The appeals and cross-appeals in this case present a potpourri of antitrust problems in the context of a newspaper distribution system before and after the publishers' conversion of the system from independent dealer-distributors to employers of the newspaper publishers.

We first identify the dramatis personae : All the plaintiffs are independent distributors of the newspapers published by Daily Review, Inc. ("DRI"). Defendants are two corporations and individual officers or employees of those corporations. DRI publishes The Daily Review a daily afternoon newspaper; The Argus, a daily morning paper; and The Daily Review Shopping News, a "throwaway" advertising circular. Bay Area Publishing Company ("BAPCO") publishes the Tri-Valley Herald, a daily morning paper; and the Tri-Valley News, a three day per week, controlled circulation afternoon paper. BAPCO is a wholly-owned subsidiary of DRI. The individual defendants are Floyd L. Sparks, William Chilcote, Dallas Cleland and John Clark. Sparks is the controlling shareholder of DRI, the president of DRI and BAPCO, and the publisher of both companies' newspapers. William Chilcote is a vice-president and business manager of DRI and a member of the Board of Directors of DRI and BAPCO. Cleland is the director of circulation of the four newspapers published by BAPCO and DRI. Clark is the circulation promotion manager for The Daily Review and The Argus.

Plaintiffs allege that defendants entered into horizontal and vertical agreements in restraint of trade. They claim that the dealership contracts used by DRI/BAPCO fixed resale prices and imposed territorial restraints in violation of the Sherman Act, Section 1. (15 U.S.C. § 1.) They also claim that DRI's termination of its independent dealer system violated Section 1. Finally, they argue that the defendants violated Section 2 of the Sherman Act (15 U.S.C. § 2) by attempting to monopolize the newspaper trade. The district court found for the plaintiffs on the price-fixing count, but awarded neither damages nor injunctive relief. The court found for the defendants on the remaining Section 1 counts and on the Section 2 claim. Defendants have not appealed the price-fixing holding. We affirm the district court on the other Section 1 counts, remand for a new trial on damages, and a limited remand on the Section 2 count.1

In 1950 Sparks adopted an independent dealer system for the distribution of his newspapers to home subscribers. Under this system, distributors purchased newspapers from the publisher and resold them to carriers (newspaper boys/girls), who delivered/resold the papers to the subscribers. There were, therefore two independent units in the distribution system: the distributors and the carriers, both of which purchased the paper from the level above. From 1950 until 1969, relationships between the papers and the distributors were governed by a standard form Dealer's Agreement which provided that the dealer would sell newspapers to the subscribers within his route or territory at a fixed subscription price and that the subscription price, the price paid to the publisher by the dealer, and the size and boundaries of the territory were subject to change by the publisher in his sole discretion. The dealer could not assign, transfer or hypothecate rights arising under the agreement without the prior written consent of the publisher.

In 1969, a dispute arose before a state agency as to whether the BAPCO distributors were independent contractors or employees. With the assistance of a consulting firm, a new agreement was drawn up and used by DRI and its distributors. A similar agreement was used by BAPCO. The agreement fixed retail prices.2 It provided for a dual rate wholesale price for newspapers supplied to a distributor for his route, i. e., a certain price up to a specified number of newspapers and a different price for papers above that number. The agreement permitted him to transfer his rights to another party on 60 days' notice; the transferee had to be bondable, qualified, and fully trained to DRI/BAPCO's satisfaction. Finally either party could terminate the agreement on 30 days' notice.

On May 14, 1973, plaintiffs' counsel wrote Sparks a letter stating that certain provisions of the dealer agreements used by DRI and BAPCO constituted unlawful restraints of trade. In July 1973, DRI and BAPCO, pursuant to the 30-day termination provision, notified all distributors that they were terminating their entire independent dealer system for home delivery and converting to a system of employee-distributors. Under the internal system of distribution, employees of DRI and BAPCO sold the newspapers directly to the independent carriers. Each terminated dealer was offered employment as a salaried district manager within the new system.

On August 6, plaintiffs filed this action; defendants agreed to continue to sell newspapers to the plaintiffs and to hold open the offers of employment.3 This agreement was later formalized in a stipulated temporary injunction.

Plaintiffs allege that defendants entered into both horizontal and vertical contracts, combinations or conspiracies. They claim that the post-1969 agreements were unlawful contracts to fix resale prices and to impose territorial restraints on the distributors. They also claim that the termination provision and the actual terminations were violations of Section 1.

* A. Intraenterprise Conspiracy.

Plaintiffs assert that DRI and BAPCO conspired to appropriate the distribution organizations without compensation by converting the entire distribution system. The district court found that there was no conspiracy in fact since the plaintiffs "presented no evidence other than Sparks' decision, such as the active participation of other corporate officers or agents in this decision (to terminate), which would support a finding that DRI and BAPCO combined or conspired to violate § 1." (383 F.Supp. at 1359.)

Section 1 prohibits conspiracies in restraint of trade between two or more people in economic entities. As a purely verbal matter, the officers or directors of a corporation can "conspire," and any contract they make restrains trade. For antitrust purposes, if the Sherman Act forbids such activities it:

" . . . would be socially inconvenient and historically surprising. So long as the business enterprise is regarded as an individual economic unit, it must be permitted to act." (P. Areeda, Antitrust Analysis 319 (2d ed. 1974).)

