"]; Biljac, supra, 218 Cal.App.3d 1410, 267 Cal.Rptr. 819 [affirming grant of summary judgment in Cartwright Act conspiracy claim].)

The federal courts have concluded that an important aspect of evaluating a defendant's motion for summary judgment in an antitrust case is the extent to which a plaintiff can avoid summary judgment by relying solely on circumstantial evidence from which to infer the requisite agreement. If the plaintiff cannot show the illegal agreement by direct evidence and relies on inferences from the defendant's overt actions and the impacts of those actions, the plaintiff may not avoid summary judgment by inferring the agreement if the overt actions and impacts are equally explicable by benign competitive motivations or the economic milieu within which the defendant's industry operates as by an illegal agreement.

Numerous federal courts, following the rules articulated in Matsushita Elec. Industrial Co. v. Zenith Radio (1986) 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538, have concluded that although the inferences to be drawn from the facts must be viewed most favorably to the party opposing the motion for summary judgment, there are limits on the range of permissible inferences that may be drawn from ambiguous circumstantial evidence. As developed by the federal courts, this rule provides that conduct "as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy." (Id. at p. 588, 106 S.Ct. 1348.) To survive a motion for summary judgment, the plaintiff in an antitrust action must present evidence "that tends to exclude the possibility" that the alleged conspirators acted independently. In Wallace v. Bank of Bartlett (6th Cir.1995) 55 F.3d 1166, the court summarized the proper approach at pages 1167-1168:

"Summary judgment may be granted even in a complex antitrust case because `antitrust law limits the range of permissible inferences from ambiguous evidence in a [Sherman Act section] 1 case.' [Quoting Matsushita Elec. Industrial Co. v. Zenith Radio, supra, 475 U.S. at p. 588, 106 S.Ct. 1348.] In Riverview Investments, Inc. v. Ottawa Community Improvement Corp., [6th Cir., 1990] 899 F.2d 474 ..., we established a two-part inquiry for evaluating a summary judgment motion in an antitrust conspiracy case: (1) [I]s the plaintiffs evidence of conspiracy ambiguous, i.e., is it as consistent with the defendants' permissible independent interests as with an illegal conspiracy; and, if so, (2) is there any evidence that tends to exclude the possibility that the defendants were pursuing these independent interests. [Id. at p. 483 (citation omitted).] We held that a plaintiff cannot demonstrate a conspiracy `if, using ambiguous evidence, the inference of a conspiracy is less than or equal to an inference of independent action.' [Ibid.]" (Original italics.)

The federal courts have adopted this limitation on permissible inference, citing one or both of two distinct rationales. The first rationale is based on the substantive policy of the antitrust laws to foster competition that leads to more and better goods and services at lower prices for consumers. (See National Society of Professional Engineers v. United States (1978) 435 U.S. 679, 695, 98 S.Ct. 1355, 55 L.Ed.2d 637.) If courts permit a plaintiff to infer a conspiracy from purely legitimate behavior, businesses would be deterred from engaging in the precise behavior antitrust laws wish to foster. (See Matsushita Elec. Industrial Co. v. Zenith Radio, supra, 475 U.S. at pp. 593-594, 106 S.Ct. 1348; Monsanto Co. v. Spray-Rite Service Corp. (1984) 465 U.S. 752, 763-764, 104 S.Ct. 1464, 79 L.Ed.2d 775; Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp. (1993) 509 U.S. 209, 223-226, 113 S.Ct. 2578, 125 L.Ed.2d 168; Rossi v. Standard Roofing, Inc. (3d Cir.1998) 156 F.3d 452; T.W. Elec. Service v. Pacific Elec. Contractors (9th Cir.1987) 809 F.2d 626, 632, fn. 4.)

The second rationale derives from the summary judgment procedures under federal rules as refined by Celotex Corp. v. Catrett, supra, All U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265, Anderson v. Liberty Lobby, Inc. (1986) 477 U.S. 242, 106 S.Ct. 2505, 91 L.Ed.2d 202 and Matsushita Elec. Industrial Co. v. Zenith Radio, supra, 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538. Because the federal rules permit entry of summary judgment against a plaintiff under the "no evidence" approach, summary judgment is properly entered against the plaintiff if his counter-showing is insufficient to establish an essential element of his case on which he has the burden of proof at trial. (Celotex, supra, at p. 322, 106 S.Ct. 2548.) In deciding whether the counter-showing satisfies the burden plaintiff bears at trial, a court ruling on a motion for summary judgment "must be guided by the substantive evidentiary standards that apply to the case." (Anderson, supra, at p. 255, 106 S.Ct. 2505.)

The federal courts have recognized that an antitrust plaintiff who relies on ambiguous circumstantial evidence from which to draw an inference of conspiracy bears the burden of proving the inference of conspiracy more likely than the inference of independent action. The federal courts have reasoned that when the circumstantial evidence is equally consistent with the defendants' permissible independent interests as with an illegal conspiracy, the plaintiff will be unable to satisfy his burden of proof at trial from that evidence alone. Accordingly, to avoid summary judgment the plaintiff must submit evidence that tends to exclude the possibility the defendants were pursuing independent interests; absent that additional evidence, the inference of a conspiracy is less than or equal to an inference of independent action, and will not satisfy plaintiffs trial burden. Absent that additional evidence, the plaintiff lacks evidence to raise a triable issue of material fact and summary judgment for the defendant is proper. (See Richards v. Neilsen Freight Lines (9th Cir.1987) 810 F.2d 898, 903-904; Wilcox v. First Interstate Bank of Oregon, N.A. (9th Cir.1987) 815 F.2d 522, 525, 528; Wallace v. Bank of Bartlett, supra, 55 F.3d at pp. 1167-1168; cf. Riverview Investments v. Ottawa Community Imp. (6th Cir.1990) 899 F.2d 474, 483-485.)

We conclude the federal cases that limit the permissible inferences to be drawn from ambiguous circumstantial evidence are applicable to a Cartwright Act claim because the rationales underlying the federal approach are consonant with existing California law.^[54] To the extent the federal approach reflects substantive antitrust principles, it is fully applicable to Cartwright Act claims. As Biljac explained at pages 1425-1426, 267 Cal.Rptr. 819:

"[The] settled expression of deference to antitrust plaintiffs has been partly undermined in two relatively recent decisions of the United States Supreme Court interpreting the Sherman Antitrust Act. In light of the policy of antitrust law to promote procompetitive, independent business judgment and to penalize only activity which is anticompetitive and concerted, [Matsushita Elec. Industrial Co. v. Zenith Radio, supra, 475 U.S. 574, 106 S.Ct. 1348] announced limits on the range of inferences that may be drawn from ambiguous evidence of conspiracy and has made it clear that those limits apply at the summary judgment stage.... "... An argument can be made that the limits should govern Cartwright Act cases. The same solicitude for lawful, independent business activity undergirds state law policy. (9) As this court has noted, `"Both the Sherman Act and the Cartwright Act proscribe price fixingโ€”the standard for lawful competitive conduct is identical under both state and federal law."` [Citations.]" (Original italics.)

Even assuming the inference limits derive from the summary judgment procedures under federal rules as refined by Celotex Corp. v. Catrett, supra, 477 U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265 and reflect the burdens of proof and persuasion that the plaintiff must satisfy, we nevertheless perceive that the federal approach is consonant with California law. First, California's summary judgment statute was amended to more closely resemble the federal rules. (Union Bank v. Superior Court, supra, 31 Cal.App.4th at pp. 579-592, 37 Cal.Rptr.2d 653.)

Second, the California courts have recognized in other contexts that a plaintiff cannot avoid summary judgment if his counter-showing lacks evidence that would satisfy his burden of proof and persuasion at trial. (Cf. Reader's Digest Assn. v. Superior Court (1984) 37 Cal.3d 244, 252, 208 Cal.Rptr. 137, 690 P.2d 610.) In Leslie G. v. Perry & Associates, supra, 43 Cal.App.4th 472, 50 Cal.Rptr.2d 785, the plaintiff was raped in the parking garage of her apartment building. She sued the landlord, claiming the negligent failure to repair a broken security gate was a proximate cause of the rape because it allowed the rapist access to the garage. The Leslie G. court noted that although the case arose from a summary judgment, the burden of proof on the causation issue was on plaintiff, and under the new version of California's summary judgment statute, a moving defendant need not support his motion with affirmative evidence negating an essential element of the responding party's case, but instead may (through factually vague discovery responses or otherwise) point to the absence of evidence to support the plaintiffs case. "When that is done, the burden shifts to the plaintiff to present evidence showing there is a triable issue of material fact. If the plaintiff is unable to meet her burden of proof regarding an essential element of her case, all other facts are rendered immaterial." (Id. at p. 482, 50 Cal.Rptr.2d 785.) The moving defendants pointed out that because there was no evidence how the rapist entered or left the garage, there was an absence of evidence of causation and it became plaintiffs burden to present evidence showing there was at least a triable issue of material fact of how the rapist entered the garage. (Ibid.)

The Leslie G. court concluded the plaintiffs counter-showing was insufficient to avoid summary judgment because it showed that causation was not and could not be established. The court stated:

"To prevail in a negligence action, the plaintiff must establish every essential element of her case by a preponderance of the evidence. [Citation.] `Preponderance of the evidence' is usually defined in terms of `probability of truth,' for example as evidence that, `"when weighed with that opposed to it, has more convincing force and the greater probability of truth."' [Citations.] In deciding whether a plaintiff has met her burden of proof, we consider both direct and circumstantial evidence, and all reasonable inferences to be drawn from both kinds of evidence, giving full consideration to the negative and affirmative inferences to be drawn from all of the evidence, including that which has been produced by the defendant. [Citations.]

"We will not, however, draw inferences from thin air. Where, as here, the plaintiff seeks to prove an essential element of her case by circumstantial evidence, she cannot recover merely by showing that the inferences she draws from those circumstances are consistent with her theory. Instead, she must show that the inferences favorable to her are more reasonable or probable than those against her. [Citations.] Since there is no direct evidence that the rapist entered or departed through the broken gate (or even that the broken gate was the only way he could have entered or departed), Leslie cannot survive summary judgment simply because it is possible that he might have entered through the broken gate. (Brautigam v. Brooks (1964) 227 Cal.App.2d 547, 556, 38 Cal.Rptr. 784 [an inference cannot be based upon mere possibility]; Estate of Gutierrez' (1961) 189 Cal.App.2d 165, 173, 11 Cal.Rptr. 51 [it is axiomatic that an inference may not be based on suspicion alone, or on imagination, speculation, surmise, conjecture or guesswork]; Reese v. Smith (1937) 9 Cal.2d 324, 328, 70 P.2d 933 [a judgment cannot be based on guesses or conjecture]; Krause v. Apodaca (1960) 186 Cal. App.2d 413, 418, 9 Cal.Rptr. 10 [inferences must be drawn from evidence and not based on mere speculation as to probabilities without evidence].)" (Leslie G. v. Perry & Associates, supra, 43 Cal.App.4th at pp. 482-83, 50 Cal. Rptr.2d 785, original italics.)

The plaintiff argued that her expert's opinion the defective lock provided an opportunity for the rapist to enter raised a triable issue of fact that he entered because the gate lock was broken. The Leslie G. court noted that other modes of entry were equally possible under the circumstances and concluded the "more probable than not" burden of proof placed "`the outer limit[s] of inference upon which an issue may be submitted to the jury'" (Leslie G. v. Perry & Associates, supra, 43 Cal.App.4th at p. 487, 50 Cal. Rptr.2d 785), and that where the "probabilities are at best evenly balanced" the test is not satisfied. Accordingly, summary judgment for defendants was proper. (Id. at p. 484, 50 Cal.Rptr.2d 785.)

We are unpersuaded by plaintiffs' attempts to avoid application of the federal approach. Plaintiffs argue that no California court has adopted the federal approach.^[55] Although no reported California decision has expressly adopted the federal approach, no California case subsequent to the amendments to section 437c has expressly rejected it. Plaintiffs also cite Eastman Kodak Co. v. Image Technical Services, Inc. (1992) 504 U.S. 451, 112 S.Ct. 2072,119 L.Ed.2d 265 as holding that Matsushita's approach is applicable only in limited circumstances that are not present here, and that Matsushita did not invent, but only rearticulated, the burden placed on a plaintiff resisting summary judgment. However, the burden that Matsushita rearticulated was derived from Monsanto Co. v. Spray-Rite Service Corp., supra, 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775, which Eastman Kodak cited with approval. (Eastman Kodak, supra, at p. 468, fn. 14, 112 S.Ct. 2072.) Moreover, federal cases decided after Eastman Kodak have continued to employ the "limits on inferences" approach (see, e.g., Rossi v. Standard Roofing, Inc., supra, 156 F.3d at p. 482; Wallace v. Bank of Bartlett, supra, 55 F.3d at pp. 1167-1168; In re Baby Food Antitrust Litigation (3d Cir.1999) 166 F.3d 112, 124, 138), suggesting that the approach was neither undercut by Eastman Kodak nor limited to unusual cases.^[56]

VII

ANALYSIS OF SUMMARY JUDGMENT RULING

We first consider whether defendants' showings in support of their motions for summary judgment were sufficient to shift the burden to plaintiffs to show an issue of material fact as to the existence of an agreement among defendants. We conclude that the burden did shift to plaintiffs and therefore also examine the evidence submitted by plaintiffs and evaluate (1) whether defendants' conduct is as consistent with permissible independent conduct as with an illegal conspiracy and, if so, (2) whether there is other evidence that makes it more probable than not that the defendants' conduct was the product of an unlawful agreement rather than the product of legitimate, independent business decisions.

A. Shifting the Burden to Plaintiffs

1. Biljac

Plaintiffs argue Biljac requires that before the burden shifts to a plaintiff in an antitrust case, a defendant's motion for summary judgment must include denial declarations from persons responsible for making the decisions. From this predicate, plaintiffs argue that because defendants had not submitted denial declarations from all persons responsible for deciding each defendant's capacity, production and pricing decisions, the trial court erred by shifting the burden to the plaintiffs and granting summary judgment for defendants; therefore, the trial court correctly granted plaintiffs' motion for a new trial. We conclude plaintiffs unnecessarily focus on the identity of the declarants rather than the evidence contained in the denial declarations.

In Biljac, commercial borrowers brought an action under the Cartwright Act and the UCA against certain banks, alleging they conspired for over a decade to move from the traditional prime rate to a new and more volatile base rate that the banks themselves controlled. The banks allegedly agreed to fix, raise, maintain and stabilize the base rate, using it as a benchmark for commercial loans, and thereby suppressed or eliminated competition among themselves. The banks also allegedly conspired to raise and standardize the spread (points charged above the new base rate) for commercial borrowers, to eliminate fixed-rate loans and institute variable-rate loans referenced to the base rate, and to base those rates on fluctuations in short-term commercial paper rates. The conspiracy was allegedly furthered through meetings held under the auspices of industry trade associations. (Biljac, supra, 218 Cal.App.3d at pp. 1416-1417, 267 Cal.Rptr. 819.)

The banks obtained summary judgment. The Biljac court first noted that federal cases are persuasive authority for Cartwright Act claims but that federal summary judgment cases are decided under rules distinct from the California summary judgment statute.^[57] (Biljac, supra, 218 Cal.App.3d at pp. 1420-1422, 267 Cal.Rptr. 819.) It then noted that plaintiffs' claims were based on the allegation of a conspiracy among the banks. (Id. at pp. 1422-1423, 267 Cal.Rptr. 819.) To support their summary judgment motions, the banks presented declarations from the officers and employees charged with setting, changing and publicizing information on their base rates. The banks presented declarations from

"[a]n unbroken chain of officers whose principal responsibility it was to determine the bank's prime rate and commercial loan pricing policies from 1970 into 1986 [that] detailed the criteria they used, the risk factors used to tailor rates for individual borrowers, the reasons for moving from fixed rates to floating rates and the reasons for changing rate criteria over the years. Each denied receiving or sharing competitive information with other banks (except for cooperative efforts such as `participation,' `interbank' or `correspondent' lending), or giving notice of rate changes to others in advance of making them known to the general public. Their declarations explained how market forces influenced rates and the importance of making changes as other banks announced changes in order to compete. In addition, employees responsible for making rate changes public during those years detailed standard procedures used to do this and how the information was received from corporate higher-ups." (Id. at pp. 1423-1424, 267 Cal.Rptr. 819.)

The Biljac court noted the declarations "contain abundant competent and admissible evidence [because] declarants, having established that they alone were responsible for determining rates and that they followed and implemented certain criteria during certain years, were competent to testify how rates and policy were set [citations] and obviously could say whether they had exchanged or agreed to exchange competitive information with other banks. They set out their objectives in attending trade association meetings and denied having agreed with other participants to take collective action or exchange confidential information on commercial rates." (Biljac, supra, 218 Cal.App.3d at pp. 1423-1424, 267 Cal.Rptr. 819.) Finally, the Biljac court noted that declarations were filed showing the bank's conduct was "consistent with independent economic activity." (Id. at p. 1424, 267 Cal.Rptr. 819.) Biljac concluded this showing was sufficient to satisfy defendants' burden under California law^[58] of negating conspiracy and to shift the burden to the plaintiffs to raise a triable issue of material fact.

