Multiflex, Inc., Plaintiff-appellee, v. Samuel Moore & Company, and Eaton Corporation, Defendants-appellants, 709 F.2d 980 (5th Cir. 1983)

This is an antitrust case under sections 1 and 2 of the Sherman Antitrust Act,, , seeking damages under section 4 of the Clayton Act, . The plaintiff, Multiflex, Inc., is a relatively new corporation that manufactures hydraulic hose bundles for use in oilfield drilling equipment. The defendant, Samuel Moore & Co., is a long-established maker of similar hydraulic hose bundles and other technical industrial hoses. At trial, Multiflex sought monetary damages for conspiracy in restraint of trade and attempted monopolization. The district court entered judgment of $549,362.98 before trebling. We affirm in part, reverse in part, and remand.I. THE STAGE SETTING

A. The Industry in Perspective.

Hydraulic hose bundles are sets of long, high-pressure hoses that are bundled, or wrapped, with plastic or other sheathing substances to produce an integral, cable-like unit. The main use for these bundles is in the oilfield equipment industry, where they are used as a component of blow-out prevention (BOP) systems. BOP devices are sophisticated, high-pressure hydraulic systems designed to monitor oilfield activities to prevent blow-outs.

The market for BOP devices is dominated by three original equipment manufacturers (OEMs), which design, manufacture, install, and guarantee a virtual turn-key system. The OEMs quote a price to the end users for a complete system that includes hydraulic hose bundles and all other system components. The OEM benefits under this "total systems" approach because it can tack a profit margin onto its wholesale cost for the hose bundles and other system components. The end user benefits from the simplicity of buying from a single vendor and relying on a single guarantee. Although an end user occasionally buys original hoses or replacement bundles directly from the manufacturer, most sales of hydraulic hose bundles are made through the OEMs as middlemen.

B. The Parties in Perspective.

The company credited with establishing the hydraulic hose bundle market is Samuel Moore & Co. (Moore). Moore holds a patent on its method of bundling the cables, and has been at the forefront of helping the OEMs incorporate these hose bundles into their BOP systems. For many years up until 1978, Moore enjoyed a dominant position in the hydraulic hose market, controlling at least an 80% share of the U.S. market.

Bruce W. McConkey, the President of appellee Multiflex, Inc., worked for Samuel Moore & Co. from 1964 to 1974. His ultimate position was national sales manager of the Synflex division, the Moore unit that sold hydraulic hose bundles. In 1974, McConkey left Moore and became associated with Flow Products of Houston, Inc., an industrial distributor. Due to McConkey's close and friendly relationship with Moore, Synflex dropped all other distributors in the Houston area and appointed Flow Products its exclusive distributor in this major market. This arrangement continued amicably for three years.

In 1977, McConkey decided to start producing hydraulic hose bundles in competition with Moore. He contracted with Gates Rubber Co. to arrange for a dependable supply of hosing, and made other preliminary arrangements. The new company, called Multiflex, Inc., was incorporated in March, 1977. Its first business goal was to purchase basic hosing from Gates Rubber Co., sheathe the hosing into bundles, and sell the hydraulic hose bundles as oilfield equipment.

Moore learned of Multiflex's existence in May, 1977, at which point Moore and McConkey reached a peaceable agreement to terminate Synflex's distributorship with Flow Products due to conflicts of interest. Multiflex was still making preparations for production at that point, and did not achieve manufacturing capability until November 1, 1977.

C. The Competitive Process in Perspective.

Multiflex approached the three OEMs in late 1977, just before it developed production capability. It hoped to show that its superior product at lower cost was a better deal than Moore's bundles. But the OEMs placed no orders with Multiflex at this time, preferring to remain loyal to Moore. Earlier entrepreneurial attempts to enter this market had met a similar lack of success.

In October, 1977, a Dallas-based drilling company, the Sedco Co., was in the market for a hydraulic hose bundle. Multiflex approached Sedco to offer its product, and Sedco was interested in a price quote. Multiflex, however, claims it asked Sedco to obtain a quote directly from the OEM that would be manufacturing the complete device, because it did not wish to disturb the established channels of distribution. Sedco obtained a quote for a Multiflex bundle from the OEM.

Moore, learning of the Multiflex bid, then submitted a lower-than-normal bid directly to Sedco, bypassing its normal distributor network. The price was slightly higher than the OEM bid based on the use of the Multiflex bundle. Moore pressured the OEM to promote the Moore product. The evidence shows that the Multiflex product was at least equal in specifications to the Moore bundle, and that Multiflex, unlike Moore, was able to promise the required short delivery time. Nevertheless, a Moore bundle was used on the Sedco job. This lost sale was later used as evidence of Multiflex's "fact of damage" in its antitrust charges.

Multiflex made a few small sales in 1977, but decided that to remain competitive it would need to bypass the traditional marketing structure utilizing the OEMs and approach the end users directly. The end users, mostly offshore drilling contractors, were quite favorably impressed with the new Multiflex line, and 1978 sales totaled over $560,000, mostly to end users. Eventually, even the OEMs began placing some orders through Multiflex. Multiflex expanded in the U.S., and established a manufacturing facility in the United Kingdom. In 1979, Multiflex's sales totaled over $1,650,000, almost doubling Moore's 1979 sales of approximately $850,000. By 1979, Multiflex was the dominant supplier in the U.S. market for hydraulic hose bundles with over a 60% market share, while Moore's share dropped from over 80% to approximately 38%.

Throughout, Moore had kept a careful eye on the upstart competitor sired by its former employee. When Multiflex bought its first hose from Gates Rubber Co. and produced its first hydraulic hose bundles, Moore closely investigated the completed bundles to see if they infringed upon the Moore patent covering many aspects of hydraulic hose bundle design and manufacture. After an investigation and a call to its patent attorneys, Moore filed a patent infringement action against Multiflex on March 22, 1978.