The problem has been to define an economic unit. In Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc. (1951) 340 U.S. 211, 215, 71 S. Ct. 259, 95 L. Ed. 219, the Supreme Court held that two separately incorporated subsidiaries within the same corporate family can conspire: "common ownership and control does not liberate (them) from the impact of the antitrust laws . . . especially . . . where (the corporations) hold themselves out as competitors." (Kiefer-Stewart, supra, 340 U.S. at 215, 71 S. Ct. at 261 (wholly-owned subsidiaries). Accord: Perma Life Mufflers v. Int'l Parts Corp. (1968) 392 U.S. 134, 141-142, 88 S. Ct. 1981, 20 L. Ed. 2d 982 (parent and wholly-owned subsidiary); Timken Roller Bearing Co. v. United States (1951) 341 U.S. 593, 598, 71 S. Ct. 971, 95 L. Ed. 1199 (parent and partially-owned subsidiary).) At the other extreme, we have held that regardless of internal corporate structure, there can be no conspiracy as long as only one corporation is involved. (Joseph E. Seagram & Sons, Inc. v. Hawaiian Oke & Liquors, Ltd. (9th Cir. 1969) 416 F.2d 71, 80-84.) The heretofore unanswered question is what, if any, limits should be and can be drawn on Section 1 conspiracies in the separate incorporation situation.

Antitrust law is concerned with the concerted action of distinct economic entities. In any case, whether such action has occurred turns on the particular facts.4 Separate incorporation is just one among many factors; it may be significant in an antitrust sense or it may be only a technicality, a byproduct of decisions with no antitrust impact. The corporate structure itself determines whether there are separate units or one entity. (In re Penn Central Securities Litigation (E.D.Pa.1973) 367 F. Supp. 1158 (subsidiaries each covered a different geographical region); I. Haas Trucking Corp. v. N. Y. Fruit Auction Corp. (S.D.N.Y.1973) 364 F. Supp. 868; See also, Beckman v. Walter Kidde & Co., Inc. (E.D.N.Y.1970) 316 F. Supp. 1321, aff'd (2d Cir. 1971) 451 F.2d 593.)

DRI and BAPCO form a single unified structure. The relationship between them far exceeds DRI's mere ownership of the subsidiary's stock, and therefore, this is not a case of parent and independent subsidiary, but of a single business unit separated only by the technicality of separate incorporation. The same individual, Sparks, is controlling shareholder of DRI, president of both corporations and publisher of all five newspapers. Both corporations have the same individuals in charge of important operations, such as circulation. The two alleged individual co-conspirators. Chilcote and Cleland, are employees and/or officers of both firms. All three daily newspapers have the same sports page, financial page, T.V. log, "Night and Day Around the Bay," editorial page (except for two editorials per week), and substantial amounts of news and advertising. All the composition work for the common features is done at one plant. Moreover, DRI and BAPCO do not compete or hold themselves out as competitors.5 The lack of intraenterprise competition (or representations thereof) is not dispositive of the conspiracy issue, but it is an additional factor in our decision. Its importance is particularly evident in this case, where the corporations are horizontally related. Since a horizontal conspiracy would attempt to restrain interbrand competition, the fact they do not compete and that plaintiffs have alleged no interbrand restraint strongly suggest there was no conspiracy.

Arguably on these facts the two corporate units were incapable of conspiracy as a matter of law. It is unnecessary for us to decide that question, however, because the same facts prevent successful attack on the district court's factual findings that there was no conspiracy.

A. Horizontal Restraints.

Even if the corporate units were capable of conspiring, plaintiffs have failed to show that the concerted action effected a horizontal restraint on trade. A termination is not unlawful because of some adverse effect on the distributor's business, even if the effect is the elimination of the distributor from the market. The complaining distributor must show that the refusal to deal was intended to or did bring about some restraint of trade beyond the loss of business suffered by the distributor or the market's loss of a distributor-competitor. (E. g., Bushie v. Stenocord Corp. (9th Cir. 1972)460 F.2d 116, 119; Ricchetti v. Meister Brau, Inc. (9th Cir. 1970) 431 F.2d 1211, 1214.) Plaintiffs rely almost exclusively on Industrial Building Materials, Inc. v. Interchemical Corp. (9th Cir. 1970) 437 F.2d 1336. Industrial was an independent distributor of sealant products manufactured by Presstite. Presstite had a very strong position in the sealant market. Industrial alleged that Presstite had tried to force it out of business by selling directly to Industrial's customers, making some sales below the wholesale price to Industrial, and hiring one of Industrial's key sales people. Presstite argued that it could legally terminate Industrial if it chose to and therefore it could terminate it by competing. We reversed a summary judgment in favor of Presstite. We held that elimination of a competitor can be an unreasonable restraint when either (a) "unfair" predatory methods are used to force out the competitor, or (b) there is a market structure in which loss of a distributor will enhance a manufacturer's market power and tend to diminish the ability of other manufacturers to compete.

Plaintiffs proved neither of the elements which might bring them within Industrial. Predatory tactics usually have little or no social or economic justification, and can potentially harm the economy by producing monopolies. (See Turner, "Antitrust Policy and the Cellophane Case," (1956)70 Harv.L.Rev. 281, 305.) The terminations here, however, were pursuant to a contract which gave either party the right to discontinue the relationship; there was no provision for compensation or for the sale of the distributorship and the exercise of the contractual right can hardly be deemed "unfair." Moreover, if plaintiffs' contention were accepted, a manufacturer could be prevented "from ever replacing a system of independent distributors with its own system of direct sales," a result that even Industrial did not endorse. (437 F.2d 1343; see Cartrade Inc. v. Ford Dealers' Adv. Ass'n (9th Cir. 1971) 446 F.2d 289, 294.) Furthermore, any of DRI/BAPCO's solicitations after the initiation of suit, even if unfair, as alleged, are irrelevant; the relationship among the newspapers and the distributors was continued only to preserve the status quo during trial.