We believe Biljac supports the trial court's grant of summary judgment for defendants. First, although Biljac concluded the denial declarations from executive officers responsible for implementing the rate system were sufficient to negate plaintiffs' claim of conspiracy, it did not hold that declarations from executive officers that deny collusion were required to negate the element of concerted action. Second, although Biljac involved declarations from the banks' executive officers, Biljac's reference to an "unbroken chain of officers" did not refer to the vertical chain of command within each bank, but to the time periods during which the alleged conspiracy existed.^[59] Finally, Biljac did not require denial declarations from officers with the "ultimate responsibility" for the decisions. (Biljac, supra, 218 Cal.App.3d at p. 1423, 267 Cal.Rptr. 819.) In Biljac the declarants were responsible for the rate setting decisions, and therefore were in a position to be "competent to testify how rates and policy were set." (Id. at p. 1424, 267 Cal.Rptr. 819.) We interpret Biljac to require only that the declarants occupy a position in which they would be competent to deny the existence of an agreement because of their personal knowledge of the process by which the challenged decisions were made.

2. Defendants' Showings to Shift the Burden to Plaintiffs

We conclude that the denial declarations defendants submitted to rebut plaintiffs' claims of an agreement to restrict capacity and CARB gas production and fix CARB gas prices were sufficient to shift the burden to plaintiffs.

a. Production Decisions: Defendants' denial declarations demonstrated that production decisions were made at the refinery level, not at the highest corporate level by the ultimate decision maker for each defendant.^[60] Defendants produced denial declarations from the managers responsible for production decisions that denied their decisions were coordinated with or influenced by any agreement with other defendants but rather were independently based on competitive economic factors of the marketplace supported by extensive documentation. (See Discussion, pt. II.B., ante, and Appendix B.)

b. Pricing Decisions: Defendants' denial declarations demonstrated pricing decisions were made at middle levels rather than at the highest corporate level.^[61] Defendants produced denial declarations from the managers responsible for pricing decisions that their decisions were independently made, detailing the independent factors underlying the decisions, and denying the decisions were influenced by any agreement with other defendants. (See Discussion, pt. II.C, ante, and Appendix C.)

c. Decisions To Enter Exchange Agreements: Defendants' denial declarations also demonstrated decisions to enter exchange agreements were generally the responsibility of middle level managers or employees rather than the chief executive officers.^[62] Defendants produced declarations from the personnel responsible for those decisions that explained how their exchange agreements were made independently based on economic considerations and denied their decisions were influenced by any agreement with other defendants. (See Discussion, pt. II.D., ante, and Appendix D.)

d. Capacity Decisions: We assume that the final decisions on the capacity plans, which required approvals for capital expenditure plans ranging from a low of $100 million (Tosco) to a high of $1.3 billion (Chevron), are ordinarily beyond the authority possessed by even the most senior managers or vice presidents and required approvals at the highest corporate levels. However, we are satisfied the defendants' denial declarations sufficiently established prima facie that the capacity decisions were independently made and not the result of an agreement among defendants.

Defendants' declarations were from persons personally involved, at relatively senior levels, in the process by which each defendant evaluated its capacity decisions. (See Discussion, pt. H.A., ante, and Appendix A.) These declarants explained the defendant-specific business and economic considerations that influenced their recommendations and each defendant's final decision and denied their recommendations tracked an agreement with other defendants to coordinate or limit capacity. Moreover, there is no evidence that more senior officers who were presumed participants to the agreement among defendants in any way constrained the declarants' recommendations, or directed the course of their recommendations, on bases other than each defendant's business or economic considerations.

We conclude defendants' evidence denying any agreement to coordinate capacity, CARB gas production or CARB gas pricing, together with detailed explanations of the independent bases for the decisions and showing plaintiffs' lack of evidence of the existence of any agreement, prima facie demonstrates the agreement element of plaintiffs' Cartwright Act claim cannot be established. Therefore, the burden shifts to plaintiffs to raise a triable issue of material fact on the agreement element of their Cartwright Act claim.

Plaintiffs argue that, even assuming defendants' showing sufficed to negate the agreement element, it was improper to shift the burden to plaintiffs because defendants' showing did not negate the possibility that the effect of the exchange agreements was anti-competitive. However, the allegations of plaintiffs' complaint and the theory articulated in their opposition to the summary judgment motions confirm that plaintiffs sought recovery based on a per se violation of the Cartwright Act arising from the alleged agreement among the defendants.^[63] Plaintiffs' complaint does not allege, and they did not argue until the new trial motion, that the economic effects of an exchange agreement between two defendants offended a "rule of reason" in violation of the Cartwright Act. A defendant need not negate, nor can a plaintiff avoid summary judgment based on, claims that are outside the issues framed by the pleadings; a party cannot gamble on the result of one theory at trial and then obtain a new trial under a different theory if the original theory proves unsuccessful.^[64] (Redwood Turkey Hatchery v. Meadowbrook Farms (1957) 152 Cal.App.2d 481, 486, 313 P.2d 146.)

B. Plaintiffs' Showing

Because we have concluded the summary judgment burden shifted to plaintiffs to demonstrate the existence of a triable issue of material fact on the agreement element of plaintiffs' Cartwright Act claim, we analyze plaintiffs' showing in response to the defendants' summary judgment motions.

1. The Alleged Agreement to Limit Capacity

Plaintiffs' theory is that the required refinery conversion to CARB gas provided defendants with the opportunity to achieve their long-term goal of limiting capacity; in furtherance of that goal, the defendants agreed to (1) coordinate their refinery modification plans, and (2) pool and share their reduced overall capacity through exchange agreements.

Defendants affirmatively demonstrated that each defendant's determination of capacity was based on independent assessments in pursuit of its own economic self-interests; no refinery modification satisfying a defendant's business strategies and investment resources was rejected.

Plaintiffs' counter-showing, from which plaintiffs infer the defendants' capacity decisions were more probably than not the product of an unlawful agreement rather than legitimate business conduct, is based on two general categories of evidence: motive to limit capacity, and opportunity to agree before the refinery modifications.

a. The Shared Motive

The evidence showed, and defendants concede, that defendants shared the common belief that an oversupply of CARB gas was undesirable and therefore had a common motive to restrict capacity. Plaintiffs cite the internal analyses from various defendants and consultants' statements that show it was in each defendant's economic interests to avoid an oversupply of CARB gas that could undermine the ability of each defendant to recoup its refinery modification investment.

However, evidence that each defendant shared a motive to maximize profits to recoup its refinery modification investment is a legally neutral fact. The presence of a shared motive has minimal probative value about whether the defendants agreed to restrict capacity. Even absent an agreement, each defendant is motivated to recoup its investment and has an independent, legitimate business concern that excess supply would jeopardize.^[65] As the court in Serfecz v. Jewel Food Stores (7th Cir.1995) 67 F.3d 591 explained at pages 600-601:

"[I]t is important to note that proof of motive is of limited utility in this context. Although a lack of motive may be evidence that parties did not conspire, the presence of an economic motive is of very little probative value. [Citation.] The mere existence of mutual economic advantage, by itself, does not tend to exclude the possibility of independent, legitimate action and supplies no basis for inferring a conspiracy. [Citation.] We agree with the district court that the mere confluence of economic interests between the parties does not establish, standing alone, the existence of a conspiracy."

b. The Opportunity Evidence

Plaintiffs cite a series of direct contacts between senior executives of ARCO and Chevron and contacts among other defendants as evidence that defendants discussed and agreed to coordinate their refinery modification plans.^[66]

The mere existence of a series of contacts providing the opportunity to make an agreement is, like the evidence of shared motivation, a legally neutral fact. To avoid summary judgment, the plaintiffs' evidence must show more than mere opportunity to conspire. Even "... when evidence shows communications which provided an opportunity for agreement, a plaintiff must still produce evidence permitting an inference that an agreement in fact existed ... [based on] more than speculation." (Alvord-Polk, Inc. v. F. Schumacher & Co. (3d Cir.1994) 37 F.3d 996, 1013; Weit v. Continental Ill. Nat. Bank & Trust Co. (7th Cir.1981) 641 F.2d 457, 462 ["the mere opportunity to conspire, even in the context of parallel business conduct, is not necessarily probative evidence" of conspiracy]; In re Baby Food Antitrust Litigation, supra, 166 F.3d at pp. 126, 133 [same].) The California courts are in accord. (Biljac, supra, 218 Cal.App.3d at p. 1431, 267 Cal.Rptr. 819 [evidence of communications "without disclosin any agreement ... does not bear further discussion"].)

The contacts among defendants cited by plaintiffs do not support the inference that defendants agreed to coordinate refinery modifications to restrict capacity to charge supra-competitive prices.^[67] Plaintiffs principally rely on evidence of a series of meetings between senior officials of ARCO and Chevron during which they explored a possible ARCO/Chevron exchange agreement.^[68] However, this evidence does not support the inference that those meetings produced an agreement to coordinate capacity. First, these meetings did not involve the other defendants with whom an overall coordinated effort to limit capacity would have been essential.^[69] Second, even as to the two participants, the evidence shows these discussions occurred between November 1993 and the summer of 1994, after ARCO and Chevron had made and approved their respective refinery modification plans.^[70]

Finally, the fact that ARCO and Chevron engaged in two-party discussions for an exchange agreement is as consistent with the defendants' permissible independent conduct as with an illegal conspiracy; there is no evidence that makes it more probable than not the meetings were the product of, or the forum for reaching, an unlawful agreement rather than representing legitimate business conduct. It is undisputed that: (1) ARCO needed CARB gas in Northern California and could have produced conventional gasoline for delivery to Chevron; (2) Chevron's production of excess CARB gas would create a corresponding need for conventional gasoline for export; and (3) Chevron sought a premium from ARCO for the CARB gas/conventional gasoline differential. These discussions, which necessarily involved some exchange of information, are as consistent with the defendants' permissible independent conduct as with an illegal conspiracy and do not provide the inference of agreement or collusion.^[71]

Plaintiffs cite ARCO's statement to Chevron that an ARCO/Chevron exchange agreement "could change some of [ARCO's] investment decisions or change investment decisions of others on supplying CARB gasoline" as evidence that capacity coordination was the focus of the exchange agreement negotiations. Again, ARCO's statement is as consistent with the defendants' permissible independent conduct as with an illegal conspiracy. As Tosco's presentation demonstrated, the availability of CARB gas from other sources, including exchange agreements, can and should legitimately affect a company's unilateral investment decisions. If a company can obtain product more cheaply by exchange or purchase agreements than by manufacturing, that choice is economically desirable and efficient; it reduces prices charged to consumers and allows the investment resources to be devoted to areas more profitable for the company. To infer anti-competitive agreements from ostensibly legitimate conduct is precisely the type of "`"mistaken inferences ... [that] are especially costly, because they chill the very conduct the antitrust laws are designed to protect."'" (Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp., supra, 509 U.S. at pp. 223-226, 113 S.Ct. 2578.)

2. The Alleged Agreement To Coordinate Production

Plaintiffs make no effort on appeal to demonstrate that there are triable issues of material fact that defendants agreed to coordinate CARB gas production distinct from their claims that defendants agreed on capacity and CARB gas prices. Plaintiffs' production claim appears to revolve around two contentions: first, by agreeing on capacity, defendants thereby agreed to limit refinery production; second, by agreeing to disseminate and use pricing information, each defendant understood they would be influencing and coordinating refinery production.

Because plaintiffs do not support their claims of production collusion apart from claims of capacity and price agreement, we discuss only the capacity and price claims and deem otherwise abandoned plaintiffs' claim of a triable issue of fact on their CARB gas production collusion claim. (Ochoa v. Pacific Gas & Electric Co. (1998) 61 Cal.App.4th 1480, 1488, 72 Cal.Rptr.2d 232.)

3. The Exchange Agreements

Plaintiffs argue the exchange agreements are direct evidence that defendants, having agreed to coordinate capacity, also agreed to allocate and control CARB gas supply and fix CARB gas prices. Plaintiffs' theory is that the purposes of the exchange agreements were to pool production, establish a common cost of goods, coordinate prices and limit supplies available to independents, who might undermine the price-fixing structure.

We begin our evaluation with several observations concerning plaintiffs' argument. First, there is no direct evidence that defendants agreed to a web of exchange agreements creating a horizontal merger of supplies, or that defendants agreed to use that web for anti-competition purposes. The only direct evidence is that each defendant entered into an exchange agreement with one or more of the other defendants.^[72] However, plaintiffs cite no authority holding that in a multi-competitor industry a two-party agreement is proscribed by antitrust laws, and the law appears to be contrary. (See, e.g., Broadcast Music, Inc. v. CBS (1979) 441 U.S. 1, 9-10, 99 S.Ct. 1551, 60 L.Ed.2d 1.)

Second, defendants' evidence showed that each two-party exchange agreement is as consistent with the defendants' permissible independent conduct as with an illegal conspiracy.^[73] Plaintiffs cite no direct evidence to support the conclusion that any one of the two-party exchange agreements was more probably than not an unlawful agreement rather than a legitimate business agreement.

Finally, courts have long recognized that two-party exchange agreements in the petroleum industry have both efficiency-enhancing purposes, and pro-competitive effects. They avoid the necessity and cost for redundant refinery operations in each area served by a refiner's marketing arm (see American Oil Company v. McMullin (10th Cir.1975) 508 F.2d 1345, 1353; Thomas v. Amerada Hess Corporation (M.D.Penn.1975) 393 F.Supp. 58, 74) and provide a company with product at competitive cost, allowing the company to compete in its competitor's geographic area (see Marathon Oil Company v. Mobil Corporation (N.D.Ohio, 1981) 530 F.Supp. 315, 321, fn. 9).^[74] For these reasons, the courts have consistently refused to conclude that a two-party exchange agreement, without more, is a restraint of trade constituting a per se violation of antitrust laws. (Blue Bell Co. v. Frontier Refining Co. (10th Cir.1954) 213 F.2d 354, 359.)

Because each two-party exchange agreement is as consistent with the defendants' permissible independent conduct as with an illegal conspiracy, drawing an inference of an anti-competitive agreement based solely on that ostensibly legitimate conduct is the type of "`"mistaken [inference] ... [that is] especially costly, because [it] chill[s] the very conduct the antitrust laws are designed to protect."'" (Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp., supra, 509 U.S. at p. 226, 113 S.Ct. 2578.) However, we examine whether plaintiffs presented evidence to support an inference that any of the two-party exchange agreements were more probably than not the products of unlawful agreements rather than legitimate business transactions.

Plaintiffs allege the exchange agreements were unorthodox and therefore conspiratorial; however, the evidence does not support the assertion that the exchange agreements were unorthodox. Plaintiffs also allege the defendants made exchange agreements not for the independent benefits they provided but to control supply and keep CARB gas from independents.^[75] However, the evidence shows some defendants continued selling to independents, and the inference that defendants were motivated by something other than their independent business interests is speculative.

Plaintiffs also argue the exchange agreements established a common cost of goods, and established a coordinating mechanism to control and elevate prices. Plaintiffs' theory is that defendants used exchange agreements to elevate and fix the price of CARB gas. First, plaintiffs argue the evidence permits the inference that defendants used the price differential contained in the exchange agreements to establish the wholesale price for CARB gas, which in turn established the base price from which rack and DTW rates were established. Second, plaintiffs argue the evidence permits the inference that defendants used the OPIS spot price in the exchange agreements as the floating price for transactions occurring in the future, and thus controlled and fixed the price for a vast number of transactions that were based on the spot prices; the defendants had the ability to manipulate and control the spot price because of the thin market. The former claim is based on two exchange agreements, and the latter claim is based on Noll's opinion and theory.

As evidence of their first claim, plaintiffs refer to the Chevron/Tosco and the Chevron/Shell exchange agreements Under which CARB gas was exchanged for conventional gasoline. The party that delivered conventional gasoline and received higher cost CARB gas was required to pay a price differential, a portion of which was fixed and a portion of which was based on the OPIS spot market price.^[76] Plaintiffs cite Chevron's statement that it believed the "fixed CARB gas/conventional gas differential [of the Chevron/Tosco agreement] may help establish, more or less, the market differential," as proof the exchange agreements were designed to fix the price of CARB gas. However, plaintiffs ignore that Chevron's Products Division recommended approval of the Tosco agreement because the formula under the Chevron/Tosco agreement would yield "about 9-11 cpg depending on market conditions [and] [w]e expect the average market differential will be similar." This evidence shows, at most, that Tosco and Chevron entered into a two-party exchange agreement to exchange products for which some price differential was essential and that each party necessarily estimated the differential and hoped its estimate would be economically advantageous. Plaintiffs do not claim that a differential was unnecessary, or that some more definitive model was available to calculate the differential, or that at the time the exchange agreement was negotiated either party agreed to a differential that was contrary to its own unilateral economic self-interests under the facts known at that time. Because the formula to account for and predict the price differential was as consistent with these defendants' permissible independent interests and conduct as with an illegal conspiracy, and there is no evidence that makes it more probable than not the differential was the product of an unlawful agreement rather than representing legitimate business conduct, a conspiracy cannot be inferred from the price differential formula.^[77]

Plaintiffs' second claim is based on Noll's theory that defendants erected an institutional structure that included the exchange agreements, which necessarily resulted in anti-competitive and collusive conduct. Noll begins with the fact the spot market is concentrated and thin in nature, and any single spot transaction, even for a small amount, has a disproportionate spill-over effect on other contract prices that are tied to the OPIS spot price. Noll averred that by linking the price in exchange agreements to the spot price, which defendants can control through ostensibly separate spot transactions, defendants fix the market price for CARB gas through a few spot trades. Noll apparently opined that defendants, who have a natural motivation to charge artificially high prices to the public, had the incentive and ability to buy and sell CARB gas at artificially high prices to drive up the cost of CARB gas in all "linked" transactions, and that this elevated price was then employed by defendants to determine the DTW and rack prices charged to retailers and passed on to consumers. Noll's theory, in essence, asserts that because of the institutional structures within which the trading and pricing activities must operate, it is impossible for defendants to act independently or non-collusively. Noll states that prices for major portions of the supply are influenced by spot transactions through the exchange agreements, and some prices for large transactions are based on a blend of a fixed and a floating differential; the cost of those goods is thereby fixed and influences the DTW and rack prices of those goods, which in turn influence a competitor's DTW and rack pricing.