Multiflex answered the suit by denying the patent infringement charges and counterclaiming with several antitrust charges. It later became clear that Gates and Multiflex had taken steps after the first few months of production to change the Multiflex design and eliminate any possibility of patent infringement. Moore eventually dropped its patent claims. Multiflex, however, continued to pursue its antitrust charges. Its amended complaint included claims under sections 1 and 2 of the Sherman Act,, , and the treble damages provision of section 4 of the Clayton Act, . The parties were appropriately realigned, and these antitrust claims constitute the present suit. Jurisdiction in the district court was proper under .

The case was presented to a jury in a trial lasting six weeks. The jury found the relevant market to be the U.S. market for hydraulic hose bundles; found section 1 liability and awarded section 1 damages of $382,963.07; found section 2 liability and awarded section 2 damages of $205,964.73. Multiflex accepted a slightly-reduced judgment of $549,362.98 and amended its pleadings to conform with the verdict. The trebled award was entered for $1,648,088.94, before costs and fees. The trial court denied motions for judgment n.o.v. or new trial. Moore files this timely appeal, .II. SECTION 1 CLAIM

Multiflex charges that Moore entered into a conspiracy in restraint of trade, in violation of section 1 of the Sherman Act. It alleges that Moore met with the three OEMs during the early days of Multiflex's existence to convince the OEMs not to enter into contracts with Multiflex.

A. The Applicable Law.

Under the Sherman Act, conspiring in restraint of trade is illegal, even if the conspiracy does not in fact lead to an actual restraint of trade. (1st Cir. 1932). Unlike a contract or other combination in restraint of trade, a conspiracy violates the Sherman Act even without proof of injury because of the surreptitious, pernicious effect a conspiracy ultimately can have upon a free market. Violation of the Sherman Act is therefore not dependent on damage, although no damages can be recovered unless proved. Federal law does not provide for automatic monetary damages as an antitrust remedy. (5th Cir.), cert. denied, (1982) (discussing Sec. 2(a) of the Robinson-Patman Act, (a)), (1977).

The term restraint of trade could not cover all possible restraints, or else the antitrust laws would eliminate the entire system of commerce. (1948). Rather, the term includes two types of unlawful acts. The major category, unreasonable restraints of trade, includes business practices that violate general principles of fair dealing and free trade. The legal standard, generally called the rule of reason, leaves the ultimate determination of reasonableness to a jury. The prohibited acts are those with a threatened impact on competition. (1919). The inquiry concerns effect on competition, not on competitors. (1962); (5th Cir.), cert. denied, (1981); (5th Cir.), cert. denied, (1980). The legality of most restraints of trade, including the vertical market restrictions at issue in this case, is determined under the rule of reason analysis.

Some violations, however, are so damaging to a system of free and open competition and so devoid of legitimate business purpose that courts have labeled them per se violations not subject to the rule of reason. (1958); (1940). These per se claims, including price fixing and horizontal exclusions from the marketplace, (1972), are not involved in the case before us. The parties have stipulated that all restraints of trade alleged in this case are subject to the rule of reason. (1977); (5th Cir. 1983).

Even though a conspiracy in unreasonable restraint of trade may be proved, to justify a private suit the plaintiff must additionally demonstrate injury under the standard of section 4 of the Clayton Act, . (11th Cir. 1982), cert. denied, (1983); (5th Cir. 1978). Under section 4, such a showing of injury requires proof of the "fact of damage" and some general proof as to the amount of damage. (5th Cir. 1976); (5th Cir. 1975), cert. denied, (1976). A violation without private injury to the plaintiff cannot support an award of money damages under section 4, even though the defendant's behavior could support injunctive relief, a monetary award in favor of another private plaintiff, or a judgment in favor of the United States as parens patria. (1969). With this general statement of the law as a backdrop, we turn to the section 1 claims involved in this case.

B. The Conspiracy and the Rule of Reason.

Multiflex alleged at trial that Moore met with the principal customers for hydraulic hose bundles (the three OEMs) and conspired to keep Multiflex out of the market by suggesting exclusive and profitable arrangements between Moore and the OEMs. The allegation is supported by several internal Moore memos. One such memo, dated October, 1977, acknowledges that Moore is no longer alone in the relevant market, and that the company must decide "just exactly how we are going to structure ourselves for this onslaught." It considers as one course of action a meeting with the three OEMs to "see what we can all do to keep these other guys out of it."

Another Moore memo, dated January, 1978, reports meetings in that same month between Moore and the OEMs at which the business activity of Multiflex was discussed. The memo claims that at these meetings, two of the three OEMs agreed not to buy bundles from any source other than Moore. All three OEMs stated, according to the Moore memo, that they hoped the days of high profit margins would return, "but for the most part it is a war situation now." The memo concluded that Moore believed the OEMs would support its plans to keep Multiflex from gaining a toehold in the market:

We are thinking over everything that was said there and how the three meetings went, and the probability is that we will go forward with the plan as we took it to Houston. We will have the support of [two of the OEMs]; however, we are not certain of the support we'll get from [the third]. It is essential to counteract McConkey's work and to try and cause him to lose every single thing that he bids, as he will only get stronger. This may indeed cause us to go much, much lower than we would care to go; however, the objective is to try and remain in the number one position in this business.

The various Moore memos suggest that Moore initiated a conspiracy with the three OEMs to exclude Multiflex from the market and retain its market dominance. In return, Moore offered the OEMs the promise of higher profit margins by agreeing to limit its direct sales to end users and to sell at "substantially higher prices to end users than to the OEMs."

Witnesses at trial provided further evidence of restraints in trade performed for the sole purpose of destroying Multiflex. An antitrust conspiracy is rarely shown by direct evidence, and usually is proved by inference and suspicion. (1948). It is "enough that, knowing that concerted action was contemplated and invited, the distributors gave their adherence to the scheme and participated in it." (1939). In this case, however, the plaintiff was able to produce the defendant's own "smoking guns" with the result that there was ample evidence of a conspiracy to restrain trade.