Plaintiffs also failed to prove that defendants' acts had any actual horizontal anticompetitive effect. Plaintiffs would have had to show DRI/BAPCO's dominant position in some relevant market6 and an actual restraint within that market. Plaintiffs claim that the relevant market is community newspapers, which Sparks unquestionably dominated, as opposed to all newspapers. We do not think they had to produce a numerical value of the cross-elasticity of demand to show community newspapers is a distinct submarket. Proofs of the factors set out by the Supreme Court in Brown Shoe v. United States (1962) 370 U.S. 294, 325, 82 S. Ct. 1502, 8 L. Ed. 2d 510, would have sufficed: industry and public recognition, peculiar characteristics or use of product, distinct customers, distinct prices, and sensitivity to price change. Plaintiffs did not produce any evidence on these factors. Their reliance on defendants' admission that Sparks' papers must be priced below the four major Bay Area newspapers in order to compete tends to refute their premise; it shows substitutability and thus one market. Plaintiff's sparse argument leaves many questions unanswered. These include: To what extent does the news, sports and feature coverage of the community papers overlap with the coverage in the metropolitan and satellite-city papers? Do the non-community papers distribute local editions or have special sections for particular communities? What is the extent of penetration of the non-community papers in the town served by the Sparks' publications? Is there a significant number of readers who subscribe to both types of newspapers, or do consumers tend to choose only one type?7 The answers to the questions will not produce a precise measurement of the cross-elasticity of demand, but they would provide the groundwork for a reasonable determination of the newspaper product market in the communities around the Bay. Only after that determination is made could a court draw any conclusions concerning the market dominance of DRI/BAPCO.

Finally, plaintiffs have been even less diligent in showing an actual restraint in that market. The district court found that no plaintiffs had ever carried a competing paper and two had declined an opportunity to do so. Moreover, plaintiffs failed to show that internal distribution costs are a substantial barrier to entry in the newspaper business, or that a demand for distribution services could not be met by independent entrepreneurs not currently in the newspaper distribution trade.

Since the plaintiffs have only shown that the terminations might restrain trade, but not that they actually did, they are not entitled to reversal on this point.

B. Termination in Furtherance of the Price Fixing Conspiracy.

Plaintiffs claim that the terminations were made in order to eliminate the uncooperative dealers and permit DRI/BAPCO to continue to control the resale price of its newspapers. The price-fixing scheme involved two different resale prices: the distributor's price to carriers (indirectly restrained) and the carriers' price to subscribers (directly restrained).8 As to the first, indirect restraint on the distributor's resale price, it is difficult to argue that the terminations were made to continue the restraint.9 In the usual refusal to deal situations, termination or threat of termination is unlawful because it coerces the distributor to adhere to a fixed price. Elimination of a nonconforming dealer notifies all other dealers that adherence to the manufacturer's resale price will be enforced. Similarly, a threatened termination or a termination coupled with a resumption of the relationship (upon a distributor's expressed willingness to adhere to the resale price) effectively restrains the distributor from setting his own resale price.

In Knutson, however, there was no selective termination of miscreant dealers; the entire wholesale level of distribution was replaced by an employee system. Thus, the terminations did not further the restraint on the plaintiffs' resale price. Sparks did not merely threaten to convert in an attempt to coerce adherence to his price; after the full conversion, no distributor was being restrained. There is, of course, no question that DRI/BAPCO now controls the price at which papers are sold to the carriers; such control results from the vertical integration. But that control is exercised by any firm (e. g., the other Bay Area newspapers) that performs its own distribution services.10

The dealership agreements permitted termination by either party after proper notice; there was no restriction on the reason for termination. If one party believed, as Sparks did, that the relationship was no longer advantageous to him, he had complete freedom to terminate it, as long as the terminations did not perpetuate the violation. In Germon v. Times Mirror Corp. (9th Cir. 1975) 520 F.2d 786, the court held that a "termination might be enjoinable even if done pursuant to contract, if the contractual clause relied upon were being used to foster an unlawful competitive scheme." (Id. at 788.) Similarly, in Noble v. McClatchy Newspapers (9th Cir. 1975) 533 F.2d 1081, 1086, we were willing to enjoin or award compensation for a termination if it were made because of a refusal to comply with territorial restrictions. In both cases, however, the issue before the court was the termination of a single distributor and the question was raised whether termination was being used as an enforcement mechanism for an unlawful restraint.11 As to the restraint on the distributor's price, no such question arises in Knutson ; a complete conversion is not an enforcement mechanism, but a choice of an alternative form of distribution.

As to the second price restraint, that on the retail price to subscribers, plaintiffs claim that the terminations were used to foster an anticompetitive scheme. They argue that Sparks eliminated the distributors so that his newspapers could influence directly the prices charged by the carriers and the remaining adult independents (motor route, retail account, and street sale dealers). Since the anti-competitive goal of the price-fixing scheme was to place a ceiling on the subscription price, the terminations could be in furtherance of the scheme if Sparks continued to unlawfully restrain the price of the newspapers.12 That is, as long as there are permissible ways for Sparks to control subscription prices, an intent to control or actual control does not necessarily violate Section 1. If Sparks' post-termination dealings with the remaining independents are not violative of Section 1, then the terminations which facilitated those dealings were not in furtherance of a price-fixing conspiracy.

While the uncontested facts demonstrate Sparks' unalloyed intent to establish uniform prices for his newspapers, the totality of his actions does not amount to the requisite quantum of coercion: no "meaningful event depend(ed) on compliance or non-compliance" with his requests of the carriers. (Butera v. Sun Oil Co. (1st Cir. 1974) 496 F.2d 434, 437.) Although the new carrier contract makes no mention of a resale price, the carriers have been sent letters which "strongly recommend" reselling at the price suggested in the newspapers.13 The letters also attempt to show a unity of interest between the newspapers and the carriers since the latter can profit from the former's promotional efforts, which are said to "depend on a uniform subscription price." The letter clearly states, however, that an offered bonus for each new subscriber does not depend on charging the suggested rate. Taken in its entirety, the letter does not amount to coercion, but relies on "individual self-interest to bring about general voluntary acquiescence." (United States v. Parke, Davis & Co. (1960) 362 U.S. 29, 46-47, 80 S. Ct. 503, 513, 4 L. Ed. 2d 505.)