Plaintiffs argue that Noll's opinion was unrebutted and that under Buttram v. Owens-Corning Fiberglas Corp. (1997) 16 Cal.4th 520, 526, 66 Cal. Rptr.2d 438, 941 P.2d 71, unrebutted expert testimony will alone suffice to raise a triable issue of fact.^[78] However, Buttram did not raise, much less analyze, the effect of unrebutted expert testimony on a motion for summary judgment. More importantly, in the context of antitrust lawsuits the courts have observed that although "[e]xpert testimony [may be] useful as a guide to interpreting market facts, ... it is not a substitute for [facts]." (Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp., supra, 509 U.S. at p. 242, 113 S.Ct. 2578.) An expert's unrebutted opinion does not preclude summary judgment if it is based on unsound reasoning (Kelley v. Trunk (1998) 66 Cal.App.4th 519, 523-524, 78 Cal.Rptr.2d 122) or on speculation and conjecture. (Exxon Corp. v. Superior Court, supra, 51 Cal.App.4th at p. 1683, 60 Cal.Rptr.2d 195; see also Leslie G. v. Perry & Associates, supra, 43 Cal.App.4th at p. 487, 50 Cal.Rptr.2d 785.) When a plaintiff must prove that certain conduct occurred, an expert who opines that the observable facts could have been produced by the conduct but has no basis for concluding the conduct occurred will not preclude summary judgment for the defendant. (Bay Area Rapid Transit Dist. v. Superior Court (1996) 46 Cal.App.4th 476, 482, 53 Cal.Rptr.2d 906.)

We are unpersuaded that Noll's opinion sufficed to raise a triable issue of fact in this case. Although Noll opined the use of the OPIS spot price in many of the exchange agreements resulted in a structure that necessarily precludes independent conduct, he did not opine and there is no evidence that defendants agreed to create this structure as a price-fixing mechanism.^[79] The deficiency in Noll's testimony is similar to the deficiency in the plaintiffs' expert testimony in Wallace v. Bank of Bartlett, supra, 55 F.3d 1166. In Wallace, the plaintiffs alleged an agreement among banks to fix the price for NSF (not sufficient funds) checks at above the actual cost to the banks. The banks offered proof of their legitimate independent business reasons for charging excessive NSF fees; plaintiffs' expert opined there must have been an agreement. The Wallace court upheld the grant of summary judgment for the banks, notwithstanding plaintiffs' expert, and stated at page 1170:

"Plaintiffs offer two affidavits in support of their allegations that the Banks have tacitly conspired to fix prices. First, they have submitted the affidavit of Allen Buckalew, an economist [who] reviewed the Banks' depositions, [and] pricing history and a presumed cost per NSF check of $.15 to $1.50 and stated that:

"`Assuming all of the above, and further assuming that defendants have historically totally failed to openly compete in the market for customers on the basis of more attractive NSF charges, I can testify that based on well-accepted economic theories dealing with supply and demand in our American economy, defendants have been able to maintain these artificially excessive NSF/DIR charges by means of a tacit agreement not to compete on such charges and to set the prices at the inflated amounts which they presently are.'

"The Banks contend that this affidavit fails to raise a genuine issue of material fact because the expert offers no additional facts or evidence and relies on conclusory allegations.

"Buckalew's affidavit is not sufficient to defeat the Banks' motion for summary judgment. In [Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., supra], the Supreme Court held that an economist's expert testimony which was based on insufficient facts could not be relied on to support a jury verdict.... The Court stated that `[e]xpert testimony is useful as a guide to interpreting market facts, but is not a substitute for them.' [Ibid.] Buckalew's conclusions are based on nothing more than facts which, as we have already stated, are equally consistent with independent actions by the Banks. For instance, the Banks readily admit that they have `totally failed to openly compete' because they do not want to attract customers who write checks with insufficient funds to cover them.

"Plaintiffs submitted another affidavit from [expert] Zate.... According to Zate, the banking industry's income derived from interest has been declining since the early 1980s while income derived from `non-interest income sources' such as NSF fees has been increasing. It is Zate's `educated and professionally-gained opinion' that the `enormous profit' from NSF fees would not have been possible without collusion between banks. Zate's opinion, however, does not exclude the possibility that the Banks wanted to keep these fees high to discourage NSF check-writing or that, by increasing NSF fees to compensate for lost interest revenue, the Banks were maintaining the safety and soundness of the institution as permitted by [law]. Plaintiffs' expert evidence is equally consistent with independent action as it is with conspiracy."

A plaintiff cannot avoid summary judgment by introducing an expert's opinion that speculates from observable market facts that the conduct was the product of an agreement, when the conduct is equally consistent with independent action. Noll did not opine or use as a basis for an opinion that any of the exchange agreements were not as equally consistent with independent action of the two-party participants as with a conspiracy.^[80]

4. The Alleged Agreement to Fix Prices

Plaintiffs cite as evidence to support their inference defendants agreed to fix the DTW and rack prices for CARB gas at supra-competitive levels that defendants: (1) agreed to use in their exchange agreements the OPIS spot price, which they could manipulate through spot trading, as the basis for DTW and rack prices; (2) jointly preplanned price increases before CARB gas was introduced, implemented those increases when CARB gas was introduced, and gave false explanations for the price increases; (3) engaged in parallel pricing for CARB gas; and (4) disseminated pricing information through third parties to coordinate prices. From this evidence, plaintiffs claim that ostensibly independent pricing decisions were more consistent with a conspiracy than independent action.

a. Use of Manipulated OPIS Price to Elevate and Fix Prices

Plaintiffs' first theory, which is premised on Noll's opinion, is that defendants agreed to use the OPIS spot price as the common cost of goods throughout the industry, manipulated and artificially elevated that price through control of the spot trading market, and then used that elevated OPIS spot price as the basis for their DTW and rack prices. We are unpersuaded by this argument.

First, there is no direct evidence of an agreement to use OPIS spot prices as the common cost of goods throughout the industry. Second, to the extent Noll's opinion is premised on the assumption all parties use OPIS spot prices as the common cost of goods, it is unsupported by the evidence: some of the exchange agreements refer to the OPIS spot price, and others, including Chevron/Tosco and Chevron/Shell, use a blended price. Plaintiffs make no effort to support their assertion that all of the exchange agreements are uniform in using the OPIS spot price.^[81] Third, plaintiffs' theory relies on the assumption that defendants had the incentive and the ability through control of the spot trading market to manipulate and artificially elevate the OPIS spot price. However, the evidence showed both assumptions were incorrect. The evidence showed the OPIS spot price is derived from numerous spot trades reported from twenty-five different sources in the supply and trading community, only two or three of which are defendants. Although defendants may participate as sellers in many spot transactions, plaintiffs' theory requires as a corollary that numerous third parties are willing to purchase at noncompetitive prices to facilitate defendants' artificial elevation of the spot price.^[82] Furthermore, defendants had no common incentive to artificially raise the spot price. Noll conceded that buyers want the lowest price possible while sellers want the highest price possible. Some defendants are net buyers and others are net sellers. Noll agreed that the "incentives of a net buyer differ from the incentives of a net seller."^[83]

Finally, the assertion that defendants used the elevated OPIS spot price as the basis to set their DTW and rack prices is not supported by the evidence. Noll assumed that DTW and rack prices are "over periods of time ... based on target markup rates over some estimate of what the market value of wholesale product is, and that [wholesale value] is based on OPIS...." Plaintiffs cite no evidence to support that assumption. Plaintiffs cite no evidence that any defendant determines its DTW and rack prices by marking up its wholesale cost, or that most defendants base their wholesale value on the OPIS spot price. Defendants instead explained how their rack and DTW prices are calculated: they do not index or coordinate those prices to the OPIS spot price; many of the defendants do not set their prices with reference to the OPIS spot price at all; and other defendants consider the OPIS spot price only as one of many factors in pricing. Finally, Noll admitted that although the OPIS spot price is taken into account in pricing decisions, the primary factors determining prices are market conditions, including the retail and DTW prices of the companies against whom a defendant competes, a defendant's cost of production and its desired margin.

Because Noll's opinion is based on unproven facts and relies on assumptions contrary to rational economic behavior, we conclude Noll's opinion does not permit an inference that ostensibly independent pricing decisions were not as equally consistent with independent action by defendants as the product of a conspiracy.

b. Planning for and Increasing Price

Plaintiffs argue defendants' agreement to fix CARB gas prices can be inferred from their conduct of jointly preplanning price increases before CARB gas was introduced, implementing those increases when CARB gas was introduced, and giving false explanations for the price increases.

There is no evidence that defendants jointly planned their price increases.^[84] Plaintiffs' evidence shows that executives of various defendants (Tosco's O'Malley in 1994 and Ultramar's Westfall in 1995) publicly predicted that CARB gas would command higher prices and could spike in the event of unplanned disruptions of refiners. However, plaintiffs do not suggest these predictions were unconnected to the increased capital expenditures and production costs reasonably anticipated by defendants or the market forces that would operate in the event of refinery shutdowns.^[85] Plaintiffs' evidence also showed that defendants unilaterally or through trade associations prepared a public relations campaign to explain the causes of the increased gas price. However, proof that defendants acted jointly to improve their public image is not probative of plaintiffs' claim that defendants acted jointly to fix prices.

The evidence shows defendants increased prices. However, the fact of price increases is a legally neutral fact. Price increases are as consistent with permissible independent conduct as with an illegal conspiracy, and there is no evidence that makes it more probable than not that the price increases were the product of an unlawful agreement rather than legitimate business conduct. As the court explained in In re Coordinated Pretrial Proceedings in Petroleum Products Antitrust Litigation (9th Cir.1990) 906 F.2d 432, 445 (Petroleum Products), a court will not infer conspiracy based solely on a claim that the price hikes were too high:

"Were we to permit such a theory, few pricing decisions would be immune from antitrust scrutiny. Any unhappy buyer in a market experiencing a general increase in prices could bring an antitrust action against the sellers, arguing that the conspiratorial nature of the price increase is indicated by the fact that the increase was `too high.' The court would then be required to receive evidence as to the wisdom of, and necessity for, the price increase, and to judge what constitutes a `fair' price.... Accordingly, the appellants must rely on additional evidence beyond mere price parallelism in order to avoid summary judgment."

Finally, plaintiffs claim defendants conducted a misinformation campaign to justify the price increases, permitting an inference defendants sought to hide the real, illegal reason for the price increase. However, plaintiffs' claim of a misinformation campaign lacks evidentiary support.^[86]

c. Parallel Pricing for CARB Gas

Plaintiffs suggest the alleged existence of parallel price increases for CARB gas permits an inference of an agreement among defendants. First, plaintiffs on appeal cite no evidence of parallel price increases, permitting us to deem this claim abandoned. (Ochoa v. Pacific Gas & Electric Co., supra, 61 Cal.App.4th at p. 1488, 72 Cal.Rptr.2d 232.)

More importantly, assuming the price increases by a defendant were parallel to price increases of other defendants, parallel pricing is a legally neutral fact. Plaintiffs allege, and defendants do not dispute, that the California CARB gas market is oligopolistic.^[87] It is well established, however, that "[t]he mere existence of an oligopolistic market structure in which a small group of manufacturers engage in consciously parallel pricing of an identical product does not violate the antitrust laws." (E.I. Du Pont de Nemours & Co. v. F.T.C. (2d Cir.1984) 729 F.2d 128, 139; see also Market Force Inc. v. Wauwatosa Realty Co. (7th Cir.1990) 906 F.2d 1167, 1172 ["conscious parallelism by itself is not enough to support an antitrust conspiracy case"]; Weit v. Continental 111. Nat. Bank & Trust Co., supra, 641 F.2d 457, 463 ["Courts have noted that parallel pricing or conduct lacks probative significance when the product in question is standardized or fungible."].)

Because parallel pricing in an oligopolistic industry is equally consistent with independent action as with the product of a conspiracy, it alone does not permit an inference of the requisite agreement. (In re Baby Food Antitrust Litigation, supra, 166 F.3d at p. 122.) The reason parallel pricing, standing alone, lacks probative value of an agreement was explained by the court in Clamp-All Corp. v. Cast Iron Soil Pipe Institute (1st Cir.1988) 851 F.2d 478. The Clamp-All court stated that a showing of parallel pricing

"... may reflect no more than an industry concentrated enough for each firm to set prices `interdependently' (each firm, aware that competitors will quickly match price cuts, may keep its prices high)....

"........................

"... A firm in a concentrated industry typically has reason to decide (individually) to copy an industry leader. After all, a higher-than-leader's price might lead a customer to buy elsewhere, while a lower-than-leader's price might simply lead competitors to match the lower price, reducing profits for all. One does not need an agreement to bring about this kind of follow-the-leader effect in a concentrated industry. [Citation.] ... "[Although] the Sherman Act prohibits agreements, [the courts] have almost uniformly held, at least in the pricing area, that such individual pricing decisions (even when each firm rests its own decision upon its belief that competitors will do the same) do not constitute an unlawful agreement under section 1 of the Sherman Act. [Citations.]" (Id. at pp. 483-184, original italics.)

California is in accord with the federal courts. (Biljac, supra, 218 Cal.App.3d at pp. 1428-1429, 267 Cal.Rptr. 819 ["Parallel movement of [prices] was not enough by itself to support inference of agreement because "market forces were as likely as not the reason for the complained-of parallel prices...."].)

Plaintiffs rely heavily on Petroleum Products to support their argument that the evidence permits an inference defendants' parallel price increases were the product of an agreement to fix prices. Because much of plaintiffs' argument relies on Petroleum Products, we examine it in detail.

The factual context of Petroleum Products is important to an understanding of the extent of, and limits on, its application here. The Petroleum Products court examined an alleged agreement to raise and stabilize prices at the DTW price level. The court noted that the "official" DTW price only occasionally changed, and that the prevalent way of changing the actual DTW price, and thus retail prices, was through adjusting the discounts given to dealers from the nominal DTW price. (Petroleum Products, supra, 906 F.2d at pp. 436-37.) Eliminating or reducing the discounts had the effect of raising actual DTW prices and restoring or increasing the discounts had the effect of lowering actual DTW prices. The court also noted that in a concentrated market in which price changes can be quickly reversed without significant losses to the price leader, the possibility parallel price movements are caused by independent pricing rather than by agreement increases because: (1) a price leader may be willing to "test the waters" with a unilateral price hike absent an advance commitment from competitors to follow because he can quickly detect whether others are following and, if not, can reverse course with absorbable losses; and (2) price followers know they can reap profits if they follow suit, but because their price movements can be detected, a price leader will quickly return to lower prices and eliminate the followers' opportunity to increase profits if his lead is not followed. Accordingly, the court refused to infer the requisite agreement solely from mere parallel price movements in an oligopolistic market. (Id. at pp. 443-445.)

However, the Petroleum Products court did find other direct evidence that combined with the parallel price movements permitted an inference the companies had an express or tacit commitment to coordinate their prices. First, the companies engaged in a practice of publicly announcing, sometimes in advance of the effective dates, withdrawal of their discounts. The companies admitted there was no business purpose for public announcements of future action other than to signal price increases to competitors "in the express hope that these competitors would follow their move." The announcements facilitated collusion in price increases by ensuring that (1) a price leader could change course if his competitors did not announce an intent to follow, and (2) assured that a price leader's move was known so that he would not suffer substantial losses during the time it took competitors to detect his move and change their retail prices. Thus, the evidence that the sole purpose and effect of announcements of future prices was to facilitate coordinated price hikes, coupled with actual parallel price increases, permitted an inference of an express or tacit agreement to conceitedly raise and stabilize prices. (Petroleum Products, supra, 906 F.2d at pp. 445-446.)

Other evidence cited by Petroleum Products was the public posting of actual discounts for each zone, the admitted purpose of which was to inform competitors of any price changes. (Petroleum Products, supra, 906 F.2d at p. 449.) The court concluded public posting had no business purpose other than to facilitate price matching. The court also noted that the posted discounts provided unusually detailed information for each price zone. The court reasoned that because disclosure of that confidential price information appeared contrary to a company's self-interests, and created a mechanism where "shading" could be detected and discouraged, a jury could infer the defendants understood and intended to discourage such "shading" to stabilize price increases. (Id. at pp. 449-450.)

In addition, Petroleum Products deemed significant the direct contacts among the defendants to exchange pricing information. The court noted that prior to the decision in United States v. Container Corp. of America (1969) 393 U.S. 333, 89 S.Ct. 510, 21 L.Ed.2d 526 (Container), in which the Supreme Court concluded an informal agreement to directly exchange price information could, under appropriate market conditions, constitute circumstantial evidence of an agreement to fix prices (id. at pp. 336-337, 89 S.Ct. 510), the defendants had widely engaged in somewhat secretive exchanges in which they traded their DTW price and discount information with competitors. The Petroleum Products court concluded that, although the companies had purportedly eliminated such contacts post-Container, the evidence that they had originally used such contacts and then switched to public announcements would permit an inference the public channels were merely a substitute for the banned private exchanges and were designed to achieve the same goal of coordinated pricing. (Petroleum Products, supra, 906 F.2d at pp. 450-53.) Moreover, the court pointed out that there was evidence that direct contacts involving price information exchanges, although used less frequently, had not ceased to be used post-Container. (Id. at pp. 453-455.)