Even though the existence of conspiracy is demonstrated, the proof must still show the intended restraint was unreasonable. The question of reasonableness is a fact question for the jury. (1977). There is sufficient evidence to uphold the jury's determination in this case that the restraint was unreasonable. The record suggests that Moore set up its meetings with the OEMs in an attempt to keep Multiflex out of the market. The OEMs realized that compliance with Moore's proposals would mean greater profits for themselves as direct-dealing competitors to Moore were eliminated. The goal of the conspiracy was to eliminate one of the only two manufacturers. While the conspiracy was in force, customers were discouraged from buying from anyone other than Moore. This evidence forms a proper basis for a jury to find an unreasonably dangerous effect on competition.

As in any case with sufficient merit to survive motions for summary judgment and reach a jury, there is evidence in this case to support a contrary result. Moore claims it was only meeting with its customers to apologize for the occasional need to make direct sales to end users with the effect of "freezing out" the OEMs as to those sales. It claims the OEMs were amenable to exclusive dealings with Moore based only on lawful considerations such as reliability in deliveries and quality of products. However, our role as an appellate court is not to adjudge the evidence de novo. We must affirm factual findings that are in accordance with the law and supported by credible evidence. There is sufficient credible evidence that Moore played the organizing and dominant role in a conspiracy undertaken to achieve an unreasonably dangerous effect on competition. We find that the record supports the claim of violation of section 1 of the Sherman Act.

To obtain any private monetary award based on the violation of section 1, a plaintiff must prove an injury to its business or property, the so-called "fact of damage." The fact of damage showing does not require the antitrust plaintiff to prove with particularity the full scope of profits that might have been earned in the absence of unlawful activity in restraint of trade. (1946); (5th Cir.), cert. denied, (1972). It requires only that a plaintiff show with some particularity an element of actual damage caused by the defendant's violations of the antitrust laws. We find that this prerequisite to a damage award was not met in the conspiracy proved under Section 1 of the Sherman Act.

To show the fact of damage, Multiflex points to the October, 1977, Sedco bid, where Multiflex presented substantial evidence that it was not given an opportunity to compete fairly for the job. Multiflex showed that it would have provided quality bundles at a lower cost than Moore with guaranteed fast delivery as required, but that Moore's anticompetitive maneuvers kept Multiflex from winning the bid. Overall, the evidence supports the inference that Multiflex would have earned this Sedco job in the absence of Moore's misconduct. The evidence also shows, however, that the OEM manufacturing the complete device for Sedco did not join with Moore in the attempt to freeze out Multiflex. The OEM quoted Sedco a competitive price on the Multiflex bundle and refused pressures from Moore employees to badmouth the Multiflex product. The angry memos presented at trial between Moore and the OEM regarding the Sedco job make it clear that the OEM was not on Moore's side of the fence during the Sedco bidding.

The evidence Multiflex has presented to show the existence of conspiracy points to the January, 1978, meeting between Moore and each of the OEMs as the starting date of the conspiracy. Moore officials may have had a conspiracy in mind at the time of the Sedco episode in October of 1977, or earlier, but there is no showing a conspiracy was born until the OEMs joined into conspiracy with Moore in January, 1978. Since the bidding for the lost Sedco job took place before the establishment of the conspiracy, this episode cannot establish a fact of damage resulting from a section 1 conspiracy claim. Yet this episode is the only one upon which fact of damage was based.

There must be a causal connection between the violation shown and the injury asserted to support a private sector award of money damages. (5th Cir. 1981), cert. granted, (1982); (5th Cir.), cert. denied, (1978). At the time of the Sedco bidding, Moore was acting alone and not pursuant to a contract, combination, or conspiracy. Therefore, Multiflex has not shown adequate "fact of damage" related to the conspiracy claim and cannot recover treble damages under section 4 of the Clayton Act based on its section 1 conspiracy claim. The district court's award of damages under the section 1 theory must be reversed.

A. The Applicable Law.

Multiflex's other antitrust claim is one of attempted monopolization under section 2 of the Sherman Act, . Monopolization as a legal offense involves more than mere possession of market power. A company can hold a monopoly through one of several legal means, including legislative grant, valid patent rights, or simply through business acumen or a superior product. (1966). Monopolization as a legal offense requires that a company engage in overt anticompetitive acts to acquire or maintain monopoly power.

Attempted monopolization under section 2 is usually defined as an unsuccessful attempt to achieve monopolization. (1946). The offense requires not a mere or simple risk of monopolization, but a "dangerous probability of success" in the relevant market. (1951); (5th Cir. 1978). This dangerous probability can exist even though the attempt ultimately is unsuccessful. Indeed, it is only the failure of the scheme that keeps the charge from becoming actual monopolization.

Attempted monopolization also requires proof of a specific intent to monopolize, unlike the offense of actual monopolization, which is unlawful even without proof of specific intent. (1953). Finally, the charge demands evidence of overt acts in furtherance of the scheme to monopolize. Claims of attempted monopolization may overlap section 1 claims of restraint of trade, but the charges are not identical: a restraint of trade may be reasonable and therefore not prohibited by section 1, yet still may constitute an attempt to monopolize if a specific intent to monopolize is shown. (1948).

Under an attempted monopolization theory, a defendant is liable for competitive damages if they are caused by the unsuccessful attempt to monopolize. A private sector suit for such damages is subject to the requirements set forth in section 4 of the Clayton Act,, just as is a section 1 claim. To support private recovery, there must be some indication of a particular injury caused to the plaintiff's business or property by the violation. A violation accompanied only by generalized damage to the marketplace but no particularized injury could support a request for injunction, or could be pursued by government agencies in an enforcement action, but a private suit for damages can be supported only on a showing of particularized economic injury. Multiflex points to the Sedco affair, discussed earlier in this opinion, as the specific episode of its damage.