The remainder of plaintiffs' evidence of a continuing restraint is based on a theory of coercion-by-example. The terminations, the treatment of plaintiff-employees, and the decision not to employ most of the plaintiffs are said to be disciplinary measures that carry an ominous threat to independents who might be tempted to sell above the suggested price. Plaintiffs are suggesting that Sparks' treatment of the distributors satisfies the Schwinn coercion standard of a "communicated danger of termination." (United States v. Arnold, Schwinn & Co. (1967) 388 U.S. 365, 372, 87 S. Ct. 1856, 18 L. Ed. 2d 1249.) They have not, however, provided any evidence of direct communication concerning the potential fate of nonconforming independents; they rely exclusively on possible incidental effects of activities the district court found not to be independent violations. Moreover, plaintiffs overlook the fact that Sparks' initial response to the distributors' lawsuit was to integrate and offer employment to each distributor, including the plaintiffs. A communicated danger of employment does not yet qualify as coercion in the context of resale price maintenance.

Plaintiffs have thus failed to show coercion of the remaining independents and have, therefore, not proven that the terminations were in furtherance of a retail price-fixing scheme. Nevertheless, the events prior to, during, and after the trial have generated a substantial amount of confusion concerning the independents and the legitimate courses of action open to Sparks. While this potentially coercive "fallout" does not justify an award to plaintiffs for the terminations, it might warrant the district court, it it chose to do so, to require Sparks to inform each independent of his or her right freely to resell the newspapers at prices other than the suggested price. This would assure that there is no residual anticompetitive effect from the terminations without penalizing Sparks for his lawful actions.

Territorial Restrictions.

The distributors cite three factors to support their claim of an unlawful restriction on areas of resale. First, they argue that the territorial reference in the dealers' agreement was a contractual restraint of trade. Second, they point to four instances in which a plaintiff-distributor's territory was realigned (a "split"), pursuant to the contractual provision for renegotiation. In at least one case they claim that the split was foisted on the distributor against his will. Finally, they contend that the wholesale newspaper price was manipulated to discourage sales outside a distributor's assigned territory. The wholesale prices to distributors varied, depending upon what the publisher and distributor believed was a reasonable income for a route. Once an income figure was decided upon, a dual rate was used, with one price per subscriber up to a certain number of subscribers, and a higher rate for each subscriber beyond that. If a contract expressly prohibits sales outside a designated territory, it is unlawful whether or not coercive devices have been employed to enforce it. In the absence of an express contractual agreement, proof of coercion that the manufacturer is "firm and resolute" in insisting on compliance with any ambiguous or implicit limitations on dealer freedom is necessary to show unlawful restraint. (See United States v. Arnold, Schwinn & Co., supra, 388 U.S. at 372, 87 S. Ct. 1856.) In Beverage Distributors, Inc. v. Olympia Brewing Co. (9th Cir. 1971) 440 F.2d 21 (territorial and customer restrictions), we held that an exchange of letters stating the distributor's intention to restrict itself to manufacturer's approval was not a contract and supported a jury's finding that there was no effort to impose the alleged restraints. (Id. at 30.) In Gray v. Shell Oil Co. (9th Cir. 1972) 469 F.2d 742 (price fixing), there was no contractual restriction and the court framed the question as whether the distributor was deprived of his free choice by some affirmative conduct of the manufacturer. (Id. at 747.)14

Thus, when the contract is ambiguous, the inquiry must focus on the seller's conduct. Designations of areas of primary responsibility or mere suggestions of territory without an express prohibition do not, by themselves, constitute a violation. A reference to territory or primary responsibility, however, may transgress Section 1 if the plaintiffs can show a course of conduct by which the manufacturer has prohibited sales outside the territory. (Hobart Bros., supra, note 14, 471 F.2d at 899-900.) The manufacturer's enforcement efforts need not be a blatant coercion. A restraint is present if "some course of action is undertaken or threatened contingent on the willingness or unwillingness" of the distributor to adhere to the restriction; the action "must involve making a meaningful event depend on compliance or non-compliance." (Butera, supra, note 14, 496 F.2d at 437.)

The contracts in Knutson do not expressly prohibit extraterritorial sales.15 They designate a district number which publisher and distributor apparently understand to define a specific area, and provide that the dealer will be supplied "such quantities (of the newspaper) as he shall order (at certain rates) to supply the needs of his territory." As the district court recognized, this is a borderline case. (383 F.Supp. at 1368.) Although there is no express prohibition, there is a strong implication that a distributor is confined to a specific area. Nevertheless, the contract seems sufficiently ambiguous to necessitate a determination whether the newspapers made efforts to enforce the implied restriction.

On its face, the dual price structure appears to be a device for confining a distributor to his assigned area. When a dealer took over a territory his wholesale price was calculated by the following method:

1. The number of subscribers in the area is multiplied by the subscription rate, giving the gross income produced by the area.16

2. The carriers' total income is subtracted from this amount, giving the remaining income to be divided between the newspaper and the dealer.

3. The newspaper and the dealer agree on a reasonable income from the area, taking into consideration the size, density, and difficulty of the area.

4. The remaining amount (the newspaper's income) is divided by the number of subscribers, giving the first wholesale rate to the dealer, i. e., the "base rate."

The base rate was initially charged for the number of subscribers in the territory at the time the territory was acquired. A higher rate was charged for each additional paper. Cleland testified that the reason for this was (1) to give a reasonable income to each dealer for his territory (the base rate calculations) and (2) to help the newspaper recover its circulation costs (the higher rate).