Plaintiffs allege defendants engaged in the same conduct condemned by the Container court and found sufficient by the Petroleum Products court to withstand summary judgment. Plaintiffs allege defendants here disseminated current and future price information through third parties to coordinate pricing. It is undisputed defendants obtained information from Lundberg and other third party services on the DTW and street prices of their competitors, and that some defendants used this information as a factor to calculate their own prices.^[88] However, this case does not otherwise resemble Petroleum Products.

First, there is no evidence that Lundberg, or any other service, reports the wholesale prices that defendants will charge in the future. Second, plaintiffs cite no evidence that Lundberg or any other service reports the actual rack or DTW prices, which are the rack/DTW prices less applicable discounts, charged by each defendant; the evidence suggests the reports contain only historical reports of the average or aggregate rack/DTW prices without reporting discounts or rebates. It was Petroleum Products' conclusion that the release of unusually detailed actual pricing information by a defendant had no legitimate business purpose but could be used to detect or reveal price shifts (Petroleum Products, supra, 906 F.2d at pp. 450-453); here, however, actual pricing information is not disseminated and a defendant's actual price change is obscured. Third, plaintiffs cite no evidence defendants exchange price information during their direct contacts.

Finally, and perhaps most significantly, defendants do not disseminate pricing information; instead, Lundberg obtains its information on the rack and DTW prices being charged by defendants from the third party jobbers and retailers who are paying those charges. A core aspect of the Petroleum Products reasoning was that announcing the actual or intended elimination of discounts served little purpose other than to facilitate price collusion, and inferring a conspiracy from the practice would not chill conduct serving a legitimate business purpose. The dealers and jobbers whose discounts were being eliminated did not need to learn through public announcements because they would learn of the action directly from the defendants. (Petroleum Products, supra, 906 F.2d at pp. 446-448.) Here, however, inferring a conspiracy based on Lundberg would chill legitimate business conduct because it is the dealers and jobbers, who pay the DTW and rack prices, to whom pricing information must be revealed and who report the information to the survey.^[89] Indeed, plaintiffs do not suggest how defendants would be able to conduct business without telling the independent dealers and jobbers the prices they are being charged, and cite no authority that permits an inference of conspiracy to be drawn from the conduct of independent third parties over whom defendants have no control.

We conclude that the rationale of Petroleum Products is inapplicable to the conduct relied on by plaintiffs to infer a conspiracy among defendants.^[90] We also conclude that defendants' conduct cited by plaintiffs is as equally consistent with independent action by defendants as by a conspiracy, and that to infer a conspiracy from the cited conduct would chill legitimate business conduct.

d. The Exchange of Information

Plaintiffs cite defendants' use of common consultants and defendants' participation in trade association activities as evidence from which to draw an inference that defendants agreed to coordinate their activities by exchanging information.

(1) Trade Association Activity

Plaintiffs argue the methods by which defendants facilitated their collusive coordination of capacity and price included exchanging information during trade association activities. However, defendants' participation in trade associations is a legitimate activity under antitrust laws, even when competitive information is discussed during those meetings. There is a "strong policy that disseminating competitive information, even pricing information, through trade association activities is condoned by antitrust law. Members may use that information in the management and control of their individual businesses. It is only where they take concerted action to restrain trade based on the information that the law is violated. [Citing Maple Flooring Mfrs.' Ass'n v. United States (1925) 268 U.S. 563, 585, 45 S.Ct. 578, 69 L.Ed. 1093.] Thus, plaintiffs must show conspiracy with something beyond [participation in trade association] activities." (Biljac, supra, 218 Cal. App.3d at pp. 1432-1433, 267 Cal.Rptr. 819.) Accordingly, evidence defendants communicated during trade association meetings that provided an opportunity for agreement, without evidence permitting an inference that an agreement in fact existed, is insufficient. (Alvord-Polk, Inc. v. F. Schumacher & Co., supra, 37 F.3d at p. 1013.)

Here, defendants explained the legitimate and non-conspiratorial purposes for their trade association activities. The totality of plaintiffs' showing was Bloom's statement that "there is also activity that takes place in trade associations that amounts to information exchange."^[91] This conclusory testimony, with no evidence of the content of those discussions or the suspicious nature of the informational exchange, is insufficient to avoid summary judgment.

(2) Common Consultants

Plaintiffs finally rely on the "unrebutted" evidence that defendants used common consultants to funnel confidential competitive information to each other as evidence of the agreement to coordinate their capacity and CARB gas pricing. We are not persuaded by this claim and distinguish between what plaintiffs showed and what plaintiffs claim they showed. Plaintiffs showed that defendants, in developing or evaluating their individual capacity decisions, employed one or more industry consulting firms to assist them and, through both internal studies and consultants, obtained information about the plans and progress of the other defendants. However, for the reasons previously discussed in detail (see pt. III.C, ante, and Appendices E & F) it is only speculation that confidential competitive information was transferred from one defendant to another through the consultants or that defendants agreed to use consultants to coordinate their capacity decisions.

Because the law permits and encourages a company to gather information to make rational economic decisions (Biljac, supra, 218 Cal.App.3d at pp. 1430-1433, 267 Cal. Rptr. 819), plaintiffs' attempt to infer an unlawful conspiracy from employing experts is precisely the type of "`"mistaken [inference] ... [that is] especially costly, because [it] chill[s] the very conduct the antitrust laws are designed to protect."`" (Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp., supra, 509 U.S. at pp. 223-226,113 S.Ct. 2578.)

C. Conclusion

We conclude that defendants' showings shifted the summary judgment burden to plaintiffs and that plaintiffs' counter-showing did not create a triable issue of material fact of an agreement in violation of the Cartwright Act. These conclusions accord with the conclusions of the trial court in its original ruling on the summary judgment motions. The trial court stated:

"As a result of the announcement [of new CARB regulations], California would become a truly unique limited gasoline market.... Billions of dollars were invested overall with widely disparate capital expenditures to convert refineries to produce the new gas.... The industry experienced profound change. No one disputes that gasoline supply was tight and prices rose dramatically.

"...................

"Having examined the evidence [separately and] in the aggregate ..., the only logical inference which can be drawn ... is that Defendants' actions were a pro-competitive response to a regulatory requirement which forced members of an oligopoly to restructure their product mix and incur substantial additional capital expenditures.

".....................

"Plaintiffs have attempted to weave a thread of conspiracy through a wide range of activities engaged in by Defendants and argue that through these activities Defendants have created an institution by which pricing decisions are coordinated. Plaintiffs' experts opine as to evidence they `have seen,' but their opinions are unencumbered by facts, reasons, or explanations as to how the evidence supports their theory. [Plaintiffs' arguments] suggest a grand scheme furthered by a cadre of personnel throughout Defendant corporations and throughout the gasoline distribution networkโ€”a scheme furthered by common consultants and allegedly `not-so-independent' pricing services[โ€”]by which Defendants have been able to raise and stabilize retail prices. But the evidence Plaintiffs provided to the Court suggests not a complex, tangled web, but nine defendants using all available information sources to determine capacity, supply, and pricing decisions which would maximize their own individual profitsโ€”without regard to the profits of their competitors. While the evidence suggests that each defendant searched out and obtained as much information as it could on its competitors' plans, such a search does not offend antitrust policies and cannot form the basis for antitrust liability without evidence that those plans were coordinated to restrict trade. Plaintiffs' evidence does not support even the inference of such an agreement." (Original underscore.)

We agree with the exhaustive and well reasoned analysis of the trial court. Defendants are entitled to summary judgment.^[92]

DISPOSITION

The order granting a new trial is reversed and the trial court is directed to grant summary judgment in favor of each defendant. The award of costs to defendants by the trial court is affirmed. Defendants are entitled to costs on appeal.

WORK, Acting P.J., and HUFFMAN, J., concur.

APPENDIX A

Mr. Waldinger, Mobil's project engineer responsible for capital projects at Mobil's Torrance refinery, testified the Torrance refinery could have produced only 25,000 BPD without refinery modifications. This capacity was less than Mobil's needs. Waldinger participated in generating the alternatives for modifying the Torrance refinery. The alternative chosen was to invest $126 million to create a 88,000 BPD capacity to meet Mobil's California needs. Alternatives to increase that capacity were substantially more costly. Waldinger denied that the capacity recommendations and decisions resulted from agreement or collusion with others. Internal Mobil documents showed the recommended refinery modifications would "enhance[] Torrance's competitive position" by providing a greater capacity for a lower investment than Mobil's competitors would likely be able to obtain.

Mr. Kiracofe, an ARCO senior vice president responsible for overseeing ARCO's Los Angeles refinery, averred he was responsible for developing the CARB gas production plans for ARCO's refineries, and the decision to convert its Los Angeles refinery to create a 150,000 BPD capacity was made independently and without consultation or discussions with other defendants. ARCO's long-standing competitive strategy, which was unchanged by introduction of CARB gas, was to sell larger volumes at lower prices than its competitors; ARCO sought to continue this strategy by creating CARB gas capacity at the Los Angeles refinery at the lowest capital expense.

Mr. Scott, Chevron's refining department engineer responsible for developing refinery modifications to produce CARB gas, averred that Chevron's refineries could not meet the new CARB gas regulations without substantial capital expenditures, and that Chevron considered many alternatives, including leaving the California market, to cope with the new regulations. He averred that Chevron's decision to invest over $1.3 billion to convert its refineries to CARB gas production was made without consultation or agreement with any other defendant. Mr. MacDonald, a planning consultant in long-term strategic planning for Chevron, explained that Chevron sought to gather information from numerous sources to assess the additional costs Chevron's competitors would incur to produce CARB gas. This information was used by Chevron to evaluate whether Chevron's strategy of converting to maximum CARB gas capacity and costs of that strategy would place Chevron at a competitive advantage or disadvantage. Chevron's declarants averred that Chevron chose a strategy of massive investments because it made Chevron the largest producer of CARB gas in the world and gave it flexibility in its refinery operations. Chevron's decision to produce more CARB gas than it needed for branded sales was also in Chevron's self-interests because it (1) allowed Chevron to build a contingency inventory that would permit continued supplies during planned or unplanned refinery outages, and (2) prevented severe shortages in CARB gas that might cause California to repeal some or all of the California Air Resources Board's regulations, thereby imperiling Chevron's investment. Chevron's capacity decisions were not influenced by its competitors' plans.

Mr. Westfall, Ultramar's Director of Strategic Planning, was involved in the decisions leading to the overhaul and expansion of Ultramar's refinery. Ultramar's decisions on whether to produce CARB gas, how to accomplish that goal, and how much to spend was not influenced by any express or implied agreements with its competitors; they were instead independently and unilaterally reached by Ultramar considering Ultramar's available credit line and financial viability. The evidence also showed that, at the end of its conversion, Ultramar had increased its refinery capacity by 20 percent.

Exxon submitted excerpts from depositions of various managers, who explained that economic considerations led them to the decisions on refinery capacity, and that lower-cost modifications were rejected because of concern they would be unable to supply their needs. Higher-cost options were rejected because the incremental additional CARB gas capacity did not justify the additional expense. Another Exxon employee averred Exxon's production levels were not the result of agreement or collusion, and that Exxon actually increased its 1996 gasoline production by 7 percent over its 1995 gasoline production. Mr. Miller, Shell's Manager of Refinery Planning, was responsible for developing refinery modifications to meet new product requirements. He described a lengthy and complicated assessment, undertaken both internally and through the use of an outside consultant, of the various options and risks associated with producing CARB gas at Shell's Martinez and Anacortes refineries. He averred that Shell attempted to develop strategies and options that would further Shell's best economic interests, and did not coordinate or agree with any other defendant in developing or implementing its refinery modifications. He also described Shell's decision-making process to allocate its capital expenditures (choosing to upgrade some of its facilities while closing others; choosing to defer construction of a proposed alkylation and butane isomerization unit to permit construction of a coker unit; expanding an existing alkylation unit and adding a catalytic distillation unit to increase CARB gas production). Shell considered the projected market conditions for CARB gas and Shell's ability to economically produce both CARB gas and more valuable light oil products at its Martinez facility.

APPENDIX B

Mr. May, Shell's engineer in the Planning and Economics Department responsible for overseeing gasoline production at Shell's Martinez refinery, explicitly denied coordinating production levels with other defendants, and explained the products and production levels were independently determined by the Martinez staff. He explained that Martinez was operated on a profit rather than cost premise and was not operated to fulfill the demands of Shell's marketing division. The refinery produced the products that would return the greatest net profits to the Martinez operation. Martinez's products and production were based on anticipated production capacities, costs of products and demands and pricing of products. A computerized program was used that recommends a six-month production schedule designed to optimize Martinez's production and net income.

Mr. Faulstich, Mobil's business optimization team leader for the Supply & Logistics Group at Mobil's Torrance refinery, had responsibility for preparing price projections for the various products manufactured by the Torrance refinery. He denied any agreement or collusion to restrict the capacity or production of CARB gas at Torrance, and explained that decisions for production and inventories of CARB gas were made independently and in Mobil's best interests. He explained that the monthly production planning process commenced with representatives from various departments discussing the refinery's capacity and expected turnarounds for the next 90-day period, and that these factors, as well as projected market price and demand projections, were placed into Mobil's computer program. The goal of the computer program was to create a production plan maximizing the refinery's net operating margin. When a projected market price for one product is higher relative to others, the program recommends greater production for the higher-valued product. The business optimization team discussed the product recommended by the computer with various departments, and then accepted or modified the product and production plans.

Messrs. Coleman and Outlaw, managers at Chevron's El Segundo and Richmond refineries, respectively, averred the production plans for the refinery product mix were developed without consultation or agreement with other defendants. They declared that production decisions were made independently, consistent with the limitations of the refinery equipment, to achieve Chevron's marketing demands while also building a contingency inventory of CARB gas.

Mr. Kiracofe, a senior vice president for ARCO Products Company with responsibility for developing manufacturing production plans for ARCO's refineries, averred ARCO does not consult or agree with competitors when developing its production or inventory plans. ARCO independently determines its production of CARB gas and other products during weekly meetings, attended by representatives of its refinery and volume planning, product supply and wholesale marketing groups, at which six-week "area/grade" plans are produced based on ARCO's internal estimates of production, marketing needs and operating restraints that may limit capacity. ARCO's strategy was to maximize the economic output of CARB gas, and although its plans were hampered by planned and unplanned outages during the first few months of CARB gas production, its refinery operated at full utilization of its available capacity.

Ms. Bird, Exxon's manager at the Benecia refinery, denied any agreement to restrict CARB gas production, asserted that Exxon acts independently and in its own best interests to determine CARB gas production, and demonstrated that its refinery operated at maximum capacity during the initial three-month period and produced more gasoline than during the same period one year earlier.

Mr. Ogden, Tosco's Corporate Vice President, and Mr. Sutton, Senior Vice President for Tosco Refining Company, denied any agreement to restrict production of CARB gas. They explained that Tosco produced at maximum levels, and through its production and acquisition agreements was able to meet both its marketing needs to independent retailers as well as to capitalize on Shell's shortage of CARB gas, created when Shell's Martinez facility was shut down by a fire, by selling CARB gas to Shell at a significant profit.

Messrs. Abay and Hall, Texaco's general managers with responsibility for overall operations of Texaco's refineries, determined their production mix "without any communication, agreement, or concerted action" with other defendants. They explained their decisions on the product mix were made independently and in the best economic interests of Texaco, consistent with the economics of the market and the condition of the refinery equipment.

Mr. Randle, Unocal's General Manager of Refining Planning responsible for coordinating and planning refinery activity, averred that Unocal's refinery produced as much CARB gas as was economically feasible, and that its refining decisions were independent of the anticipated demand of its retail marketing system. Excess production was disposed of through term contracts or sales on the spot market. Unocal adjusted its production of CARB gas in relation to conventional gasoline and turbine and diesel fuel production in response to market conditions during 1996.

APPENDIX C

Mr. Johnson, ARCO's Manager of Price Policy and Strategy responsible for administering and implementing ARCO's wholesale price policy, denied involvement in or knowledge of any agreement to fix prices. He explained that ARCO's strategy was to charge lower prices to increase its volume of sales, and that its pricing policies are closely guarded trade secrets involving numerous price zones among which prices vary. ARCO sets a DTW price, which is then used as a benchmark against which the net dealer price is calculated based on a discount that varies among zones and is determined by zone managers. Similar pricing systems are also used to calculate rack prices to branded distributors. ARCO does not base its DTW price on either Lundberg DTW reports or the OPIS spot price, and ARCO's increases or decreases in its DTW prices do not track changes in prices reported by OPIS or Lundberg.

Mr. Ogden, Tosco's Corporate Vice President, and Mr. Sutton, Senior Vice President for Tosco Refining Company, denied any agreement to fix the price of CARB gas. Tosco's evidence attributed the price spike in the April through June 1996 period to two factors: the price for Alaska North Slope crude oil increased by 45 percent between January and April 1996; and there was a shortage of CARB gas when Shell's Martinez facility was closed by a fire, which caused spot prices to rise.

Mr. Huccaby, Pricing Manager for Chevron, denied any agreement to fix the price of CARB gas, and averred that Chevron sets its prices independently and in its own best interests. He explained that prices are constantly adjusted to permit Chevron to remain competitive, and are set by considering a complex mix of factors, including Chevron's sales volumes, inventory levels, refinery utilization and breakdowns, competitors' refinery problems, street prices of competing retailers, data on wholesale and retail prices obtained from OPIS and Lundberg, and market developments. He also averred that Chevron needed to be aware of competitors' prices to avoid the risk of running out of fuel supplies if its prices were too low or running out of inventory storage space if its prices were too high. He also averred that Chevron's DTW price for CARB gas changed at times, and in increments, different from the DTW price changes of other defendants.