The claim of attempted monopolization in this case is an unusual one. The evidence shows that Moore enjoyed market dominance in the hydraulic hose bundle market, due in large part to its valid patent covering several stages of the assembly process. Multiflex's charge is that Moore, while holding this lawful market dominance, acted intentionally and overtly to exclude Multiflex and illegally maintain a monopoly. This case, then, is not an attempt to create monopoly power, but an attempt through illegal means to protect a lawful monopoly against rising competitive challenge. Such a claim could constitute the offense of actual monopolization under section 2. It fails as actual monopolization only because Moore's efforts were unsuccessful. Rather than maintaining market dominance, Moore's market share plummeted from over 80% to approximately 38% in just over a year. Multiflex is arguing, in effect, that its own market share would have jumped from almost zero to over 80%, rather than "merely" to the 62% share it achieved in its first two years of operation.

At first sight, it might strain credulity to consider imposing monopolization damages against a firm that rapidly lost, rather than gained, market dominance. The claim is more surprising with the recognition that the plaintiff in this case actually gained in market power during the relevant period. Even an unsuccessful attempt to monopolize, however, can cause substantial injury both in the marketplace and to a competitor. The central issue in this section 2 claim is whether an unsuccessful attempt to maintain monopoly power through anticompetitive acts can support a section 2 claim under any circumstances and, if so, whether the claim is cognizable under section 4 of the Clayton Act in this private suit.

B. Attempted Monopoly and Lack of Success.

Moore urges on appeal that section 2 cannot apply to this case under any circumstances. It points out that its market share dropped from almost complete control of the U.S. market to a 38% share as Multiflex, the company it allegedly was overpowering, grew from nothing to over a 60% share in only two years. With such overwhelming evidence that Moore lost rather than gained any market power it might have had, and that Multiflex prospered rather than suffered during the relevant period, Moore urges that as a matter of law there could be no showing of a "dangerous probability of success." Moore relies heavily on (2d Cir. 1980), where the Second Circuit ruled that a loss in the defendant frozen waffle maker's market share from 48.3% to 33.0% from 1969 to 1974 could not, as the only evidence presented, support a claim of a dangerous probability of success under a charge of attempted monopolization.

Multiflex, on the other hand, contends that its ultimate success in the marketplace should not preclude us from finding a violation of section 2. Although it might have pulled out of serious ultimate injury from Moore's conduct, it claims that Moore's anticompetitive acts threatened its ability to continue as an ongoing concern when it was a fledgling venture. Its evidence goes beyond share of the marketplace. It charges that Moore disparaged Multiflex to bankers and prospective customers, so that Multiflex could not obtain the bank financing it needed to remain in business or bid on equal footing for jobs to the OEMs or the end users. Multiflex claims its problems created a "dangerous probability of success" at that time. Only through a substantial infusion of capital from McConkey's brother-in-law and a radical shift from normal industry marketing strategies was Multiflex able to overcome what it characterizes as a willful, unlawful campaign to crush Multiflex at all costs. Multiflex asserts it was damaged by Moore's acts viewed from a short term perspective, even if it ultimately recovered and overwhelmingly prospered in the marketplace.

Characterization of the claims in this case is not an easy matter. We do not conclude, nor have we found a case to suggest, that a drop in defendant's market share alone can defeat a claim of attempted monopolization. As the Ninth Circuit stated in (9th Cir.), cert. denied, (1977):

The record indicates that IBM's market share is declining. A declining market share may reflect an absence of market power, see United States v. International Harvester Co., supra, 274 U.S. [693,] at 709, [at 754, (1947) ]; [, 295 n. 1,] (1920), but it does not foreclose a finding of such power., 328 U.S. [781,], at 794-95, [at 1132, (1946) ].

(7th Cir. 1981), cert. denied, (1982) and cases cited.

In the case, the Second Circuit dismissed the claims of attempted monopolization when the only evidence of antitrust injury was the defendant's drop in market share, which alone, of course cannot possibly establish wrongdoing by a defendant. Further, the decline in market share was a gradual drop continuing over a period of years. This data, the only evidence presented, could not establish the plaintiff's case. We agree, therefore, with the Second Circuit's holding. The Second Circuit the following year in (2d Cir.), cert. denied, (1981), made clear that a drop in market share does not defeat an attempted monopolization claim when there is evidence to support the claim. The court said: "We do not doubt the significance of market share evidence as an indicator of either the presence or absence of monopoly. As the cases and commentators have cautioned, however, the true significance of market share data can be determined only after careful analysis of the particular market." (9th Cir. 1980), cert. denied, (1981) (large and increasing market share, under proper circumstances, could demonstrate market power).

The fact that a plan violative of the spirit of the antitrust laws ultimately fails does not inexorably lead to the conclusion that there was no "dangerous probability of success." As the court said in (7th Cir. 1981), cert. denied, (1982): " [T]he Sherman Act's prohibition against attempted monopolization does not require that the attempt in fact ripen into an actual monopoly. It is the attempt which is the offense." The time to examine "dangerous probability" is when the acts occur. We do not accept the use of hindsight plus evidence of plaintiff's successful response to the defendant's acts to exonerate an antitrust violator who did cause damage to the plaintiff.

We find adequate evidence to support the jury's implicit finding that there was a "dangerous probability of success," when we look at the situation at the time of the violation. Viewing this appeal of a jury verdict in the manner most favorable to the verdict, we can state that Samuel Moore & Co. memos showed a clear intent to freeze Multiflex out of the market. We note that Moore told Multiflex's bankers and competitors that Multiflex's products were inferior, that the company was closing down, and that the plant had been shut in anticipation of bankruptcy with Pinkerton security guards posted at the door. Viewing the danger from the proper time perspective, we can find a sufficiently dangerous probability that Moore would have maintained its monopoly power through illegal means and would have frozen Multiflex out of the market. Multiflex amply demonstrated Moore's "overt acts" in furtherance of the attempt to monopolize, and showed anticompetitive behavior from which a specific intent to monopolize can be inferred. The fact that Multiflex was able to overcome the hurdles Moore erected and establish a strong market presence should not allow Moore to escape liability, although market success could be expected to affect the quantum of a damage award. We find that Moore's market position in 1977 plus its overt anticompetitive acts indicate a specific intent to monopolize and created a dangerous probability that Moore would injure both Multiflex and the open, public market place in violation of section 2. This dangerous probability did not materialize because of the quality of Multiflex's product and the dedicated and imaginative way the company built a market for it.

