Liggett Group, Inc., charges Brown & Williamson Tobacco Corporation with pursuing a primary-line predatory pricing scheme in the sale of generic cigarettes during the period 1984-85 in violation of § 2(a) of the Robinson-Patman Act, 15 U.S.C. § 13(a). Liggett contends that Brown & Williamson charged below-average-variable-cost prices1 to force Liggett either to raise the prices of its generic cigarettes or to cease selling them, with the expectation of preserving high profits theretofore earned on sales of branded cigarettes by the industry-wide oligopoly.2 Following a 115-day trial, a jury returned a verdict in favor of Liggett in the amount of $49.6 million which the district court trebled for a judgment of $148.8 million. The district court, however, granted Brown & Williamson's motion for judgment notwithstanding the verdict, 748 F. Supp. 344 (M.D.N.C.1990), and this appeal followed. We now affirm.

* Cigarettes in the United States have been manufactured in recent years primarily by six companies, Philip Morris, Inc., R.J. Reynolds Tobacco Corporation, Brown & Williamson, Lorillard, Inc., American Tobacco Company, and Liggett. Philip Morris and R.J. Reynolds, with respectively over 40% and 28% of the market at the time of trial, held the two largest market shares throughout the 1980's. During the relevant period Brown & Williamson's sales never represented more than 12% of the market. Prior to 1980, cigarettes were sold by these companies to distributors at the same price, and when one company increased the price, the others followed. Liggett has characterized this market as "one of the most highly concentrated oligopolies in the United States," which has produced what Liggett's economist has characterized as "supracompetitive profits."3

In 1980 when Liggett's share of the cigarette market in the United States had declined to 2.3%, a level that threatened its viability, it introduced a line of generic cigarettes in black and white packaging. Liggett discounted these generic cigarettes and offered volume rebates to yield an effective price to distributors about 30% lower than that charged for branded cigarettes. By 1984 Liggett's share of the total cigarette market in the United States increased to over 5% and its generics became the fastest growing segment of the market, in which overall sales of cigarettes were declining.

Other manufacturers began responding in 1983. In July of that year, R.J. Reynolds introduced "Century" cigarettes. It sold this brand in cartons of nine packs containing 25 cigarettes each, for the same price as other branded cigarettes sold in cartons of 10 packs containing 20 cigarettes each. These "25s" thus cost consumers 12.5% less than other brands. Brown & Williamson followed suit later that year with "Richland," its own brand of 25s.

Nonetheless, in the 1983-84 period, Brown & Williamson began to observe that it was losing more sales to generic cigarettes, proportionally, than were other manufacturers, and that it sustained a "variable margin loss"4 of over $50 million in 1983 alone. It concluded that "unchallenged, [Liggett] could continue its total dominance of this segment ..., becoming the third largest company in the U.S. cigarette market." A Brown & Williamson memorandum introduced at trial stated about the Liggett activity:

Stipulating that the industry's interests--other than [Liggett's]--would be far better served had generics never been introduced, they are an immediate and growing threat to all other manufacturers. Competitive counter-actions are essential and inevitable.

After examining each company's capacity and expected response to Liggett's introduction of generic cigarettes, Brown & Williamson determined to enter the generic segment to recapture the sales lost to Liggett. It described the opportunity as follows:

Generics represent B & W's most immediate opportunity to increase volume. This volume can be achieved within current manufacturing capacity, without incremental manpower and without negatively impacting trading profit. No other option offers similar potential to recover lost volume/share with such minimal investment risk. This is true because our goal is to capture existing demand[,] not create new consumer demand.

The same marketing memorandum revealed a pricing strategy that was intended to produce "a full variable margin" in excess of $4 per thousand cigarettes. But it went on to state,

B & W is prepared to spend up to net variable margin as a first step response to competitive counter-offers; if required, we are also prepared to go up to, but not beyond, full variable margin to gain entry into the generics market.

* * * * * *

If [Liggett] goes below full variable margin, Brown & Williamson would not plan to match their offer.

In May 1984, before Brown & Williamson implemented its plan, R.J. Reynolds "repositioned" its brand "Doral," cutting the list price to that charged by Liggett for its generic cigarettes. Brown & Williamson reanalyzed its strategy in light of the Doral move, concluding in the final strategic memorandum,

B & W believes that branded generics will enhance the growth of the economy segment and will draw volume from popular priced brands.