Mr. Stepp, responsible for setting Shell's rack price to Shell's branded jobbers, and Mr. Broderick, responsible for setting Shell's DTW prices for stations supplied directly by Shell, explained that Shell competes to obtain and retain jobbers and must adjust its rack price to assure that its jobbers maintain competitive margins in order to retain their business. Shell also must adjust its rack prices to avoid having rack prices that are too low compared to the prices Shell could obtain for spot sales. If Shell's rack price is too low, (1) Shell loses money it could obtain from a higher spot sale, and (2) jobbers would have an incentive to "lift" supplies in excess of their immediate marketing needs and then sell the excess through unbranded channels or through brokers. Broderick explained the complex factors he considers in setting Shell's DTW price, including twice-weekly proprietary surveys comparing retail prices in various areas that are integrated into Shell's computer program, Shell's sales volume information, daily reports of Shell's DTW prices, and market surveys by Lundberg of DTW prices. Broderick considers Lundberg's information but does not index Shell's prices to those reports. He considers many factors to set DTW prices, which vary for the trade area within California in which each retail station is located. In some areas, competitive pressures require Shell to set DTW prices at unprofitable levels to retain the dealer and maintain sales volumes. As an example, he noted that Shell's Martinez fire required Shell to purchase supplies at spot prices that were above the competitive DTW price level, and Shell therefore lost money on each DTW sale during this period.

Mr. Strong, the marketing manager responsible for setting Mobil's DTW price, and Mr. Soraci, another Mobil employee responsible for planning and marketing during the pertinent time frame, explained that Mobil has over 100 pricing zones in California, and assesses its retail price within each price zone based on retail prices from specific "market drivers" within each pricing zone. DTW prices are then set by "backing out" taxes and a competitive dealer margin within each pricing zone, with various discounts given to dealers for incentives or volume bonuses. A similar market drivers approach is used for rack rates. The prices are set independently of other defendants.

Mr. Zellinger, Unocal's Manager of the Fuels Pricing Group, explained that its prices were not indexed to OPIS prices. Instead, Unocal had 1,100 pricing zones and the business managers for those zones select one of three "market delta strategies" (a delta to an average of the majors; a delta to a lowest prices charged by selected majors; or a delta to ARCO) to set its DTW price within a particular zone, The DTW price against which the delta was applied was obtained from reports from Lundberg and PMI, but these reports only provided "gross" DTW prices from competitors without factoring in rebates, dealer assistance or volume discounts that might change actual DTW prices of other defendants. Approximately 20 percent of Unocal's sales were to jobbers, of which about 5 percent were unbranded gas. For rack sales of branded gas, Unocal used a targeted average price as its delta strategy using either a delta to a targeted average major or a delta to a targeted low major. Its unbranded sales generally used the OPIS average rack price.

Mr. Korte, Ultramar's General Manager for Wholesale Marketing, denied any agreement or collusion in fixing Ultramar's prices. He averred that 42 percent of its CARB gas sales were to unbranded or independent stations. Prices charged to both branded and unbranded stations are determined for each station on a daily basis using a proprietary pricing formula that takes into account competitors' retail pricing, alternative sales values for the supplies, current and anticipated market prices and demands, and historical spot price information.

Mr. Ramsey, Texaco's marketing manager for Texaco's Southern California marketing area, and Ms. Walker, Texaco's marketing manager for Texaco's Northern California marketing area, denied any agreement or collusion with other defendants to set DTW, branded rack or unbranded rack prices. Prices are set daily at each of Texaco's 14 terminals, and those prices vary based on the specific competitive market conditions. Its wholesale pricing considers terminal-specific market conditions, general conditions in the industry, historical data on competitors' DTW, branded rack and unbranded rack prices from Lundberg and OPIS reports, retail prices, and prospective market conditions.

APPENDIX D

Tosco explained it had a 10-year exchange agreement with ARCO beginning in 1986 in which Tosco provided 35,000 BPD to ARCO from Tosco's Northern California refinery in exchange for crude oil from ARCO. Under the 1986 agreement, ARCO had options to extend the agreement, but if gas specifications changed, a renegotiation of price would be required to account for Tosco's increased capital costs and operating expenses required to meet new specifications. ARCO and Tosco unsuccessfully tried to negotiate a new price for CARB gas. Tosco therefore needed to find a new exchange partner because it could not economically modify its refinery to produce quantities of CARB gas sufficient for its marketing needs. Tosco ultimately agreed to exchange 30,000 barrels per day of conventional gasoline for 30,000 BPD with Chevron, and to purchase an additional 5,000 BPD from Chevron. Tosco agreed to pay a fixed differential of 5.75 cpg for 17,500 BPD plus a price differential on the remaining 17,500 BPD that floated with the market price based on the actual market differential between CARB gas and conventional gasoline as published by OPIS. This agreement was a gamble by Tosco because if the actual market differential was below 5.75 cpg Tosco would acquire 17,500 BPD at a locked-in, above-market price. In 1996, because the average differential between CARB gas and conventional gasoline was only 4 cpg, Tosco lost that gamble.

ARCO explained that when it lost its Northern California supply from Tosco, it sought to negotiate exchange agreements with other refiners to satisfy its projected needs for 50,000 BPD for Northern California. ARCO, with a refinery in Los Angeles but none in Northern California, and Shell, with a refinery in Northern California but none in Southern California, agreed to exchange 30,000 BPD as a pure location exchange, which saved both companies transportation costs. Although the contract provided that imbalances in volume delivered would be settled using the OPIS spot pipeline price, that provision has not been used because imbalances are carried forward to the following month and no settlement of imbalances has been necessary. ARCO also explained that it has no incentive to fix an artificially high spot price for CARB gas because it is a significant net purchaser of CARB gas on the spot market; it purchased between 100,000 and 700,000 BPD during March through September 1996 to satisfy its marketing needs.

Mr. Diaz, Texaco's General Manager for Supply and Distribution responsible for coordinating the supply needs of Texaco's marketing department with the production of its refineries and for disposing of surplus and acquiring product to cover shortfalls, averred he acted in Texaco's independent interest without collusion or agreement with other defendants. He explained that Texaco, although usually "long" on CARB gas, regularly obtains CARB gas from other refiners if the location, type or amounts of Texaco's own supplies do not match its needs. Texaco uses exchange agreements of three general types: "location exchanges" of the same type of product in different locations to eliminate the costs and environmental risks involved in transporting the product to the location in which it is needed; "timing exchanges," trading product today for product next month, thereby saving storage expenses; and "product exchanges," trading CARB gas for conventional gasoline, which gives both parties a stable source of a product mix. These agreements are negotiated at arms length and any price term contained in the agreements that may be necessary to compensate for products of different value are based on a good faith estimate of the market price or by reference to reported market prices.

Mr. Rober, Unocal's General Manager for Product Supply and Refinery Sales during the relevant period, averred that Unocal's studies showed that by 1998 Unocal anticipated having a market of 127,000 BPD in California (as well as another 23,000 BPD outside California), and because it expected to have production of only 95,000-100,000 BPD, it expected to be a net purchaser of CARB gas. Unocal participated in exchanges, buy-sell transactions and spot purchases to offset its shortage of CARB gas and to dispose of its surplus of diesel and jet fuel. It was not party to term exchange agreements for CARB gas. Unocal used its buy-sell and spot purchases to balance its system by preventing shortages or gluts of CARB gas in various Unocal markets. Unocal met the needs of its branded dealers by supplementing its production of CARB gas with buy-sell agreements and spot purchases. Any surplus product was sold to other refiners under term contracts or through sales on the spot market to traders, brokers or other refiners.

Mr. Blackwell led Chevron's team examining its projected CARB gas availability and analyzing possible outlets for its excess production. Chevron equipped both its refineries to produce nearly 100 percent of its gasoline as CARB gas, but still had requirements for conventional gasoline for other markets. Chevron therefore had two options: scale back its CARB gas production to allow production of conventional gasoline, or take advantage of its investment by producing 100 percent CARB gas and exchanging it at a premium for conventional gasoline made by others. He concluded the most profitable course was to trade Chevron's excess CARB gas for conventional gasoline and reap the premium. Because it was not operationally feasible to use the spot market to attempt to sell 50,000 BPD of CARB gas each day or to acquire 50,000 barrels per day of conventional gasoline each day, Chevron entered into exchange agreements. Chevron's negotiations with ARCO were unsuccessful because they could not reach agreement on the premium differential to be paid by ARCO, and therefore Chevron negotiated an agreement with Tosco. The price differential negotiated with and to be paid by Tosco resulted in a projected $16 to $23 million CARB gas profit to Chevron. Chevron also modified an existing exchange agreement with Shell by which Chevron obtained conventional gasoline from Shell's Washington refinery (where Chevron had no refinery) as well as a price differential from Shell. These agreements were made to further Chevron's best interests without regard to what actions other defendants might take.

Exxon provided evidence that it uses exchange agreements to keep its gasoline supply in balance. Exxon produces about 100,000 BPD when at full operation, but requires only 20,000 to 30,000 BPD to satisfy its branded market. It sells 30,000 BPD to Tower Energy, which is resold to independent unbranded retailers, and also sells to others who resell to independent retailers.

APPENDIX E

Mr. Walz, Texaco's senior executive with direct responsibility for Texaco's two California refineries during the relevant period, was continuously and personally involved in the decision-making process and the decisions reached about Texaco's CARB gas modification projects. When Texaco was faced with the need to comply with CARB gas regulations, it assessed whether to secure its CARB gas needs by purchasing or producing CARB gas. Texaco compared the costs of refinery modifications with the risks that the market price for CARB gas might not provide an adequate return on Texaco's investment, and considered that it might prove less costly to import CARB gas from outside California than to produce CARB gas in California. Texaco conducted a multi-year study to evaluate the alternatives. Texaco used both internal analyses and outside consultants to project the market future, the best way to reconfigure its refineries, and the engineering for the modifications. Because use of consultants necessitated divulging competitively sensitive business information to the consultants, Texaco instituted safeguards to protect that information, including confidentiality and nondisclosure agreements. Walz averred Texaco hired consultants to assist Texaco in conducting its business, not as a method to transfer information to competitors.

Mr. Kiracofe, an ARCO senior vice president responsible for developing the gasoline production plans for ARCO's refineries, averred that ARCO's decision on the expenditures to convert its Los Angeles refinery to create 150,000 BPD capacity was made independently and without consultation or discussion with other defendants, and that he was unaware of anyone involved in ARCO's decision-making process using non-public information regarding other refiners' investment decisions as part of ARCO's decision-making process. ARCO also produced evidence from consultants that PACE did not do consulting work for ARCO, and the Wright Killen & Co.'s consultants who did work for Unocal were not aware of or involved in the consulting work performed for ARCO; they used only publicly available information about ARCO in consulting for other refiners. ARCO also produced evidence that Booz-Allen & Hamilton's consulting work was performed for ARCO after all the substantial decisions for CARB gas production had been made, and that Booz-Allen & Hamilton took precautions against disclosing ARCO confidential information, made no contacts with other defendants, and prohibited its personnel working on the ARCO project from working on similar projects for other defendants.

Mr. Randle, the manager responsible for planning refinery activities for Unocal's gasoline refineries, and Mr. Van Sluyters, a planning engineer responsible for Unocal's strategic planning, explained that Unocal used PACE to review its "Strategic Plan, 1993-1997" for refining. PACE advised Unocal to maximize its CARB gas production in Los Angeles by increasing its alkylation capacity in Los Angeles, to seek off-shore sources for alklyate, or to consider an alkylate joint venture. PACE was later retained to study sources for alkylate and to examine CARB gas production for Unocal's San Francisco refinery. Unocal initially planned to build an alkylate facility, but chose a different course. When considering its revised options (invest $450 million to produce 120,000 barrels per day of which 100,000 would be CARB gas, invest $300 million to produce 120,000 barrels per day of which 50,000 would be CARB gas, or liquidate all refining and marketing operations), it commissioned PACE to perform a supply/demand study on the anticipated balances for CARB gas, and to study the relative production costs of other refiners to assess whether Unocal would be a high-cost producer of that fuel. When using PACE and other consultants, Unocal expressly prohibited the consultants from disclosing to third parties any confidential information provided to the consultants by Unocal.

APPENDIX F

Plaintiffs cite a survey by PIRA, which was confidential and solely for its client's use, that contained the projected total production capability of California refiners, opined that the market was tight but balanced, and predicted that there would be "extensive price volatility." The report also opined that CARB gas prices would be 10-15 cpg above conventional gasoline prices, with "spikes above this at times" in the event a refiner experienced problems supplying CARB gas. However, plaintiffs do not explain how this report, released in late January 1996, could have impacted capacity decisions made years earlier, and plaintiffs do not identify any refiner-specific confidential information that this report disclosed to other competitors.

Plaintiffs cite Chevron's cooperation in a late 1994 PIRA survey on an industry-wide supply/demand projection, the results of which PIRA agreed to provide to Chevron after the survey was completed. PIRA asked for a capacity number from Chevron, as well as a maximum and minimum expected production, and Chevron told PIRA that Chevron would have a yearly average capability of 240 MBD [240,000 BPD] at 100 percent CARB gas, and peak capability with "everything running" might be 5-10 percent higher. Chevron also told PIRA that "the last increment of gasoline could be CARB or conventional depending on economics, we expected the economics would support 100 [percent] CARB, but anticipate making at least 200-210 MBD as CARB." Plaintiffs do not identify how this refinery capacity information was confidential or to which competitor this information was disclosed by PIRA.

Plaintiffs cite a 1994 study conducted for Unocal by Wright Killen & Co. in which Wright Killen provided a "Retail Marketing And Supply Competitive Assessment." The objective was to compare Unocal's retail marketing with its competitors' retail marketing. Wright Killen gathered information from public pricing information, from internal Unocal data, from public 10-K's and Annual reports, and from "[industry intelligence gathered from periodicals and Wright Killen files." Although plaintiffs conclude the reference to information from "Wright Killen files" demonstrates that competitors' confidential information was transmitted to Unocal, Wright Killen's consultants testified to the contrary: the consultants who did work for Unocal were not aware of or involved in the consulting work performed for ARCO and used only publicly available information about ARCO in consulting for other refiners.

Plaintiffs cite a December 1994 report by Mr. Hileman, an industry consultant, that detailed refinery modifications by the various refiners. However, plaintiffs ignore Hileman's testimony that (1) he obtained much of his information from publicly filed EIR's, (2) most of the information he obtained from "industry contacts" merely confirmed information he already had, and (3) he could not recall having obtained any information from Exxon or Texaco. Although plaintiffs cite Mr. Hileman's providing Mobil with information about competitors' plans to modify hydrocracker operations or increase their alkylation capacity, plaintiffs ignore that Hileman's hydrocracker modification project information was based on a "review of refinery [EIR] statements," and his statements to Mobil about alkylation capacity projects merely "summarize[d] announced refinery alky projects."

It is undisputed that defendants, either directly or through consultants, obtained detailed information about their competitors' proposed projects from their competitors' EIR's or other permitted activities.

APPENDIX G

On appeal, plaintiffs have cited a number of exchange agreements they claim are unorthodox because they involve one or both of the exchange partners receiving product in areas where they have refinery capacity. Plaintiffs' claims are unsupported by the record.

Plaintiffs cite Shell's exchange agreements for deliveries of product to Shell in Northern California, where Shell already has a refinery. However, as discussed above, Shell has refineries in Northern California and Anacortes, Washington, but none in Southern or Central California, and Shell meets its Southern and Central California marketing needs through five exchange agreements. The Shell/ARCO agreement involves exchanging 30,000 BPD with Shell delivering to ARCO in Northern California and ARCO delivering to Shell in Southern California. The Shell/Chevron agreement involves Shell's delivery of 20,000 barrels per day of conventional gasoline to Chevron at Anacortes in exchange for 12,000 BPD from Chevron to Shell in Northern California and 8,000 BPD to Shell in Southern California; the Southern California deliveries help Shell satisfy its retail needs in Southern California, and the Northern California deliveries provide Shell with product it can then trade to other refiners for product Shell needs elsewhere. The Shell/Ultramar agreement provides for Shell to deliver between 2,000 and 3,000 BPD premium to Ultramar in Northern California in exchange for like amounts of regular CARB gas in Southern California. The Shell/Texaco agreement, involving only 40,000 barrels per month of CARB gas, exchanged CARB gas from Shell's Northern California facility for CARB gas from Texaco's Bakersfield facility. Under the Shell/Exxon agreement, Shell delivers conventional gasoline to Exxon from Shell's Anacortes facility in exchange for conventional gasoline from Exxon's Spokane facility. However, because Exxon receives more gas from Shell's Anacortes facility than Shell receives from Exxon's Spokane facility, Exxon satisfies the imbalance by delivering CARB gas to Shell in Northern California, which provides Shell with product to trade for additional Southern and Central California product needed by Shell.