Finding each of the necessary elements to support the jury's verdict, we affirm the district court's holding of a section 2 violation.

C. Antitrust Violation as the Cause of Damage.

Finding the likelihood of attempted monopolization does not end our analysis, however. Since this is a private sector suit for money damages, the plaintiff must also satisfy the requirements of section 4 of the Clayton Act. This means Multiflex must show by a preponderance of the evidence some actual damage to its own business or property that was proximately caused by the antitrust violation.

Although Multiflex in this case ended up the victor in this competitive war, this fact does not foreclose interim damages even though this victim of attempted monopolization ultimately benefitted from the monopolistic conduct. The attempt at monopolization could well have hindered or delayed the natural growth of the victim's market share. A competitor's lawful and successful response to otherwise illegal monopolization activities does not negate the possibility of short term damage. The plaintiff, of course, carries the burden of showing damage.

Multiflex relies on the lost Sedco job, discussed earlier in connection with the section 1 claim, as proof of section 2 damage. We find that this evidence is sufficient to establish injury for purposes of section 4 of the Clayton Act. There is substantial evidence to indicate that Moore mounted a vicious attack against Multiflex during the early days of Multiflex's existence, and that it was prepared to do "whatever it takes" to remain "number one" in the business. The Sedco affair failed as "fact of damage" under section 1 because Moore had not yet induced the OEMs to join in the conspiracy. The attempted monopolization by Moore, by contrast, did not require concerted action. Accepting the jury's verdict as revealing a belief that Multiflex would have earned the Sedco job in the absence of Moore's misconduct, we hold that Multiflex has met its burden of showing injury stemming from the section 2 violation.

To complete the burden necessary to recover under section 4, Multiflex need only show that its injury was proximately caused by the antitrust violation and a showing of the actual dollar amount of the damage. Multiflex must first establish a link between its damages and the values the antitrust laws were designed to remedy. Antitrust plaintiffs

must prove more than injury causally linked to an illegal presence in the market. Plaintiffs must prove injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful. The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation.

(1977) (emphasis in original). The question that must be addressed is whether Multiflex's proven specific instance of damage is related to the section 2 violation.

Moore argues that there can be no private damage award in this case under any circumstances, since Multiflex's market share increased so dramatically in so short a time. It believes that even if it were to concede anticompetitive acts, Multiflex's success in the market negates the requisite connection between conduct and injury, and conclusively offsets any possible recovery. We cannot accept such a sweeping argument. Moore should not be entitled to use Multiflex's successes to offset actual damage resulting from its own anticompetitive behavior. If Multiflex can prove that its injury was proximately caused by Moore's antitrust violation, we must allow recovery without offsetting this amount by the beneficial effects of Multiflex's strategic ingenuity.

Yet we cannot allow treble damages under the Clayton Act unless the injury was the type the antitrust laws were designed to protect. In this respect, charges of attempted monopolization differ from claims of actual monopolization. In the case of actual monopolization, there can be little doubt that the violation injures the marketplace and violates the public policies that the antitrust laws seek to protect. This helps explain, in part, why actual monopolization is an offense even without a showing of specific intent.

Attempted monopolization, on the other hand, does not invariably lead to the type of injury the antitrust laws seek to prevent. The attempt may fail to produce any actual antitrust injury to the marketplace. In this case, for example, the threatened injury to free and open competition never materialized except at most for a brief period. The market quickly beat a path to Multiflex's door in search of the fabled better mousetrap. The "dangerous probability of success" never progressed to success.

The seeming conflict is actually illusory, however. The offense of attempted monopolization need not cause actual market damage, but need merely threaten to produce the type of market damage contemplated in the antitrust laws. The relevant test is whether the "dangerous probability of success," if it had materialized, would have caused the type of market damage the antitrust laws seek to prevent. The mere attempt is the offense. (7th Cir. 1981), cert. denied, (1982).

Under such an inquiry, we have little trouble finding the necessary link between Multiflex's injury and the public policies behind the antitrust laws. Moore clearly was trying to maintain its market dominance and exclude any and all competitors. It stepped far beyond the bounds of reasonable competitive practices in trying to maintain its market position. The jury implicitly found that Moore had a specific intent to monopolize the market, a determination we find supported by substantial evidence. The Sedco incident is an example of the kind of sale Multiflex wanted to gain but that Moore was determined to win at all costs. Within the relevant market, Moore and Multiflex were the only substantial suppliers. It is therefore almost tautological to find that Moore's intent plus actions with an eye toward monopolization would injure Multiflex, its chief and perhaps only, competitor. Furthermore, it seems clear that these actions interfered with the ultimate customers' free choice in the marketplace and forced them into the position of following the well-trodden path to Moore's doorstep or hewing a treacherous, uphill path to Moore's only competitor. This is the stuff of which antitrust cases are made.

Armed with the findings of overt acts, injury, and proximate relation to the intent of the statutes, we affirm Multiflex's right to recover under section 4 of the Clayton Act for Moore's attempt to monopolize in violation of section 2 of the Sherman Act. The final issue in this case is the proper quantum of damages.

We have concluded that Multiflex may recover on its section 2 claims and not on its section 1 claims. Normally this would lead us to uphold the section 2 award made in a district court and nullify the section 1 award. The appellate court's power to review jury damage awards is narrowly limited. (5th Cir. 1981), cert. denied, (1982); (5th Cir. 1981). However, we cannot uphold the jury's section 2 award in this case and must instead remand for a redetermination of section 2 damages.