Whatever the newspapers' justification for the dual rates, charging a higher rate for newspapers above the amount needed for the territory inhibits extraterritorial sales. Not only does the dealer pay a higher price for additional papers, but he is also likely to incur greater costs in acquiring and serving customers outside his territory. This profit squeeze, resulting from the dual rates, inhibits dealer sales beyond his assigned area.17

Nevertheless, the dual rate structure is not, by itself adequate to meet the restraint standard of making "a meaningful event depend upon compliance or non-compliance with the restriction." A similar situation arose in Butera, supra. There, the oil company used a dual pricing system for sales to its dealers: the "tankwagon price" was the base rate, but a competitive allowance (i. e., a price reduction) was given on gasoline sold in areas and at times when the local competitive situation required retail sales at a lower price. (496 F.2d at 435.) The company also varied its suggested retail price depending on the local conditions. The First Circuit refused to prohibit the practice as unlawful price fixing because " 'adherence to a suggested price (was not) the quid pro quo for . . . receiving (the) competitive allowances.' " (Id. at 437, quoting Lehrman v. Gulf Oil Corp. (5th Cir. 1972) 464 F.2d 26, 39.) Although the wholesale price structure clearly narrowed the range in which a dealer could price, the court refused to constrain the company's wholesale pricing decisions, as long as the wholesale prices were applied without regard to a dealer's adherence to the alleged restraint.

In Knutson, the dual price applied to all additional newspapers, whether sold within or without the territory. Of course, the initial determination of the break-point for the base rate was made on the basis of the number of subscribers within the territory, but the higher rate applied regardless of whether additional customers came from the territory or outside it.

The interaction between the splits and the dual prices is more problematic. As pointed out by the district court, the mere fact of a split does not show that territorial restrictions were imposed on the dealers. (383 F.Supp. at 1368.) Areas of primary responsibility are as subject to realignment as exclusive territories. The question is whether a dealer was restrained from selling outside his area, whatever that area happened to be. When a dealer resists a split, however, and is coerced into accepting it, the realignment can be construed as forcing a distributor to cease sales in the area. It should be noted that once a split is effected, the carriers of the lost area are given to the new dealer and new subscribers secured by professional solicitors (hired by the newspapers) go to the new dealer. Since nearly all new subscribers are acquired by the carriers or the solicitors, after a split the new dealer is unlikely to continue any operations in the lost area.

Plaintiffs have not provided any substantial evidence of blatant coercion, such as a termination threat, to accept a split. Even in the one instance where the dealer actively resisted the split (1970, Knutson), plaintiffs do not seriously claim that Knutson succumbed because of threats. But defendants admit and the district court found that "whenever a split resulted in a loss of circulation, the dealer's rate was adjusted to avoid any loss of income of the dealer." After Knutson's 1970 split, his rates were favorably adjusted. (383 F.Supp. at 1368.) This admitted manipulation of the wholesale rate (and thus the dealer's income) comes perilously close to giving a quid pro quo for refraining from sales in a lost area. If plaintiffs had shown some sort of pattern in which rates were adjusted downward when dealers sought to sell outside their territories, the Butera standard would probably be met. No such pattern has been shown. The contracts provided for renegotiation of the rates upon initiation by either party, and rates were at times lowered at dealers' request with no indication that territorial disputes were involved. The small number of total splits of plaintiff-areas (four) and the isolated instance of adjusting Knutson's rates when he resisted a split are inadequate to show an anticompetitive use of the dual rate structure. Although the system was clearly subject to abuse, plaintiffs have not shown that it was abused.

This is a very close case. It is undeniable that newspapers and dealers were operating under a system of territories; there is no evidence that any dealer attempted to sell outside his area prior to the suit. (383 F.Supp. at 1369.) But neither is there adequate evidence that these territorial divisions were imposed and enforced by the newspapers. It may be that the system was actually an implicit horizontal division of territories among the distributors or even a vertical-horizontal arrangement requiring a minimum of coercion from above. Each dealer had strong incentives to adhere to his territory, including ease of delivery and response to errors and reciprocal respect for territorial integrity from other dealers. Plaintiffs have chosen, however, to pursue a purely vertical theory and thus must show that they were restrained. In this they have failed. We conclude, therefore, that the district court's conclusion that the plaintiffs failed to prove that the territorial divisions were a part of the price-fixing scheme is unassailable on appeal.

Damages.

Plaintiffs claim damage due to lost profits.18 After they had initiated this suit, the seven Daily Review distributors and two of the four Argus distributors raised their subscription prices. Because the trial on damages began on April 16, 1974, the largest new-price period considered by the lower court was six months (September 1, 1973 to February 28, 1974); plaintiffs state in their brief that further data is now available. The proof of damages is based on a comparison of dealer profits between this "after" period and the "before" period when the restraint was in effect (1969-73). The district court held that the evidence was inadequate to prove either the fact of damage, i. e., that plaintiffs suffered some actual injury, or the amount of damage in a manner other than by speculation and conjecture. (383 F.Supp. at 1379.)

The district court found that the data for the "after" period was so "inherently suspect and patently artificial" as to destroy the probative value of the evidence. (383 F.Supp. at 1382.) Each Daily Review dealer increased his or her price in the first two months after the price restraint was removed. The increase yielded a higher profit per subscriber and a decrease in the number of subscribers. From these two figures, the Daily Review plaintiffs calculated a net increase in monthly income. This they multiplied times their number of months as a dealer to arrive at a total dollar amount of damages (ranging from $1,800 to $36,000). Plaintiffs claim that their estimates are conservative since not all of the circulation drop is attributable to the price increase and during the "after" period there were no attempts made to acquire new customers.

Two Argus dealers raised their prices on April 1, 1974, two weeks before the damage trial. Subsequent to the damage trial they filed affidavits giving the increased profits of their distribution businesses for the months of April and May. Despite the apparent availability of data for these two months in the "after" period, plaintiffs have used the average circulation loss of Daily Review dealers (15 percent) to calculate the two Argus dealers' net increase in profit for the "after" period. Plaintiffs state that the two Argus dealers did not raise their prices until April 1 (and did not claim damages after March 1, 1973) because from March of 1973 until February of 1974, the Argus suggested price was 25cents higher than the Daily Review price. Since the Daily Review was home delivered in the districts of the Argus dealers, they felt a price hike would lead subscribers to switch to the Daily Review.