Plaintiffs cite the Tosco/Chevron agreement as unorthodox because Tosco delivered 17,000 barrels per day of conventional gasoline in Northern California and 10,000 BPD in Southern California, where Chevron already had refinery capacity. Plaintiffs argue that because this agreement required Chevron to absorb the costs of shipping those conventional gasoline deliveries (to the Pacific Northwest and Phoenix, respectively) where Chevron could still sell conventional gasoline, the "tortured logistics [demonstrate the agreement was] negotiated by both parties with an eye to something other than efficiency." However, plaintiffs' argument ignores Chevron's need for conventional gasoline that would be unmet with 100 percent CARB gas production, and Chevron calculated that it could best capitalize on its CARB gas modification investment by maximizing CARB gas production, by charging Tosco a price differential, and still obtain the conventional gasoline needed by Chevron for export to other regions. Moreover, although plaintiffs quote Blackwell as testifying that exchanging CARB gas for conventional gasoline to export was "too far out," plaintiffs have wrenched from context and misquoted Blackwell. Blackwell actually testified that, when Chevron was considering various strategies for disposing of its excess CARB gas production, it considered and rejected exchanging CARB gas for conventional gasoline "that we can export to Singapore. And it is just such a far-out case" that Chevron rejected further consideration of that type of arrangement.

Plaintiffs seize on Exxon's San Francisco Bay Area agreements as evidence defendants have pooled their supplies. The Exxon/Chevron Northern California "convenience exchange" agreement and a similar agreement covering Southern California provide arrangements under which they agree to deliver product in amounts mutually agreed upon. Exxon has analogous exchange agreements with Shell, Tosco, Unocal and ARCO. Although plaintiffs claim these convenience agreements pool the supplies of Exxon and Chevron, plaintiffs ignore that the express purpose of those agreements was to handle "occasional activity to cover runouts and/or scheduling related convenience activity" (emphasis added) and permit the requesting party to obtain product only if the sending party has product and is willing to provide it ["[b]oth parties must mutually agree prior to any transaction on this Agreement"]. There is no evidence that convenience exchanges were invented as a method of control in the post-CARB gas era; these types of agreements appear to have predated CARB gas.

NOTES

[1] The defendants, all of whom operate oil refineries in California, are Atlantic Richfield Company (ARCO), Chevron Corporation (Chevron), Exxon Corporation (Exxon), Mobil Oil Company (Mobil), Shell Oil Company (Shell), Texaco Refining and Marketing, Inc. (Texaco), Union Oil Company of California (Unocal), Tosco Corporation (Tosco) and Ultramar Inc. (Ultramar) (collectively referred to as defendants). During the pendency of this appeal, plaintiffs and defendant Unocal entered into a settlement of the action and Unocal is no longer a party to this appeal. References to Unocal in this opinion are included to establish background information and to discuss alleged collusion among defendants, including Unocal.

[2] The word "capacity" as used in this opinion refers to the capacity of an oil refinery to produce CARB gas, unless otherwise specified or the context otherwise requires.

[3] For example, Tosco operates one refinery in California, the Avon refinery. Tosco estimated it would cost at least $400 million to convert Avon to 100 percent capacity. Tosco instead devoted approximately 50 percent of Avon's total gasoline production to CARB gas, and 50 percent to conventional gasoline that could be exported for sale outside of California. Tosco met its shortfall of demand for CARB gas by entering into exchange agreements with other refiners to trade CARB gas for conventional gasoline.

[4] The production of CARB gas was curtailed from March through June 1996 by unplanned refinery operational problems. Additionally, Shell's Martinez refinery, which had produced 87,000 BPD in March 1996, was damaged by a fire on April 1, 1996, and could produce only 20,000 BPD in April 1996 and 47,000 BPD in May 1996.

[5] The studies concluded that Texaco's modification plans would be less capital intensive than those of its competitors and would therefore provide Texaco with a competitive advantage.

[6] Unocal's evidentiary showing resembled Tosco's. It described Unocal's independent business reasons motivating its capacity decisions without expressly denying that its investment choices on refinery capacity were influenced by an agreement with other defendants. Mr. Randle, Unocal's manager responsible for planning refinery activities for its gasoline refineries, explained Unocal considered three options: invest $450 million to produce 120,000 barrels per day, of which 100,000 would be CARB gas; invest $300 million to produce 120,000 barrels per day, of which 50,000 would be CARB gas; or liquidating all refining and marketing operations. Unocal commissioned studies to evaluate the likely demand for and price of CARB gas, and assess whether Unocal would be a high-cost producer. Unocal ultimately elected the first alternative and explained the business reasons why various other alternatives, including spending an additional $140 million on an alkylation plant, were not pursued.

[7] Oil Price Information Service (OPIS) is an independently owned and operated commercial information service that collects pricing information for numerous petroleum products and reports historical price information to its subscribers. The OPIS spot price for CARB gas summarizes the prevailing market prices for bulk purchases of CARB gas on the open market.

[8] A Chevron employee involved in buying and selling CARB gas on the West Coast spot market explained that at least 20 entities not affiliated with any of defendants buy and sell on the spot market, have competing interests and no incentive to allow manipulation of the spot price. Some of these entities are exclusively buyers, without any incentive to maintain artificially high spot prices, and others are trading companies that have different price incentives at different times, depending on whether they have taken "long" or "short" positions at any given time.

[9] Defendants also showed that exchange agreements between competitors is a long-standing industry practice. There are several forms of Exchange agreements: an exchange of like-kind products between competitors with geographically matching needs; an exchange of different products between a competitor who is long on one product but short on another with a competitor who has correspondingly opposite needs; and exchanges of product needed for current marketing needs in return for deliveries to satisfy future marketing needs. The geographical exchanges allow both parties to reduce transportation and storage costs, and avoid the environmental and safety risks associated with storing and transporting products to marketing outlets distant from their refinery.

[10] The Southern California deliveries help Shell satisfy its retail needs in Southern California, and the Northern California deliveries provide Shell with product it can then trade to other refiners for product Shell needs elsewhere.

[11] Shell also had an exchange agreement with Ultramar providing that Shell deliver between 2,000 and 3,000 BPD premium to Ultramar in Northern California in exchange for like amounts of regular CARB gas in Southern California, with Ultramar paying a price differential for the difference in grade.

[12] Shell's other exchange agreement is with Exxon. Shell delivers conventional gasoline to Exxon from Shell's Anacortes refinery in exchange for conventional gasoline from Exxon's Spokane refinery. However, Exxon receives more conventional gasoline from Shell's Anacortes refinery than Shell receives from Exxon's Spokane facility, so Exxon equalizes the imbalance by delivering CARB gas to Shell in Northern California, providing Shell with product to trade for additional Southern California product needed by Shell.

[13] Barnes averred that from March through December 1996 sales to independents accounted for 32 percent and sales to third party wholesalers and traders accounted for 16 percent of all CARB gas sold by Ultramar in California.

[14] Chevron's March 1993 strategic study recognized that because California's relative isolation and unique product requirements created barriers to imports, there were "opportunities for well positioned refiners" and the market was "dominated by a limited number of large, committed refiner/marketers whose individual actions can have significant market impact." A Shell analyst stated in his deposition that an historical problem was that the industry would "over-build [and that's] why we have such low margins." At an ARCO CARB gas strategy meeting in March 1994 the advantages of not encouraging oversupply because "excess supply can hurt integrated margins" were discussed.

[15] By the end of 1993 negotiations to extend the exchange agreement ended because Tosco and ARCO could not agree on the price for the CARB gas to be delivered by Tosco to ARCO under a new agreement.

[16] During this same time period of November 1993 through April 1994, ARCO was also discussing potential exchange agreements with Exxon, Ultramar and Tosco, and Chevron had direct contacts with Shell and Texaco.

[17] A March 1994 memo informed Chevron's management that Chevron would have excess capacity, while the industry supply "appears balanced too short," and that "[s]everal uncertainties emerge from the supply/demand balances [including] independent marketers have no known or foreseeable source of CARB [gas]." The memo suggested that "[downstream marketing does] not have a thorough understanding of the CARB [gas] opportunity or the widespread rearranging of supplies that are needed. [ถ] ... The opportunity for Chevron lies in finding better outlets than .. . traditional sales into the spot market or a variation on exports. [ถ] If nothing changes, some or most of Chevron's physical excess CARB [gas] barrels will flow through the independent marketers at spotlike realizations. [ถ] The downside to having no integrated strategy will be last minute scrambling to balance, foregone opportunities to get the best for ourselves, and perhaps a major market imbalance."

[18] Chevron's internal documents following its March 1994 meeting with ARCO reflect that Chevron learned gasoline balances were "per industry intelligence [ถ] -Arco to [thirty] (30) CARB [ถ] -Unocal to balance [ถ] -Tosco-cut CARB [gas] production to 54. .. ." A senior manager at Chevron was given the job of contacting Shell and Texaco for an "update on plans."

[19] After Tosco decided to limit its capital expenditures at its Avon facility to $100 million and to produce a targeted 60,000 BPD, Tosco negotiated the exchange agreement with Chevron to "provide Tosco additional Marketing volume and to gain control of [Chevron's] excess CARB production...."

[20] The parties signed a letter of intent on September 15, 1994, and the following week announced Tosco had reached agreement with Chevron to exchange 30,000 BPD for conventional gasoline.

[21] Although plaintiffs claim Ultramar's executives also signaled competitors on production capacity, plaintiffs cite only a syllabus from a December 1995 trade association seminar and a February 1996 newspaper interview, both of which allude to capacity information for the industry. Plaintiffs do not explain how this information could have affected capacity decisions made years earlier by the respective refiners.

[22] Plaintiffs claimed O'Malley's signal was then seized by Exxon in an internal Exxon memo several days later detailing Exxon's strategy. Although the Exxon internal report discusses strategies for maximizing profitability with the prospect for a shortage of CARB gas, the internal memo does not mention O'Malley or his estimates, and indeed, Exxon's estimate of the overall CARB gas shortfall was 30 percent less than O'Malley's estimate. Neither O'Malley's comments nor the Exxon internal report discloses the source for their disparate assessments of the prospective shortfall.

[23] Although plaintiffs suggested this memo shows a strategy being developed to limit supply and reduce sales to independents, plaintiffs overlook the memo's discussion, as a part of Exxon's strategy, of a desire to "increase unbranded netbacks [versus] spot market." The memo also discusses Exxon's unbranded marketing options, including extending Exxon's contract with Tower Energy, developing Exxon's unbranded rack marketing business, negotiating supply contracts with Tosco or Ultramar (who are "aggressive sellers at unbranded rack"), or negotiating a fixed fee marketing contract with Tower Energy.

[24] PACE explained that its extensive experience, databases and ability to simulate California refineries would enable it to assess the future impacts of CARB gas on production and consumption in California. PACE planned to examine the viewpoints of refiners, transporters, distributors, blenders, and state and federal regulators to prepare its study, which would enable it to give definitive answers to logistics, blending, crude oil supply, competitive assessment, and facilities construction questions.

[25] In early January 1994, PACE planned to contact each defendant and ask for general information about verification of refinery modifications, construction completion outlook, logistics of importing and exporting gasoline blendstocks, and expected total gasoline and percentage of CARB gas. Plaintiffs do not specify whether this information was confidential competitive information unavailable through public channels, or which refiner-specific confidential information PACE passed to any other defendants. Instead, it appears PACE was gathering information to develop its computer models to develop scenarios for CARB gas supply, demand, price and price volatility.

[26] Plaintiffs have not on appeal cited any confidential competitive information obtained by PACE that was integrated into or influenced PACE's report.

[27] Plaintiffs speculated that this multi-client survey was a method to use PACE as a clearinghouse to pass confidential information to the competitors. However, a review of the evidence does not support this thesis. For example, plaintiffs cite a "call report" between ARCO's Ms. Richards (who performs the competitive assessment work for ARCO) and PACE in which Ms. Richards "admitted" some of her work involved "passing of information." However, this cryptic comment does not identify to whom information is passed, and whether that information is passed to consultants subject to confidentiality agreements to enable consultants to work on ARCO-specific projects. There is no evidence she or anyone else at ARCO passed confidential information to PACE, and ARCO did not use PACE as a consultant. Moreover, the inference plaintiffs drawโ€”that PACE obtained confidential information during this survey and then passed it to othersโ€”is inconsistent with the materials submitted by plaintiffs. For example, a "call report" between Unocal and PACE, in which Unocal disclosed it had canceled plans for an alkylation project, states, "The project is now [canceled]. Even though everyone `knows' it is [canceled], the announcement is not public. (We [PACE] cannot confirm the [cancellation] to other[s].)" Similarly, in a report synopsizing contact between PACE and Tosco in February 1994 in which PACE tried to convince Tosco to purchase the study, Tosco asked about the types of information the study would contain, and specifically asked how much detail PACE would provide about individual refineries. PACE cautioned that "for legal reasons we will have to limit details. However ... [PACE] can provide such details on a single client basis[, and Tosco] indicated that they might be interested in evaluations of their actual plans in light of our findings." The import of these comments was that PACE would provide detailed analysis of a client's individual refinery plans directly to that client, but could not divulge details of other defendants' plans because of legal constraint.

[28] As evidence for this proposition, plaintiffs cite only a quote from the deposition of Chevron's Mr. Blackwell: "The only reason for doing an exchange [is] if you needed product in an area where you didn't have a refinery...." However, Blackwell actually stated, in response to a question asking whether it was a requirement that exchanges have a geographical component, "That would be the only reasonโ€”that would be a reason for doing an exchange: If you needed product in an area where you didn't have a refinery...." (Emphasis added.) The italicized statement, which plaintiffs chose to omit, demonstrates that other reasons may motivate exchange agreements.

[29] But see Appendix G.

[30] Plaintiffs alleged defendants conspired to limit supplies to independents. However, the record shows that during 1996 Mobil had contracts with three independent gas wholesalers, who were free to resell to anyone, under which Mobil sold approximately 12,000 BPD. Exxon sold 30,000 BPD to independent Tower Energy, and smaller amounts to other independents. Ultramar (with a capacity of 60,000 BPD) averred that almost 50 percent of its CARB gas sales between March and December 1996 were to independents and large, third party wholesalers.

[31] Plaintiffs cited Chevron's June 1994 CARB gas strategy, which was to improve West Coast profitability through "profitable exports and/or processing, exchange or supply arrangements with other companies" and its 1995 strategy for improving profitability of its El Segundo refinery, which was to "place surplus [CARB gas] in California market with other refiners to maximize return on refinery investments until adequate marketing outlets are developed." However, the evidence showed that Chevron did not want to produce CARB gas in excess of its existing retail marketing needs unless (1) a specific buyer was identified and an appropriate price established and (2) the sale to that buyer would not "adversely impact the current market structure or undermine our long term retail gasoline strategy...." Chevron's O'Reilly explained that Chevron wanted to focus on its Chevron brand sales, and did not want to sell to "enemies who will then undermine our own branded market customers. [ถ] And it also means that we don't want to ... have some other company take the growth opportunities out there that we see. We're trying to expand our retail marketing system, and we don't want to ... help somebody else get there before us."

Plaintiffs also cited evidence that Exxon, developing its CARB gas strategy during 1995, recognized it was a major supplier to the unbranded market and had the opportunity to increase its share of that market "post-CARB" because Tosco would be required to direct its conventional gasoline outside of California. However, Exxon also recognized that a supply guarantee to the unbranded market could depress prices. Exxon pursued a strategy to place its additional volume with "branded `anyone'" or through term sales, and to avoid placing product into the San Francisco Bay Area spot market (where supplies were expected to be "long").

[32] One of Ultramar's contracts had a rebate provision with the individual purchasers. However, if a purchaser bought more than one million gallons in any given month he would not receive the rebate for the additional gas purchased.

[33] The Chevron/Tosco agreement and the Chevron/Shell agreement involved exchanging CARB for conventional gasoline. The Chevron/Tosco agreement involved a two-tiered price differential: Tosco paid a set price differential (5.75 cpg plus a variable "MTBE pass-through" charge) for 50 percent of the CARB gas it received from Chevron, and a floating price differential (priced at the average spot market differential between CARB and conventional gasoline) for the other 50 percent of the CARB gas it received from Chevron. This agreement involved a gamble by Tosco, which Tosco lost, because the actual market differential was below the 5.75 cpg Tosco paid for 17,500 BPD. (See Appendix D.) The Chevron/Shell agreement involved a similar formula, utilizing a fixed differential and a floating differential, to account for the difference between CARB gas and conventional gasoline being exchanged.

[34] Chevron's Products Division, in recommending to Chevron's management approval of the Tosco agreement, opined that the formula devised under the Chevron/Tosco agreement would yield "about 9-11 cpg depending on market conditions [and] [w]e expect the average market differential will be similar."

[35] Plaintiffs make no effort on appeal to demonstrate that the OPIS spot price was the uniform pricing mechanism used by defendants in their exchange agreements. Although some agreements in addition to the Chevron/Tosco and Chevron/Shell agreements use the OPIS spot price in whole or in part to price CARB gas, many others did not. For example, many of the exchange agreements, including the ARCO/Shell agreements, do not use OPIS to set prices but instead use a fixed "location differential" based on pipeline tariff charges.

[36] Although a price is reported, OPIS does not report sales volume.

[37] Plaintiffs' brief also asserts Lundberg reports the wholesale prices that defendants will charge in the future. However, the record cited by plaintiffs for this proposition contains no reference to future wholesale prices.

[38] Noll opined that defendants had a "general agreement that they wanted to reduce capacity in the supply of CARB gas, and they exchanged information for the purpose of engaging in more effective joint planning in the sense of one firm taking into account, in a more comprehensive way than would otherwise be possible, the supply decisions of another." However, when asked whether he could identify any examples of direct contacts among defendants in which they exchanged information not obtainable from public documents, Noll was unable to do so.