A. The Applicable Law.

Even after showing the "fact of damage" sufficient to establish liability under section 4 of the Clayton Act, Multiflex held the burden at trial of showing a rational quantum of actual damages stemming from the antitrust violations. This showing of the "amount of damage" enjoys a relaxed burden of proof as compared to the initial "fact of damage" proof requisite to establishing liability for damages under section 4 of the Clayton Act. The burden is relaxed, in part, because once a plaintiff has shown injury through anticompetitive behavior, we do not require that plaintiff prove with exact particularity what its success might have been in the market in the absence of the illegal conduct of another. (1981), after remand, (5th Cir. 1982), cert. denied, (1982); (1969); (1969); (1946); (5th Cir.), cert. denied, (1981); (5th Cir. 1977), cert. denied, (1978); (5th Cir. 1975), cert. denied, (1976); (5th Cir.), cert. denied, (1972). However, even under a relaxed burden of proof, there must be a rational basis by which a jury can assess the amount of antitrust damages or the award cannot stand. (5th Cir.), cert. denied, (1982); Household Goods Carriers' Bureau v. Terrell, 452 F.2d 152, 159 (5th Cir. 1971) (en banc), appeal after remand, (5th Cir. 1974), cert. dismissed, (1974). We examine two aspects of the award in this case: the damage model upon which the award was based, and the relationship between the verdict and the evidence.

B. The Damage Model.

In its pleadings, Multiflex asked for approximately $550,000.00 in damages growing out of both section 1 and section 2 antitrust claims, before trebling or attorney's fees. It based this amount on a model of lost profits developed for its use by an economic consultant. The damage model was the expert's attempt to indicate what the market shares for Moore and Multiflex would have been in the absence of anticompetitive acts. It looked to the few sales made directly to end users during that period, where the expert assumed little, if any, market interference had taken place. He found that Multiflex had just over 80% of that submarket. The economist used this figure as a "yardstick" to project an 80% market share over Multiflex's entire market. The economist's model, in short, assumed that Multiflex's market share should have been just over 80% for the entire period in question, rather than the actual rise from zero to 63% over two years. The difference between the actual market share and the 80%+ theoretical market share established the basis of the expert's damage model. Typical profit margins in each submarket then produced the expert's final damage figure, based on his "yardstick" model.

At trial, the expert's damage model was attacked on many grounds. First, Moore attempted to show that the model was based on lost gross profits, rather than on the proper measure of lost net profits. (3d Cir.), cert. denied, (1973). The expert agreed at trial with Moore's charge that his estimates of lost profits did not account for increases in marginal production costs, as a net profit model should. This made his model more analogous to an impermissible accounting of gross profits. However, the expert expressed his opinion that marginal costs of production in this case were so small as to be irrelevant, since the majority of the costs of production were fixed rather than incremental, and Multiflex was operating below its production capacity at the time in question. He showed uncontroverted evidence that the marginal labor and electricity costs for the lost production were minimal and that to include these costs in the model would have had little effect on the damages claim. Moore presented no countering evidence at all that the marginal costs of production would be any different than the expert's estimates. In any event, the jury was entitled to weigh and evaluate the expert's opinion. We find no error in the jury accepting the expert's cost estimates.

Similarly, the expert adjusted the damage estimates to create a net present value for the previous injury. Expressing a past damage figure in present terms required an estimation of the lost opportunity cost of the capital. The estimate normally equals the interest that might have been earned on the funds if placed in alternative investments. The expert in this case developed a discount figure based on the yield available on three-month Treasury bills on the dates in question to project the 1977-79 damages into terms of present value. Moore claims that these projections amount to nothing more than prejudgment interest, which normally would not be recoverable in this antitrust action. . Cf. (Supp. V, 1981) (prejudgment interest allowable under certain circumstances for actions commenced after September 12, 1980).

However, this Court in condemned prejudgment inflation of damage claims only when the adjustments amount to nothing more than bare interest. In for example, we affirmed a net present value adjustment to a claimed loss in capital. Ruling that the adjustment was for the lost opportunity cost from the capital formation rather than a bare claim to prejudgment interest, we affirmed that portion of the award. Similarly in the present case, we find that the claims were made to reflect what the asserted lost profits would have earned in an open investment market, not as a statutory claim for prejudgment interest. Since the claim was based on an economic theory and not in a claim to statutory or common law prejudgment interest, we find no error in this portion of the award.

Moore also contests the method by which the damage model classified sales in the relevant market and time frame as foreign or domestic. The model developed by Multiflex's expert looked to the domestic or foreign location of the buyer's purchasing agent, rather than to the location of actual installation. Moore argues that a sale to an American company for installation abroad should not enter into the damage award that is concerned only with the U.S. market. The relevant market, however, is defined as sales in the U.S. market. A sale is made when a contract is formed. A contract is formed when a bid is accepted. See 1 A. Corbin, Corbin on Contracts Secs. 55-56 (1963 ed.). We find it logical to assume that the contract bids were accepted by the buyers at their mailing addresses rather than at the location of their drilling sites. Under this view, contracts formed in the U.S. are part of the U.S. market, and those established elsewhere are not. We find no error in this aspect of the expert's analysis.

C. Support in the Evidence.

The damage model, although reasonable in other respects, does contain a fatal flaw as applied to the jury's verdict. The damages claimed were not split into components of section 1 and section 2 damage. Rather, the model projected only one total figure for the Sherman Act damages alleged under both section 1 and section 2 theories. There is no clear error in such calculations--section 1 and 2 theories often overlap in a case, (1946), and similar facts may not only establish both violations but may result in similar or identical injury. However, the plaintiff's failure to distinguish between its section 1 and 2 injuries is relevant in this case because the jury split its verdict into an award for the section 2 damage, which we recognize, and the section 1 damage, which we reverse. Since the legal underpinnings of the jury's award do not both survive this appellate review, we must remand for retrial of the quantum of damages, based on the factors explained below.