Two of the remaining three Argus dealers made no price changes. The third increased his price on February 1, 1974. These three plaintiffs claimed damages based on a comparative study using the data of the Daily Review dealers and the Argus dealers' assertion that the price in the "before" period would have been 25cents higher had they been free of restraint. The ratio of 25cents to the actual Daily Review price increases was multiplied by the average monthly damages of the Daily Review dealers to calculate the damage per subscriber month of these three Argus dealers. This figure was multiplied by the dealer's number of subscriber months prior to March 1, 1973, to arrive at a total damage figure.

Different standards govern proof of the fact and proof of the amount of damages. (Story Parchment Co. v. Patterson Parchment Paper Co. (1931) 282 U.S. 555, 562, 51 S. Ct. 248, 75 L. Ed. 544; Flintkote Co. v. Lysfjord (9th Cir. 1957) 246 F.2d 368, 390. But see Weinberg, "Recent Trends in Antitrust Civil Action Damage Determinations," 1976 Duke L.J. 485, at 495-96.) Proof of the fact of damage is closely intertwined with proof of causation. Thus in Story Parchment, supra, the Court held that "the natural and probable effect" of the antitrust violation was depressed prices, so that if earlier prices had been fair and reasonable a plaintiff could recover for the difference in price. (282 U.S. at 561, 51 S. Ct. 248.) (See also Zenith Radio Corp. v. Hazeltine Research Inc. (1969) 395 U.S. 100, 124-125, 89 S. Ct. 1562, 23 L. Ed. 2d 129; Bigelow v. RKO Radio Pictures (1946) 327 U.S. 251, 264, 66 S. Ct. 574, 90 L. Ed. 652.) A plaintiff need not negative all possible alternative explanations for his decline in profits, but the defendants may show that other events caused the injury. (Zenith Radio Corp., supra.) The plaintiff must prove some damage, but "proving the fact of damage . . . is satisfied by . . . proof of some damage flowing from the unlawful conspiracy; inquiry beyond this minimum point goes only to the amount and not the fact of damage." (Zenith Radio Corp., supra, (1969) 395 U.S. 100, 114 n. 9, 89 S. Ct. 1562, 1570, 23 L. Ed. 2d 129.) Even as to this minimal quantum of injury, the standard is relaxed; otherwise, it would defeat the loose standard applied even to the amount of damages in antitrust cases. (See Bigelow, supra, 327 U.S. at 274, 66 S. Ct. 574 ("just and reasonable" damage estimates are permissible); Flintkote Co. v. Lysfjord, supra, 246 F.2d at 392.)

The Supreme Court has also established a relaxed standard for proving the amount of damages in an antitrust case once the fact of damage has been shown. In Story Parchment, the Court noted that where the violation makes difficult a certain determination of damages "it will be enough if the evidence show the extent of the damages as a matter of just and reasonable inference, although the result be only approximate." (282 U.S. at 563, 51 S. Ct. at 250; see also, Bigelow, supra, 327 U.S. at 264, 66 S. Ct. 574.) More recently, the Court in Zenith emphasized the "practical limits" on the standard of proof of damages in antitrust cases. (395 U.S. at 123-25, 89 S. Ct. 1562.) We followed and explained the reasoning of those cases in Flintkote, supra, 246 F.2d at 391:

"A study of the adjudicated cases in this area readily dispels any impression that this question of damages is governed by an application of the common law rule of reasonable certainty. The cases have long since departed from this rule in antitrust litigation.

"Preliminarily, it should be observed that the reasons underlying the evolutionary trend toward liberality in proving damages are grounded in logic and sound policy. Two principal factors have influenced the courts. First, the self-evident intangible nature of the subject matter. To ascertain what would have been is as difficult as trying to determine what should be.

"Second, the legal maxim that a wrongdoer should not profit by his wrong. In light of the intrinsic uncertainty surrounding this problem, the responsibility for which lies in large measure with the defendant found liable, it has long been felt that this presents an ideal situation for application of that doctrine." (246 F.2d at 391.)

The district court found two major flaws in the plaintiffs' proof. First, it said that they did not show a reasonable probability that, absent the restraints, they would have raised their prices during the "before" period. While recognizing that the plaintiffs need not "prove that they actually engaged in the potentially futile act of violating a price-fixing agreement," the court characterized plaintiffs' proof as "conjectural hindsight." Second, the district court rejected the fact-of-damage proof on the ground that the plaintiffs did not meet the Story-Flintkote criteria. We agree with the district court in respect of the Argus plaintiffs, but we disagree as to the other plaintiffs.

The fundamental difficulty with the district court's reasoning is that it creates a nearly insurmountable barrier to recovery in maximum price-fixing cases. A determination of what a dealer would have done is bound to involve speculation and "second-guessing." Even if plaintiffs had supplied the corroborating circumstantial evidence apparently sought by the trial judge, assertions of their past intentions would still entail "conjectural hindsight."

Rather than imposing the nearly impossible burden of proving what each dealer would have done if he had been free to make his own pricing decision, we assume that, absent evidence to the contrary, a dealer would have raised his prices had it been profitable to do so; that is, dealers are profit maximizers.19 This assumption merely amounts to a recognition that a " restraint" in fact restrains. The defendants can attempt to show plaintiffs would have kept their prices beneath a maximizing point despite their violative behavior. Contrary evidence might include dealer testimony that he would not have raised his price, or a showing by the defendants that the dealers had reasons other than the restraint for selling below a profit-maximizing price.