[39] Noll did not submit a declaration in opposition to defendants' motions for summary judgment; instead, plaintiffs submitted portions of Noll's deposition testimony. Although plaintiffs urge on appeal that Noll's deposition testimony supports several subsidiary propositions, plaintiffs' citations to Noll's deposition often do not support the proposition urged by plaintiffs. Accordingly, we must speculate on the precise limitations on and basis for Noll's opinions. able to coordinate buying/selling activity without independent conflicting goals.

[40] Noll opined that because there is no separation between the buying and selling functions in a defendant's trading department, defendants' employees engaged in trading have frequent contact with their counterparts and are cognizant of which companies are buying or selling at any given time and are

[41] Plaintiffs also argue defendants cannot rationally explain the elevated CARB gas prices. Plaintiffs again argue based on facts that their record citations do not support. Plaintiffs claim changes in crude oil prices do not explain the CARB gas price increases, but they cite pages of Noll's testimony in which crude oil prices are never mentioned. Plaintiffs cite a declaration by Noll, submitted in support of plaintiffs' motion for new trial, that peremptorily asserts the crude oil price increase was "quite small" without any detailed discussion of the basis for this assertion. Plaintiffs next assert the cost to produce CARB gas is "substantially less" than the change in price and cannot explain all of the price increase, but can only account for a 6 cpg increase. Noll stated neither the basis for his 6 cpg estimate nor whether that estimate includes or excludes amounts to provide for a reasonable return on defendants' capital investments. Noll also claimed the refinery outages at two refineries cannot account for the price increase because neither production nor inventories were "low" following the outages. However, Noll apparently assumed the refinery outages had no effect on the availability of product or the market spot price for filling the gap between supply and demand, even though another part of plaintiffs' theory was that defendants purposefully implemented limits on capacity to create a "tight" supply/demand balance.

[42] Bloom described trade association meetings as providing an opportunity for exchanging confidential information, but acknowledged that it was not per se improper to attend these meetings. When asked to identify what information defendants actually exchanged at these meetings, he identified only O'Malley's speech and claimed that O'Malley disclosed Tosco's production plans. However, the only evidence of O'Malley's speech was that he projected there would be a shortage of CARB gas in 1996, that price rises were probable, and that price spikes were possible. Tosco had already publicly announced its production plans and its agreement with Chevron to trade CARB for conventional gasoline. Bloom also identified a speech by Brown at an OPIS conference, but did not identify any confidential competitive information disclosed by Brown in that speech.

[43] At the 1994 Pacific Oil Conference, O'Malley predicted price increases of 6 cpg for manufacturing and 4 cpg for oxygenate, and that any unplanned shutdowns could lead to significant price spikes. At the 1995 Pacific Oil Conference, Ultramar's Westfall predicted tightened spot markets with price spikes occurring in the event of unplanned shortfalls. Chevron's MacDonald noted in August 1995 that CARB production capacity was "delicately balanced with expected demand" and that any "prolonged and significant incidents in the summer could lead to shortages .. . [b]oth of these conclusions imply a price spike above long term equilibrium pricing...." The consultants agreed with those assessments.

[44] ARCO prepared information for use by their personnel, and the Western States Petroleum Association prepared a May 1996 set of brochures that attributed recent gas price increases to higher crude oil costs, unplanned refinery shutdowns, and the increased costs associated with producing CARB gas.

[45] The California Air Resources Board chairman's press release complaining of a lack of a satisfactory explanation was apparently premature. By January 1997, the California Air Resources Board reported to the California Legislature that the price increases were attributable to a confluence of market forces that included (1) increased crude oil prices, (2) seasonal demand increases for gasoline and diesel fuel, (3) lower inventories, (4) a significant unplanned outage at Shell's Martinez refinery and a series of minor outages at other refineries, (5) the increased operating and capital costs to produce CARB gas rather than federal RFG gas, and (6) an increased sales tax component associated with rising retail prices. Reports by federal agencies reached similar conclusions and in 1997 the Department of Justice closed its antitrust investigation pertaining to retail gasoline prices. Plaintiffs' experts simply disagree with the state and federal government assessments that market forces provide an explanation for the price increases.

[46] All further statutory references are to the Code of Civil Procedure unless otherwise specified.

[47] The trial court rejected plaintiffs' claims of irregularity in the proceedings, accident or surprise, newly discovered evidence and insufficiency of the evidence as grounds for granting plaintiffs' motion for new trial.

[48] Plaintiffs' motion also asserted insufficiency of the evidence, but the trial court recognized this assertion merely recast plaintiffs' claim that there was an error of law in shifting the burden to plaintiffs because defendants' showing was allegedly insufficient.

[49] Plaintiffs suggest that no burden is shifted to plaintiffs to make any counter-showing until the defendants conclusively negate an essential element of the plaintiffs' claim. However, the Union Bank court noted that the Legislature, when it amended section 437c, did not adopt the language of the judicially declared rule of "conclusively negating" an element of plaintiffs' claim, but instead chose to describe the defendant's burden as providing evidence showing that one or more elements of plaintiffs' cause of action "cannot be established." (Union Bank v. Superior Court, supra, 31 Cal.App.4th at pp. 585, 586, fn. 8, 37 Cal.Rptr.2d 653.) Other courts have recognized that the continued use of the "conclusively negate" terminology as the requisite showing continues to cause uncertainty as to the level of proof required. (See Scheiding v. Dinwiddle Construction Co. (1999) 69 Cal. App.4th 64, 81-82, 81 Cal.Rptr.2d 360.) We are convinced the decisions in Allyson v. Department of Transportation (1997) 53 Cal. App.4th 1304, 62 Cal.Rptr.2d 490 and the other cited cases correctly hold that a defendant satisfies his burden by producing evidence that prima facie shows an essential element of the plaintiff's claim cannot be established. Application of the more stringent "conclusively negate" standard is inconsistent with the statutory structure. The statute contemplates that a sufficient showing by a defendant will shift to plaintiff the burden of making a counter-showing raising a triable issue of material fact. However, if the initial showing shifts the burden only if it has conclusively negated an essential element, by definition plaintiff's counter-showing would be an exercise in futility; the defendant has already conclusively negated the element. This court sub silentio approved the "prima facie" approach in McManis v. San Diego Postal Credit Union (1998) 61 Cal.App.4th 547, 558-559, 71 Cal.Rptr.2d 617 (but see Kovatch v. California Casualty Management Co. (1998) 65 Cal.App.4th 1256, 1266, 77 Cal.Rptr.2d 217), and it is the only interpretation that gives content to statutory shifting of the burden of proof and makes meaningful the opponent's counter-showing.

[50] A plaintiff may not defeat a summary judgment motion by producing evidence to support claims that are outside the issues framed by the pleading. (Robinson v. Hewlett-Packard Corp. (1986) 183 Cal.App.3d 1108, 1131-1132, 228 Cal.Rptr. 591; City of Hope Nat. Medical Center v. Superior Court (1992) 8 Cal. App.4th 633, 639, 10 Cal.Rptr.2d 465.)

[51] Inferences to raise triable issues of fact must be reasonable and cannot be based on "speculation, conjecture, imagination, or guesswork." (Joseph E. Di Loreto, Inc. v. O'Neill (1991) 1 Cal.App.4th 149, 161, 1 Cal. Rptr.2d 636.) Because a plaintiff has the burden of proof by a preponderance of the evidence, the inferences he relies on to defeat summary judgment must satisfy the "more likely than not" burden that he will bear at trial. (Leslie G. v. Perry & Associates, supra, 43 Cal.App.4th at p. 487, 50 Cal.Rptr.2d 785; Cf. Kidron v. Movie Acquisition Corp. (1995) 40 Cal.App.4th 1571, 1580-1581, 47 Cal. Rptr.2d 752 [grant of nonsuit proper when plaintiff's evidence does not support logical and reasonable inference in his favor and is nothing more than speculation or conjecture]; California Shoppers, Inc. v. Royal Globe Ins. Co. (1985) 175 Cal.App.3d 1, 44-45, 221 Cal. Rptr. 171 [an inference may not be illogically and unreasonably drawn or be based on mere possibility or suspicion, imagination, speculation, supposition, surmise, conjecture or guesswork]; Salter v. Keller (1964) 224 Cal. App.2d 126, 128, 36 Cal.Rptr. 430 [". .. it is incumbent upon the plaintiff to produce evidence which supports a logical inference in his favor and which does more than raise a mere conjecture or surmise that the fact is as alleged."].) Finally, as discussed below, substantive law may limit the inferences that may be drawn from ambiguous circumstantial evidence. (See pt. IV.B., post.)

[52] Although plaintiffs' complaint contains a UCA claim in addition to a Cartwright Act claim, the core of both claims is the alleged agreement among defendants to fix and control the capacity, production and price of CARB gas. Defendants' motions for summary judgment focused on the crucial element of "agreement." Although it is not invariably necessary to show an agreement or conspiracy in a UCA claim in contrast to a Cartwright Act claim (Biljac, supra, 218 Cal. App.3d at p. 1422, 267 Cal.Rptr. 819), and unilateral action can violate the UCA (Motors, Inc. v. Times Mirror Co. (1980) 102 Cal. App.3d 735, 741-742, 162 Cal.Rptr. 543), the UCA allegations in this case require a conspiracy. Plaintiffs' UCA claim alleged that defendants, by agreeing to coordinate their actions, violated the UCA. Plaintiffs did not allege in their UCA claim that any individual defendant, by making its individual capacity, production or pricing decisions, violated the UCA. Indeed, in plaintiffs' discovery responses, they admitted that defendants' business practices, to the extent the practices were "otherwise separate, distinct conduct not alleged to be coordinated in an unlawful manner," did not violate California law except to the extent the practices violated the Cartwright Act. Additionally, apart from plaintiffs' generic citations on appeal asserting that the UCA is distinct from and can be violated without engaging in conduct violating the Cartwright Act, plaintiffs do not cite any unfair-competition-law authority demonstrating that their UCA claim is applicable to defendants' conduct apart from the alleged antitrust conduct. We could deem this an abandonment of their UCA claim. (Biljac, supra, at pp. 1423-1424, 267 Cal.Rptr. 819.) However, because the UCA provides remedies that are cumulative to those provided by the Cartwright Act (People v. McKale (1979) 25 Cal.3d 626, 632, 159 Cal.Rptr. 811, 602 P.2d 731; Bus. & Prof.Code, ง 17205), we assume plaintiffs' lack of separate analysis means that the alleged acts are "unlawful" under the UCA only to the extent that they violate the Cartwright Act. (See Blank v. Kirwan (1985) 39 Cal.3d 311, 329, 216 Cal.Rptr. 718, 703 P.2d 58.) Accordingly, we analyze the alleged conspiracy as do the parties: we use only those cases interpreting the Cartwright Act and federal antitrust laws.

[53] Indeed, the comparative scarcity of state-court precedent often means that we must rely heavily on federal cases. (Redwood Theatres, Inc. v. Festival Enterprises, Inc. (1988) 200 Cal.App.3d 687, 694, 248 Cal.Rptr. 189.)

[54] The court in Biljac declined to accept or reject the application of the federal limitation on permissible inferences. The majority acknowledged that to the extent the limitations were derived from the substantive policies underlying antitrust laws, it would be fully applicable to Cartwright Act claims because the policies underlying both statutes were identical. (Biljac, supra, 218 Cal.App.3d at pp. 1425-1427, 267 Cal.Rptr. 819.) However, the concurring opinion perceived that the limitation was based on the evidentiary burdens applicable to summary judgment motions in federal court under Federal Rules of Civil Procedure, rule 56, 28 U.S.C., and California's section 437c was sufficiently different from the federal approach to allow application of the federal cases. (Biljac, supra, at pp. 1435-1445, 267 Cal.Rptr. 819 [cone. opn. of Kline, J.].)

[55] Plaintiffs claim Biljac rejected the federal approach. Although Biljac ostensibly declined to decide whether the federal approach was applicable, it relied on Wilcox v. First Interstate Bank of Oregon, N.A., supra, 815 F.2d 522 (which used the Matsushita approach), and de facto employed the federal approach in its analysis. For example, rejecting the plaintiffs' claims that the conspiracy could be inferred from participation by the defendants in trade association meetings, Biljac noted there was a strong policy that disseminating competitive information through trade association activities is condoned by antitrust law, and therefore that "plaintiffs must show conspiracy with something beyond" those activities. (Biljac, supra, 218 Cal.App.3d at pp. 1432-1433, 267 Cal.Rptr. 819.)

[56] Plaintiffs finally argue that Matsushita applies only when a plaintiff relies on "ambiguous" circumstantial evidenceโ€”conduct that could have either pro- or anti-competitive purposes and effectsโ€”to show the requisite agreement, and that this case involves unambiguous evidence of a combination to fix prices that can never have a significant procompetitive effect. This is a classic bootstrap argument: because defendants' conduct was undertaken pursuant to a conspiracy to fix prices, their conduct cannot have a pro-competitive effect and therefore their conduct is not "ambiguous"; because defendants' conduct is not ambiguous, defendants' conduct may alone serve as a springboard from which to infer the conspiracy existed. Plaintiffs may not assume the existence of a conspiracy as a predicate to proving defendants engaged in anti-competitive conduct from which the conspiracy may then be inferred.

[57] Biljac was decided before the California summary judgment statute was amended to more closely resemble the federal rules.

[58] Biljac also referred to federal precedents holding that conspiracy allegations can be rebutted by showing justifiable reasons for practices consistent with independent business practice. Biljac referred to the decision in Matsushita Elec. Industrial Co. v. Zenith Radio, supra, 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 in which the U.S. Supreme Court held that, although on summary judgment the inferences to be drawn from the facts must be viewed most favorably to the party opposing the motion, "antitrust law limits the range of permissible inferences from ambiguous evidence in a [Sherman Antitrust Act section] 1 case. Thus, in Monsanto Co. v. Spray-Rite Service Corp.[, supra,] 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775 ..., we held that conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy. [Id. at p. 764, 104 S.Ct. 1464.] [Citation.] To survive a motion for summary judgment . .., a plaintiff seeking damages for a violation of ง 1 must present evidence `that tends to exclude the possibility' that the alleged conspirators acted independently. [Id. at p. 764, 104 S.Ct. 1464.] [Plaintiffs], in other words, must show that the inference of conspiracy is reasonable in light of the competing inferences of independent action or collusive action that could not have harmed [them]. [Citation.]" (475 U.S. at pp. 587-588, 106 S.Ct. 1348.)

[59] Our interpretation of Biljac's reference to an "unbroken chain" finds support in several observations. First, the Biljac court described as the "unbroken chain of officers" those "whose principal responsibility it was to determine the bank's prime rate and commercial loan pricing policies from 1970 into 1986." (Biljac, supra, 218 Cal.App.3d at p. 1423, 267 Cal.Rptr. 819.) Second, that description was immediately preceded by the observation that the moving defendants presented declarations from the officers and employees "charged over the years in question" with setting their rates, suggesting that Biljac was concerned with the time element. (Ibid.) Third, the Biljac court repeatedly describes the plaintiffs' conspiracy theory as an agreement between the banks that evolved over a multi-year period (see Biljac at pp. 1417, 1423-1424, 1429-1434, 267 Cal.Rptr. 819), which reinforces our conclusion that Biljac was concerned with the chain of personnel involved in rate setting over the years of the alleged conspiracy, rather than the vertical chain of command within the banks between 1970 and 1986. Finally, the legal authority cited by Biljac as supporting its conclusions that the declarations were adequate (Id. at p. 1424, 267 Cal.Rptr. 819) were cases involving declarations from persons competent to testify about the practices during the relevant time periods even though those declarants were not the chief executive officers for the defendants. (See Chatman v. Alameda Flood Control etc. Dist. (1986) 183 Cal.App.3d 424, 428-429, 228 Cal.Rptr. 257 [averments by engineer for flood control district, who declared district did not approve, accept, have easements in, or take responsibility for culvert, competent to establish these issues on district's behalf in summary judgment motion]; Zuckerman v. Pacific Savings Bank, supra, 187 Cal.App.3d at p. 1404, 232 Cal.Rptr. 458 [declaration of one individual, holding position as collections and foreclosure manager during relevant time period, sufficient to establish bank's practices]; Kiernan v. Union Bank (1976) 55 Cal.App.3d 111, 116, 127 Cal.Rptr. 441 [declarations from a vice president and an operations officer, employed during relevant time period, sufficient to establish bank's practices].)

[60] For example, Texaco's Vice President for Planning and Administration established that product slates are determined at the refinery level, and Messrs. Abay and Hall, Texaco's general managers responsible for the overall operations of Texaco's two refineries, averred they had "ultimate authority and responsibility" over their production mix and made those decisions "without any communication, agreement, or concerted action" with other defendants. Similar representations were made by other defendants. (See, e.g., declaration by Shell's Mr. May ["1 am responsible for overseeing gasoline production planning at Martinez . .. [and] [a]ll decisions regarding the slate of products and production levels at Martinez are made independently by myself and other employees at Martinez ...."] and ARCO's Mr. Kiracofe [as "Senior Vice[] President for ARCO Products Company ... I had responsibility for .. . overseeing ARCO's two refineries and the development of manufacturing volumetric plans for those refineries" and ARCO's plans for production slates are determined during weekly meetings attended by representatives of its refinery, volume planning, product supply and wholesale marketing groups].)

[61] For example, Mr. Johnson, ARCO's Manager of Price Policy and Strategy, averred he had "full . .. responsibility for administering] and implementing] [] ARCO's gasoline price policy at the wholesale level" and denied involvement in or knowledge of any agreement to fix prices. Similarly, Chevron's Mr. Huccaby averred that he was "Pricing Manager" for Chevron and responsible for gasoline and diesel fuel pricing for the United States. He denied any agreement to fix the CARB gas price, and averred that Chevron sets its prices independently in its own best interests.