The reason we must reverse the jury's award of section 2 damages is that there is no rational basis in the record for distinguishing between the section 1 and section 2 damage award. Nowhere in the record does either party suggest what damages might be appropriate if one theory and not the other were valid in this case. The jury's bifurcation of the award into $382,963.07 for section 1 damages and $205,964.73 for section 2 is completely unsupported in the evidence and does not represent the "substantial, credible evidence" that is a prerequisite for our upholding a jury verdict. (1930); (5th Cir. 1978); Boeing Co. v. Shipman, 11 F.2d 365 (5th Cir. 1969) (en banc). Since we are reversing the section 1 component that served as partial support for the jury's award, we must nullify the jury's damage verdict and remand for a new trial on the damages growing out of the section 2 violation alone.

Moore raises an additional point for us to consider in examining the jury's damage award. After the jury returned its verdict, the district judge obtained the affidavits of the jury foreman and another juror. These apparently were taken because the total award was greater than the plaintiff's proof at trial. These affidavits suggest that the jury verdict might have included $200,000 to compensate Multiflex for "injury to its reputation" or to "punish" Moore for its conduct. One juror also stated that the $200,000 figure was reached by taking an average of each juror's predilection regarding such an award--a "quotient verdict." Neither theory is a proper basis for recovery in a Sherman Act claim, and Moore now asks us to overturn the jury award on the basis of jury misconduct.

Punitive damages are impermissible in an antitrust award. The treble damage provisions accomplish, in part, a similar purpose. However, we are also cognizant of the need to respect the finality of a jury verdict. In general, it is unwise to leave open to doubt the propriety or finality of a jury's decision if there is any credible basis for supporting its verdict. Rule 606(b) of the Federal Rules of Evidence discusses the limited permissible inquiry into the validity of a verdict. An affidavit discussing the jury's decision-making process is cognizable only to discover whether "extraneous prejudicial information was improperly brought to the jury's attention or whether any outside influence was improperly brought to bear upon any juror." Fed.R.Evid. 606(b). The court, therefore, properly could not consider Moore's charges that the jury awarded impermissible punitive damages or arrived at a "quotient verdict." (5th Cir. 1983); (5th Cir. 1977).

The district judge in this case examined the affidavits for the single purpose of determining whether the award was supported in the evidence. He found that the award was larger than the damages shown at trial, at which point he properly offered Moore the option of remittitur in the amount of the unsupportable excess., 582 & n. 14 (5th Cir. 1981); (5th Cir. 1980); (5th Cir. 1974). The remittitur, accepted by the plaintiff Multiflex, properly reduced the award only to the extent necessary to conform to the evidence. (5th Cir. 1970), modified, (5th Cir. 1972), cert. denied, (1972). The district court properly handled the questions raised by the excessive amount of the verdict. Therefore, we refuse to consider Moore's arguments regarding the foundation of the verdict. However, as discussed earlier, we find the verdict unsupportable in its breakdown of damages into the section 1 and section 2 components. We still must remand, therefore, for readjudication of the damage award.

Since the jury was never instructed on the delineation between the section 1 damage and the section 2 damage when it rendered its verdict, we cannot allow the section 2 damage award to stand. It is clear that the jury intended to compensate Multiflex for the profits it lost in the years 1977, 1978, and 1979 when it was fighting its uphill battle against Moore for its rightful market share. Yet, it would be pure speculation to calculate what the jury might have awarded if presented only with the meritorious section 2 theory.

In remanding for calculation of quantum, however, we caution the district court to observe the distinction between the fact of damage growing out of the Sedco incident, which already has been determined by a preponderance of the evidence, and the amount of damage which must be calculated on remand on the basis of credible evidence presented to the jury. Since the fact of damage already has been shown with a reasonable certainty, the damages claimed on remand may be shown with a somewhat relaxed standard in this bifurcated procedure. (7th Cir. 1983); (5th Cir. 1978); (5th Cir. 1976); (5th Cir. 1974), cert. dismissed, (1974).

In summary, we reverse the jury's finding of a conspiracy in restraint of trade, violative of section 1 of the Sherman Act, ; we affirm the finding of attempted monopolization, . We reverse and remand for a new trial on the issue of damages growing out of the Section 2 violation, in accordance with this opinion.

AFFIRMED in part; REVERSED in part; REMANDED.

District Judge of the Northern District of Illinois, sitting by designation.

. Shortly before trial, Samuel Moore & Co. was acquired by the Eaton Corporation and ceased to exist as a separate corporate entity. However, for clarity's sake, the district court referred to the defendant as Samuel Moore & Co., a convention we adopt in this opinion as well. Eaton Corp. has been properly named a party to this proceeding.

. The jury awarded damages of $382,963.07 under the conspiracy claim and $205,964.73 under the attempted monopolization claim, for a total of $588,927.80 before trebling. However, before the entry of judgment Multiflex filed its consent to reduce damages under the verdict to $549,362.98. The district court entered judgment for this reduced figure on September 4, 1981, five days after the jury's verdict. Damages were then trebled to $1,648,088.94. The damages figure is discussed further in Part IV of this opinion.

. U.S. Patent No. 3,905,398.

. The parties dispute Moore's exact share of the "relevant market," as that market is defined in this case. Multiflex asserts a virtual 100% stranglehold on the market until 1978, while Moore admits to "at least an 80-85% market share." Either figure, however, is sufficient to establish market dominance for the purposes of this case.

. The OEMs, Multiflex's best customer prospects, are not named in this suit. It is easy to see how Multiflex would hinder its long-term success in the marketplace if it sued its intended customers. "The antitrust laws do not require a business to cut its own throat." (1st Cir. 1961). However, a conspiracy charge need not be filed against all co-conspirators to be valid. A conspiracy conviction may lie against a single indicted party, even though the legal definition of conspiracy requires two or more persons acting in concert. (5th Cir. 1982); (5th Cir. 1981), cert. denied, (1982) (conspiracy defendant may be convicted of conspiring with unindicted parties if the evidence supports their existence and the existence of a conspiracy). (5th Cir.), cert. denied, (1961) (conspiracy conviction may not lie against only one of several indicted parties in a combined trial); (5th Cir. 1983) (conspiracy conviction may lie against one of several indicted parties in severed trials).