With respect to the non-Argus plaintiffs, Knutson is easily distinguishable from Story-Flintkote. The distributors were operating the same businesses in both time periods and there was substantially less uncertainty as to the comparability of the two periods. The only question is the effect of a price increase on circulation and these plaintiffs have provided data on the degree of circulation drop. The district court implicitly recognized the differences between Flintkote and the Knutson in that it did not address the comparability of the two periods, but focused on the reliability of the data from the "after" period. Yet, the factors upon which the court relied in finding unreliability, at least as to the Daily Review plaintiffs, are relevant to the amount rather than the fact of damage. In keeping with the dictates of Story Parchment, Bigelow, and Zenith, the Daily Review plaintiffs have shown that the reduced net profits were "precisely the type of loss that the claimed violations of the antitrust laws would be likely to cause." (Zenith, supra, 395 U.S. at 125, 89 S. Ct. at 1577.) The district court found a contractual resale price restraint and plaintiffs produced evidence that despite circulation drops their net profits would have been higher in some amount. Although there are some infirmities in their evidentiary showing, unlike Flintkote, the infirmities are not so significant as to call into question the fact of damage, but relate only to the amount of damage.

In respect of the Argus plaintiffs, we agree with the district court that these plaintiffs did not meet the-fact-of-damage Flintkote test. Both groups of Argus dealers relied on average Daily Review figures to compute their prices, and two of the Argus plaintiffs never raised their prices. Thus, the Argus dealers produced no independent data, and they did not offer any proof of the comparability of the Daily Review and Argus dealerships and markets. Unlike other plaintiffs, the Argus plaintiffs had a complete failure of proof of damages, not simply some deficiencies in the proof of amount of damages.

On remand of the amount of damages issue as to the non-Argus plaintiffs, the district court is not asked to accept unreliable evidence.20 Rather, the district court is to make factual findings as to the amount of damages as to which each non-Argus plaintiff has offered reliable evidence, having in mind that certitude is not required and that lack of certainty justifies scaling down the award but not total rejection. Defendants, of course, can offer proof that market conditions, intra- and interbrand forces of competition, or other factors would also have scaled down the plaintiff's claimed loss of profits.

Section 2 Attempt to Monopolize.

" The phrase 'attempt to monopolize' means the employment of methods, means and practices which would, if successful, accomplish monopolization, and which, though falling short, nevertheless approach so close as to create a dangerous probability of it." (American Tobacco Co. v. United States (1946) 328 U.S. 781, 785, 66 S. Ct. 1125, 1127, 90 L. Ed. 1575.) Although the completed offense of monopolization requires only a general intent to do the acts leading to a monopoly, "a specific intent to destroy competition or build monopoly is essential to guilt for the mere attempt." (Times-Picayune v. United States (1953) 345 U.S. 594, 626, 73 S. Ct. 872, 890, 97 L. Ed. 2d 1277.)

In Lessig v. Tidewater Oil Co. (9th Cir. 1964) 327 F.2d 459, we rejected defendant's argument that proof of a dangerous probability of success requires an evaluation of a firm's economic power in the relevant market:

"We reject the premise that probability of actual monopolization is an essential element of proof of attempt to monopolize. Of course, such a probability may be relevant circumstantial evidence of intent, but the specific intent itself is the only evidence of dangerous probability the statute requires . . ..

. . .

"When the charge is attempt (or conspiracy) to monopolize, rather than monopolization, the relevant market is 'not in issue.' " (327 F.2d at 474.) (Footnotes omitted.)Despite the substantial criticism of Lessig, it is still the law of this Circuit to which this panel is bound. The sole issue, then, in this attempt case, is the presence or absence of a "specific intent to destroy competition or build monopoly." In Lessig itself, the court relied on the defendant's Section 1 violations (resale price fixing and exclusive dealing) to support a possible inference of specific intent. (Id. at 475.) In Bushie v. Stenocord Corp. (9th Cir. 1972) 460 F.2d 116, 121, we reemphasized the importance of a Section 1 violation in inferring the requisite specific intent. (See also, Moore v. Jas. H. Mathews & Co. (9th Cir. 1973) 473 F.2d 328, 332.) More recently, we reaffirmed and expanded on the Lessig formula:

". . . dangerous probability may also be shown through proof of specific intent to set prices or exclude competition in a portion of the market without legitimate business purpose. This specific intent must be accompanied by predatory conduct directed to accomplishing the unlawful purpose. Ordinarily specific intent is difficult to prove and will be inferred from such anticompetitive conduct." (Hallmark Indus. v. Reynolds Metals Co. (9th Cir. 1973) 489 F.2d 8, 12.)

In sum, we require (1) only specific intent and (2) some illegal (under Section 1) or predatory activity from which specific intent can be inferred. Where the conduct is justified by legitimate business reasons or merely exemplifies a healthy spirit of competition, an intent to monopolize is more difficult to support. The court below refused to find the requisite specific intent. The question, therefore, is whether the acts of the corporate and individual defendants require an inference of specific intent. We discuss each form of conduct separately, but it should be noted that all of the acts should be viewed together in determining whether there was an attempt to monopolize. (Morning Pioneer, Inc. v. Bismarck Tribune Co. (8th Cir. 1974)493 F.2d 383, 387.)