[62] For example, Texaco's Mr. Diaz averred that as "General Manager of Supply and Distribution I have the ultimate . . . responsibility for" coordinating the supply needs of Texaco's marketing department with the production of its refineries, and for disposing of surplus or acquiring product to cover shortfalls, which tasks include making spot sales, term sales and exchange agreements. He averred he acted in Texaco's independent interest without collusion or agreement with its competitors, and negotiated agreements at arms length at prices based on good faith estimates of the market prices or by reference to reported market prices. Similarly, Ms. Barnes, Ultramar's General Manager of Product Supply and Distribution, was responsible for balancing gasoline supply and demand for Ultramar and was a participant in the negotiation of Ultramar's exchange agreements. She averred Ultramar's exchange agreements were independently negotiated in its independent business judgment.

[63] Plaintiffs' theory, asserted throughout these proceedings, was that defendants' agreements constituted a per se violation of antitrust laws.

[64] Moreover, even had plaintiffs' complaint focused on the effect of an exchange agreement, plaintiffs offered no evidence that any single exchange agreement had a sufficiently pernicious effect on competition to violate the Cartwright Act. Plaintiffs' theory, based on Noll's testimony, is that the aggregate of all the exchange agreements had an anti-competitive effect. However, plaintiffs offer no authority for the proposition that a two-party contract, otherwise permissible under antitrust laws, is transformed into a contract violative of antitrust laws merely because other actors independently enter similar contracts and the cumulative effect of these independent contracts is anti-competitive. Instead, we perceive that before Noll's cumulative impact theory could result in antitrust liability, plaintiffs are required to prove the two-party agreements were components of or methods to implement an overarching agreement among defendants.

[65] Plaintiffs cite the lawsuit by several defendants challenging the "small refiners exemption" as evidence of concerted action to restrict capacity. We are not persuaded by plaintiffs' argument. First, a defendant's concern that this exemption from CARB regulations permits small refiners to produce gas at lower costs, and thereby undermine the CARB "market premium" in some local markets, is not probative of an agreement to restrict overall capacity. The lawsuit is equally attributable to an independent legitimate business concern that the exemption gives an unfair competitive advantage to the exempt refiner against whom the defendants must compete to recoup their investment. Second, a company is entitled to pursue legal action, even in concert with other competitors, without violating antitrust laws. (Blank v. Kirwan, supra, 39 Cal.3d at pp. 320-321, 216 Cal.Rptr. 718, 703 P.2d 58.)

[66] Plaintiffs also claim defendants used consultants as a vehicle for passing information to assist their agreement to coordinate refinery modification plans, a claim we discuss post.

[67] Plaintiffs' experts suggested that there was actually excess refinery capacity. Dr. Leffler explained his damage calculations did not make any adjustment for price increases attributable to the reduction in supplies caused by the Shell Martinez fire in April 1996 because he understood that plaintiffs' expert Mr. Brown had opined "there was 14 percent excess capacity available at the time of the Martinez fire that would have been more than sufficient to make up for any decreases in supply as a result of that fire."

[68] Plaintiffs also rely on a March 1993 contact between the chairmen of ARCO and Chevron as evidence of the conspiracy to restrict capacity. However, the evidence cited by plaintiffs is an internal Chevron memo stating that in March 1993 Chevron reassessed the possibility for lobbying to delay implementation of the new regulations, unanimity would be needed in that effort, Chevron's chairman contacted ARCO's chairman and learned ARCO still supported the regulations and the timing, and therefore Chevron dropped the effort. There is no suggestion this contact concerned capacity. Furthermore, there is no prohibition against joint lobbying activity. (Blank v. Kirwan, supra, 39 Cal.3d at p. 320, 216 Cal.Rptr. 718, 703 P.2d 58.)

[69] Plaintiffs note that between November 1993 and April 1994, ARCO discussed potential supply contracts with Exxon, Ultramar and Tosco, and Chevron had direct contacts with Shell and Texaco. From this evidence, plaintiffs infer Chevron and ARCO "held the proxy" for all other defendants to discuss and agree on coordinated refinery modifications. We decline to draw this conclusion. First, Unocal and Mobil are absent from the alleged "assembly of the whole." Second, it is pure speculation that the negotiators of a two-party exchange agreement controlled by proxy the investment decisions (involving billions of dollars) of non-participants.

[70] Plaintiffs conceded that "[b]y September 1993, ARCO had decided to completely convert the Los Angeles Refinery to the production of CARB gasoline," and actually increased its capacity in Los Angeles. Similarly, Chevron's capital expenditure project was approved in September 1993, which ultimately increased its capacity, and plaintiffs cite nothing to suggest Chevron altered its plans as a result of an agreement arising out of later ARCO/Chevron exchange agreement discussions.

[71] Plaintiffs' efforts to prove there were direct contacts among defendants to discuss and agree on coordinated capacity also rely on a December 1992 internal Texaco memo by a Texaco employee summarizing a conversation with a Shell governmental relations official, in which Shell expressed "[extreme concern] about Texaco's silence and lack of activity concerning [Texaco's] plans" for coping with the new California Air Resources Board regulations. The memo reflects that most other defendants had already announced their plans or had begun steps to secure the permits necessary to modify their refineries. Shell was nervous that Texaco was "positioning itself to be the `wild card' on this issue" and stated that if Texaco sought to use imported sources of CARB gas, Shell would take legislative action to prevent that course. It is paradoxical that plaintiffs use evidence of Texaco's silence and non-cooperation, and its possible use of a different strategy that could undermine the competitive position of Shell, as proof that Texaco agreed to coordinate capacity.

[72] The fact any two defendants entered an agreement is directly shown by the exchange agreements. However, entry into an agreement is not direct evidence of a different agreement to restrict and pool supply and fix prices. Antitrust law bars only agreements designed to achieve an unlawful objective. (Monsanto Co. v. Spray-Rite Service Corp., supra, 465 U.S. at p. 764, 104 S.Ct. 1464.) To find that defendants entered into an unlawful agreement requires inferences from the exchange agreements. (Biljac, supra, 218 Cal. App.3d at p. 1430, 267 Cal.Rptr. 819.)

[73] Plaintiffs' expert acknowledged that "it was in the self-interests of each company entering into an exchange to do that, regardless of what other competitors ... were doing."

[74] Exchange agreements are also beneficial to the environment because they eliminate the risks of accidental spills if product is required to be stored and then shipped great distances.

[75] For example, plaintiffs claim that Exxon and ARCO entered into exchange agreements to prevent unbranded sales that depressed prices. However, the Exxon/ARCO agreement involved a convenience exchange that had independent business justifications. (See Appendix G.) Second, the only evidence cited to support plaintiffs' propositions are two mid-1995 internal Exxon memos that show when Exxon was developing its CARB gas strategy it recognized it was a major supplier of the unbranded market and had the opportunity to increase its share of that market "post-CARB" because Tosco would be required to ship its non-CARB gas outside of California. However, Exxon also recognized that a supply guarantee to the unbranded market could depress prices. The memo proposed that Exxon's strategy should be to place its additional volume with "branded `any-one'" or through term sales, and to avoiding placing product into the San Francisco Bay spot market where supplies were expected to be "long"). Plaintiffs ignore that one of Exxon's listed priorities was sales to "BE Trib unbranded," and that it in fact continued selling 30,000 BPD to independent Tower Energy, plus smaller amounts to other independents. Evidence that Exxon considered methods of increasing its profitability (1) shows nothing about ARCO's reasons for entering into the convenience exchange and (2) is as consistent with Exxon's permissible independent interest as with an illegal conspiracy.

[76] The Chevron/Tosco agreement involved a two-tiered price differential: Tosco paid a set price differential (5.75 cpg plus a variable "MTBE pass-through" charge) for 50 percent of the CARB gas it received from Chevron, and a "floating" price differential (priced at the average spot market differential between CARB gas and conventional gasoline) for the other 50 percent of the CARB gas it received from Chevron. The Chevron/Shell agreement involved an analogous formula, utilizing a fixed differential and a floating differential, to account for the difference between CARB gas and conventional gasoline being exchanged.

[77] Moreover, there is no evidence there was an agreement among all defendants to use a set differential to fix the price of CARB gas based on a formula. Plaintiffs make no effort on appeal to demonstrate that the formula in the Chevron/Tosco and Chevron/Shell agreements was the agreed uniform pricing mechanism or even was used by other defendants in their exchange agreements, or that the price differentials were identical. Although some agreements used the OPIS price in whole or part to price CARB gas, others did not.

[78] Plaintiffs also cite Donchin v. Guerrero (1995) 34 Cal.App.4th 1832, 41 Cal.Rptr.2d 192 for the same proposition. However, the issue in Donchin was whether a landlord knew of the vicious tendencies of a rottweiler dog on his premises. On summary judgment, the court concluded there was a triable issue of fact as to the landlord's knowledge. The court focused extensively on the fact that, after the dog attacked, he denied knowing of the dog's presence and denied that he gave the dog's owners permission to keep the dog, both of which denials were demonstrably false exculpatory statements. (Id. at pp. 1841-1843, 41 Cal.Rptr.2d 192.) Although the court also concluded the plaintiff's affirmative evidence of the landlord's knowledge was "fairly strong ... we find the [evidence of false exculpatory statements] more persuasive and thus concentrate on it" as a ground for concluding a triable issue of fact existed. (Id. at p. 1840, 41 Cal.Rptr.2d 192.) Although the expert's testimony, with other evidence, reinforced the "inference already drawn from [landlord's] false exculpatory statement" (id. at p. 1845, 41 Cal.Rptr.2d 192) that the landlord had knowledge of the dog's vicious tendencies, the opinion makes clear that numerous bases existed for rejecting summary judgment.

[79] In addition to our conclusion that Noll's opinion does not provide a basis to infer the requisite agreement, Noll's opinion is also based on unsound assumptions, which we discuss more fully at pt. VII.B-4, post.

[80] Noll also appears to have disavowed any testimony that all the companies expressly agreed to create this institutional structure.

[81] We admit some confusion over this aspect of plaintiffs' claim. In the exchange agreements with no imbalances in the amounts exchanged, the cost of the goods received by a company would appear to be the cost to produce the product that the defendant sent in exchange, with any appropriate transportation adjustments. Thus, we cannot perceive how the OPIS spot price would be at all relevant to the cost from which those companies would base their DTW and rack rates.

[82] For example, from March through October 1996 Ultramar engaged in 139 spot transactions with non-defendants who presumably had no interest in facilitating artificially high spot prices.

[83] Although Noll theorized a defendant would be willing to pay an artificially high spot price in one transaction in order to reap the benefits of the higher OPIS spot price when it subsequently sold CARB gas using that elevated price, he declined in his deposition to identify any examples of transactions involving that conduct.

[84] The evidence cited by plaintiffs does not show joint planning of price increases. The industry recognized prices would increase, but plaintiffs' proffered evidence of joint activity is evidence of unilateral activity. Plaintiffs cited a late 1995 internal ARCO document (without any showing it was distributed to other defendants) stating "begin price increases to recover costs" on April 1, 1996. Plaintiffs do not suggest that ARCO or any other defendant could not legitimately adopt some price increases to recoup their investments. Mr. Johnson, ARCO's Manager of Price Policy and Strategy, explained the unilateral and legitimate reasons for ARCO's actual price increase in April 1996: ARCO's lower pricing policy was generating such additional demand that ARCO was concerned about "runouts" of product; increases in crude oil prices and production costs were causing ARCO's core refining and marketing operations to lose money; and ARCO (because of increased demand and refinery problems) was forced to purchase product in spot transactions at high cost. Plaintiffs also cite Chevron's internal predictions that prices would rise and spikes "could well occur" in the event of prolonged and significant supply interruptions. Again, however, the cited evidence was an internal Chevron document, without evidence that it was distributed to other defendants, and does not in any event show "joint preplanning" activity.

[85] Indeed, Tosco's projected price increase of 6 cpg for manufacturing accorded with the California Air Resources Board's 1997 report that recovery of capital expenditures and increased operating costs translated into an increased cost of between 5 and 8 cpg to produce CARB gas.

[86] Plaintiffs cite as evidence of misinformation only the April 1996 statements by Texaco's Mr. Walz. He stated at a California Air Resources Board meeting that part of the upward pressure on prices was attributable to supply imbalances cause by refinery outages. Although the statement was certainly information, the only basis for asserting it was misinformation is Noll's opinion the refinery outages were not a cause of price increases. Noll's opinion, which conflicted with the state and federal findings (see fn. 45, ante), is not a substitute for evidence.

[87] In economic and antitrust terminology, an oligopoly is a market with a relatively limited number of sellers and a relatively fungible or standardized product. (See Reserve Supply v. Owens-Coming Fiberglas (7th Cir.1992) 971 F.2d 37, 50.)

[88] It is also undisputed that collecting and using that information is legitimate conduct and cannot serve as a predicate for an inference of a price-fixing agreement to avoid summary judgment. The courts have uniformly recognized that competitors may and should be free to collect and use information about their competition in conducting their business (Biljac, supra, 218 Cal.App.3d at pp. 1430-1433, 267 Cal.Rptr. 819), and that consideration of their competitors' prices is an essential component of their own pricing decisions. (Wallace v. Bank of Bartlett, supra, 55 F.3d at p. 1169.) To avoid chilling that legitimate business conduct, an inference of conspiracy will not be drawn to defeat summary judgment merely because a competitor's price information is found in the internal files of a defendant (Stephen Jay Photography, Ltd. v. Olan Mills, Inc. (4th Cir.1990) 903 F.2d 988, 996) or from evidence a defendant attempts to and succeeds in gathering information about his competitor's pricing policies. (In re Baby Food Antitrust Litigation, supra, 166 F.3d at pp. 125-126.) Even the Petroleum Products court recognized that "[t]o allow an inference of agreement from such evidence, standing alone, would have the effect of deterring competitors from obtaining information about other companies' actual retail prices, even through legitimate means. We can think of few inferences that would have a more undesirable distorting effect on market conduct." (Petroleum Products, supra, 906 F.2d at p. 455.)

[89] For this reason, we reject plaintiffs' claim that Bloom's unrebutted testimony precludes summary judgment. Bloom's opinion asserted that there was no competitive justification for disseminating the detailed DTW and rack prices contained in these third party services, and that the only rational explanation for disseminating the information provided to OPIS, Lundberg and other media services was to signal the competition and facilitate collusive and coordinated pricing decisions. However, Bloom ignores the fact that defendants could not do business without telling the retailers and jobbers the detailed price information that Bloom claims should not have been revealed. (Wallace v. Bank of Bartlett, supra, 55 F.3d at p. 1169, fn. 5 [a defendant must make public its prices to inform customers how much they will have to pay].) That these third parties chose to disclose the information, and defendants then considered it, is of no significance because neither evaluating other companies' prices (ibid.) nor labeling the conduct a "signal" (E.I. Du Pont de Nemours & Co. v. F.T.C., supra, 729 F.2d at p. 139) converts lawful conduct into evidence of a price-fixing conspiracy.

[90] For similar reasons, we conclude several other cases on which plaintiffs rely are inapplicable. In Container, supra, 393 U.S. 333, 89 S.Ct. 510, 21 L.Ed.2d 526, the evidence showed (1) there were direct competitor-to-competitor exchanges of price information; (2) the exchange occurred pursuant to an industry-wide understanding or agreement, which itself constituted concerted action (id. at p. 335, 89 S.Ct. 510); and (3) the type of information exchanged was customer-specific prices, which enabled price matching. (Id. at pp. 336-337, 89 S.Ct. 510.) None of those facts are present here. In United States v. Socony-Vacuum Oil Co. (1940) 310 U.S. 150, 60 S.Ct. 811, 84 L.Ed. 1129 and People v. National Association of Realtors (1981) 120 Cal.App.3d 459, 174 Cal.Rptr. 728, there was direct evidence of an agreement to a program that restrained trade. Neither case involved ambiguous circumstantial evidence from which to infer the agreement.

[91] Plaintiffs also cite a statement by a defense expert that "trade associations may be used as a conduit for [unlawful communications." That falls short of the proof required. (Biljac, supra. 218 Cal.App.3d at p. 1431, 267 Cal. Rptr. 819 [evidence of "discussion[] without disclosing any agreement ... does not bear further discussion"]; Alvord-Polk, Inc. v. F. Schumacher & Co., supra, 37 F.3d at p. 1013.)

[92] Plaintiffs also argue the court erred in awarding costs to defendants. A prevailing party is entitled to costs "[e]xcept as otherwise expressly provided by statute." (ง 1032, subd. (b).) Plaintiffs argue that because the Cartwright Act permits a plaintiff to recover "reasonable attorneys' fee[s] together with the costs of the suit" (Bus. & Prof.Code, ง 16750, subd. (a)) but does not authorize a similar award to a defendant, the latter statute creates the express exception to costs otherwise required under section 1032, subdivision (b). The identical argument, albeit arising under a different consumer protection statute, was rejected in Murillo v. Fleetwood Enterprises, Inc. (1998) 17 Cal.4th 985, 73 Cal.Rptr.2d 682, 953 P.2d 858. In Murillo, the court affirmed an award of costs to defendant because the court concluded a provision of the "lemon law" (Civ.Code, ง 1790 et seq.) that mandated recovery of costs and attorney fees by the prevailing plaintiff-buyer but did not mention a prevailing defendant-seller was not an "express" exception to the general rule in section 1032, subdivision (b). (Murillo, supra, at pp. 989-991, 73 Cal.Rptr.2d 682, 953 P.2d 858.)