. While the section 1 rule of reason analysis requires an impact on competition, section 4 of the Clayton Act requires evidence of private injury to support a private sector recovery. A private plaintiff pursuing a restraint of trade case therefore must prove both marketplace and private injury at some point in the case. Yet such evidence may overlap, since evidence of injury to a competitor may suggest damage to the competitive marketplace.

. In, 317 & n. 17 (5th Cir. 1978), we explained the distinction between a Sherman Act violation and private recovery:

Proof of a violation of the Sherman Act standing alone does not establish civil liability under Sec. 4 of the Clayton Act. There must under Sec. 4 be proof of "injury to business or property" before a Sherman Act violation becomes cognizable as a private civil remedy.... [A]wareness of the distinction between conduct which violates Sec. 1 of the Sherman Act, and the elements which establish liability in private party litigation under Sec. 4 of the Clayton Act, is vital.

An example of the importance of this distinction would be a conspiracy to establish maximum prices. This form of price-fixing would clearly violate the Sherman Act, (1951), but would not be a cognizable action under Sec. 4 on behalf of retail customers because they most likely would not have suffered injury as a result of such acts.

(Emphasis in original.)

. The "fact of damage" is an element of the plaintiff's burden of proof, which requires a showing by a preponderance of the evidence. Once the fact of damage is shown, a jury may use its discretion in determining the exact amount of damage resulting from the antitrust violation. By separating the fact of damage determination from the amount of damage computation, an antitrust plaintiff benefits from a "relaxed standard" of showing quantum once the fact of damage is proven. (1946). Application of this rather unique damage scenario under the antitrust laws is described also in our discussion of the claimed section 2 violation (parts III, C, and IV, infra) .

. If a conspiracy had been in effect at the time of the Sedco bidding, parties later entering into that conspiracy properly could be charged with knowledge of, hence liability for, earlier acts of the coconspirators. (5th Cir. 1982). However, no one else was conspiring with Moore to keep Multiflex from the market in October, 1977. One party does not a conspiracy make. (5th Cir.), cert. denied, (1961). Hence, we cannot allow damages under section 1 for events occurring before the instigation of the conspiracy, namely January, 1978.

. We therefore need not rule conclusively on the theoretical question of whether section 1 was violated in the abstract. The evidence, regardless of merit, was not presented by a party deserving of the relief sought.

. While this requirement is closely comparable to the "fact of damage" requirement discussed under Section 1, the courts typically do not use "fact of damage" as a phrase of art under section 2.

. Claims of attempted monopolization under Section 2 must be presented within the confines of a well-defined product market. (5th Cir. 1975), cert. denied, (1976). The definition must set forth both the products in question and the geographic limits of the relevant market. The jury found the relevant market to be hydraulic hose bundles in the United States. It excluded sales of hard metal pipe or other products that might, under certain limited circumstances, substitute for hydraulic hose bundles. Similarly, the jury excluded the United Kingdom market and other foreign markets for offshore drilling equipment.

Moore argues on appeal that the relevant market should be the worldwide market for BOP devices. Under this definition, the jury would have had to consider markets such as the United Kingdom, where other manufacturers hold significant market share and where Moore's sales are minimal, in assessing liability. A worldwide definition of the relevant market would make it more difficult for a jury to find that Moore possessed market power.

The record supports the view that hydraulic hose bundles have no practical substitute in the BOP market, and that the demands of the market and of foreign governments in overseas markets makes the United States market a reasonable definition of "relevant market" or "relevant submarket" for the purpose of this case. We therefore find no error in the jury determination of relevant market.

. Professor Areeda described the connection between public injury and private recovery in Areeda, Antitrust Violations Without Damage Recoveries, 89 Harv. L. Rev. 1127, 1127 (1976):

Money awards under the antitrust laws do not serve only to compensate injured plaintiffs. Treble damages also punish wrongdoers and enlist private plaintiffs in the work of detecting, punishing, and thereby deterring wrongdoing. But the desire to encourage private enforcement and to penalize antitrust violations is no excuse for awarding damages that are non-existent, inconsistent with antitrust policy, or unconnected with the true rationale for imposing antitrust liability.

(Citation omitted.)

. The testimony in the record suggests a long list of Moore's anticompetitive practices. In addition to the allegation discussed earlier that Moore told customers Multiflex was insolvent and had Pinkerton guards posted at its gates, there is also evidence that Moore disparaged the quality of the Multiflex product, overstated the pressure ratings of its own competing products, and advised potential investors that the company had no customers. Multiflex has dropped its earlier claim that the patent infringement claims that began this litigation were themselves part of an anticompetitive scheme to retain market dominance.

. In Multiflex's single reference to the breakdown of the damage claim, made during closing argument, its attorney told the jury:

Finally, I want to talk to you a minute about one area that may be a slight problem for you on the question of damages.

Necessarily, you are going to have to look at the question on how much damages was caused by the conspiracy, how much damages was caused by the attempt to monopolize.

Of course, Multiflex lost their sales as a result of not being able to sell through the three OEM's. This was kind of a mixture of things that happened.

So, you, as the jury, are in the position of having to use your discretion in how to answer those two damage questions.

Certainly Multiflex is not claiming any more than that as their total damages in this case.

Record on Appeal, Vol. 14, at 48.

. The draft of Fed.R.Evid. 606(b) as passed by the House of Representatives would have permitted juror testimony about objective matters occurring during the jury's deliberation, such as the reaching of a quotient verdict, see H.Rep. 93-650, U.S.Code Cong. & Admin.News 1974, p. 7051, but the version passed by the Senate, see S.Rep. No. 93-1277, and approved by the Joint Conference Committee expressly denounces such a practice, H.Rep. 93-1597. This makes clear the drafters' intent that a jury's verdict should not be open to searching the minds and motives of the jurors, and must be affirmed if there is adequate support for the award.