The district court held that the papers had violated Section 1 by fixing resale prices in their contracts with the distributors. Lessig, Bushie, and Moore all endorse the proposition that a Section 1 violation itself can support an inference of specific intent. The district court, however, refused to make the inference because there was no evidence of "predatory or knowingly unlawful activity" in the fixing of prices. The court characterized the illegal contracts as a "technical violation" and noted that the defendants took prompt action to comply with the law when notified by plaintiffs' counsel that the practices were unlawful. Assuming arguendo, that there is a requirement of "knowingly unlawful" conduct, there is no question that the defendants engaged in it. Price fixing has been illegal since the appearance of the Sherman Act in the nineteenth century; resale price fixing contracts were held to be illegal as early as 1911 in Dr. Miles; maximum resale price fixing was clearly outlawed in Kiefer-Stewart in 1951; and maximum resale price fixing specifically in the newspaper industry was found to violate Section 1 in Albrecht in 1968, the year before DRI/BAPCO adopted its illegal contracts. In the light of these cases beginning with the definition of the general offense of price fixing and culminating in the specific offense in the specific industry at issue in Knutson, the district court's determination that there had been no knowingly unlawful activity cannot stand. Even in the absence of predatory acts, the contracts themselves are unlawful and they accomplished the unlawful purpose of maintaining a maximum resale price.21

During the period from 1944, when he bought the Daily Review, until 1972, Sparks and his corporations bought 11 local newspapers and throwaways, six of which no longer exist.22 The mere fact that Sparks purchased numerous newspapers does not in itself lead to an inference of an intent to acquire a monopoly. But Sparks was not a diminutive Hearst; he did not build a kingdom of noncompeting newspapers in different areas. The papers acquired in 1945, the 1950's, 1962, 1965, and 1970 were all distributed in the Daily Review home delivery area. The 1972 acquisitions were competitors of either the Daily Rewiew, the Argus, or the Daily Review Shopping News. Three advertising throwaways were purchased in 1972 and then discontinued, thus leaving the Shopping News without their competition. Moreover, the 1972 purchases included a covenant that the seller corporation and its stockholders would not compete for a period of 10 years in the DRI/BAPCO areas of operation. Finally BAPCO was acquired for $300,000 only $20,000 of which could be allocated to the physical assets; the Village Pioneer was acquired for $3,000, but all that was acquired was "the name," and the name was later changed; and the 1972 papers were purchased for $800,000, one-half of which was the price of the covenant not to compete. Sparks' history of purchasing competitors for prices greatly out of proportion to the value of their tangible assets provides support for, but does not compel, an inference of specific intent.

The Audit Bureau of Circulation ("ABC") is a non-profit corporation which issues statements on its members' circulation and distributes the statements to advertisers and publishers who rely on the data for the sale and purchase of advertising space. Prior to the periodic determinations of circulation by the ABC, DRI would sponsor promotional contests in which dealers purchasing additional copies of newspapers for a specified time would be rewarded with trips or cash. The additional purchases would be recorded as paid circulation even if the dealers were unable to acquire new subscribers to purchase the papers. The "prizes" awarded for these contests were not reflected on the dealer's monthly statements as credits, nor were the ABC auditors informed of the contests. Plaintiffs also alleged that employees of DRI had set up a system of phony start orders whereby dealers would write (for a fee) start orders for nonexistent subscribers. Professional solicitors were also alleged to have participated in this scheme.

These uncommendable tactics might have anticompetitive effect because new entrants might be discouraged by the false impression that market strength of DRI would prevent successful entry. But an inference is also permissible that high circulation would actually attract competition into what was falsely presented as an especially juicy market. This factor thus becomes a neutral element in the search for the requisite specific intent.

In short, the Section 1 violation alone, or in conjunction with Sparks' newspaper acquisitions, the questionable promotional practices, and the padding of circulation figures would have supported a district court finding of specific intent, but that finding was not compelled and we cannot say that the district court's contrary determination was clearly erroneous.

The injunction is ordered dissolved upon issuance of mandate. Affirmed in part, reversed in part, and remanded for further proceedings consistent with the views herein expressed. The parties shall bear their own costs on appeal.

RUSSELL E. SMITH, District Judge (concurring and dissenting):

I would affirm.

I think that the trial court's findings of fact are sufficient and are supported by the evidence. The trial court here simply did not believe the plaintiffs' witnesses. The trial judge considered the plaintiffs' interests and motives in the case; the lack of corroborating evidence; the plaintiffs' self-contradictions; the times at which the claims here made were first made.1 The court concluded:

. . . Because of the paucity and doubtful credibility of the evidence on this question, plaintiffs have not satisfied their burden of proof on the first element of the fact of damage . . . . Knutson v. Daily Review, Inc., supra note 1, as 1381.

Absent believable witnesses, the plaintiffs who did have the burden failed unless their burden of proof was satisfied by the operation of some rule of law.

Perhaps that rule of law is stated by the majority in these words:

Rather than imposing the nearly impossible burden of proving what each dealer would have done if he had been free to make his own pricing decision, we assume2 that, absent evidence to the contrary, a dealer would have raised his prices had it been profitable to do so; that is, dealers are profit maximizers. This assumption merely amounts to a recognition that a "restraint" in fact restrains . . . (Footnote 19 of the majority opinion is deleted, and footnote 2 of this dissent is added.)

I am not aware of any rule of law which permits appellate courts to make assumptions of fact. An appellate court may create presumptions, even the mandatory kind which bind the trier of fact in the absence to the contrary. Although the opinion does not say so, that seems to be what the majority has done here. If so, then I disagree. I think it could be said that it is the universal intention of dealers to make a profit but that there is an intent to profit does not warrant the conclusion that all dealers are in fact "profit maximizers," i.e., that they will do all things necessary to make a maximum profit. Here the effect of the presumption is that the dealers, if free, would have considered raising prices as a means of increasing profits, would have taken the trouble to do so, would have risked the effect of a raise in prices in a competitive market, and the possibility that the publishers might have retaliated by raising their prices to the dealers.

The fact is that some dealers do not do all things necessary to maximize profits. That is why some dealers profit much less than others, and some go broke. Between the intent to profit and the production of a maximum profit lie the factors of energy, imagination, intelligence, and the willingness to take a risk. If the presumption created here goes as far as it must go to support the conclusion, then, in my opinion, it is artificial. I think the testimony in this case discloses the artificiality of it.3

To sum it up: I think the trial court did not believe plaintiffs' witnesses. I think that, in the absence of credible evidence, there was no rule of law compelling the trial court to find that the loss of profits had been proved.