* * * * * *

The earlier concern of expanding the economy segment is no longer tenable, given RJR's recent action. It is clear that the economy segment is significant, and growing. Accordingly, recognizing the importance of minimizing increased cannibalization and concomitant share erosion, as well as maintaining trading profit targets, it is imperative that B & W enter this segment.

Shortly thereafter, in July 1984, Brown & Williamson introduced its line of generic cigarettes in black and white packages to compete directly with Liggett's black and white packaged generics. Liggett responded immediately with this lawsuit alleging initially that Brown & Williamson violated the trademark laws. The complaint was amended to add the Robinson-Patman claims that are the subject of this appeal. Liggett also responded by increasing rebates and other incentives to its distributors. During the rest of the summer, the two manufacturers traded moves and counter-moves four more times in setting prices, offering rebates, and otherwise promoting black and white generic cigarettes. The incentive schemes established when the dust settled in August remained in force until June 1985, when Liggett raised its list price for generics. Brown & Williamson matched the rise in October of that year.

By 1988 all other market participants, except for Lorillard, were selling generics and discounted branded cigarettes, and both R.J. Reynolds and Philip Morris added a black and white generic line. By the time of trial, Lorillard too was in that segment of the market. With the exception of an aborted attempt by Brown & Williamson in December of 1985, no manufacturer raised the price of black and white cigarettes until the summer of 1986. Since then, generic cigarette prices have risen twice a year, in tandem with those of branded cigarettes, reducing the proportional price discount between branded and generic cigarettes from a high of 40% in 1985 to 27% in 1989.

While the United States market for cigarettes has been generally declining, the growth of discounted cigarettes has been dramatic. In 1981 Liggett, which held 97% of the generic sales, sold 2.8 billion cigarettes, representing .4% of the United States cigarette market. By 1988 it sold over 9 billion generic cigarettes. The total sales of generic and discounted cigarettes during the same period increased from 2.8 billion cigarettes to 61.6 billion, representing 11.1% of the entire United States cigarette market. By trial all manufacturers were selling generics and discounted cigarettes, and yearly sales had reached nearly 80 billion cigarettes, representing 15% of the United States cigarette market. The percentage of the market represented strictly by black and white generics peaked in 1985 with 4.7% and was continuing to decline at the time of trial.

No one denies that after Brown & Williamson introduced its black and white generic cigarettes in 1984, the companies fought "tooth and nail" and that the market "got very competitive." Liggett's president confirmed that "competition had substantially increased in the total cigarette market." When all was said and done, Liggett ended up with 3.25% of the market for cigarettes in the United States and Brown & Williamson with 11.36%.

Liggett contends that from June 1984 to the end of 1985, Brown & Williamson sold black and white cigarettes at prices below its "average variable cost," although it acknowledges that Brown & Williamson realized profits in the overall sale of cigarettes in the United States, the agreed upon relevant market. Its Robinson-Patman Act claim focuses on Brown & Williamson's pricing activity for that period, which it contends was predatory, designed to force Liggett to raise its prices for generic cigarettes and thereby to destroy or limit the generic cigarette segment.

After a 115-day trial, the jury returned a verdict in favor of Liggett on its Robinson-Patman Act claim in the amount of $49.6 million, which the district court trebled under 15 U.S.C. § 15 for a judgment of $148.8 million. On Brown & Williamson's motion for judgment notwithstanding the verdict, the district court, after thoroughly analyzing the wide range of issues presented, set aside the judgment and entered judgment in favor of Brown & Williamson.

Section 2(a) of the Robinson-Patman Act makes it unlawful for a person engaged in commerce "to discriminate in price between different purchasers of commodities of like grade and quality ... where the effect of such discrimination may be substantially to lessen competition ... or to injure, destroy, or prevent competition" with the person charging the discriminatory prices. 15 U.S.C. § 13(a). The requirement to show that the effect of pricing "may be substantially to lessen competition" may be satisfied by proof of predatory pricing.5

Liggett contends that this case is unusual because Brown & Williamson's intent to destroy competition from Liggett's sales of generic cigarettes is so clearly stated in corporate memoranda that, when coupled with evidence that its pricing plan in fact caused injury to Liggett, liability is established. It relies on Utah Pie Co. v. Continental Baking Co., 386 U.S. 685, 87 S. Ct. 1326, 18 L. Ed. 2d 406 (1967).

While Utah Pie is difficult to understand in light of recent economic theory and has been the subject of some criticism over the years, see, e.g., A.A. Poultry Farms, Inc. v. Rose Acre Farms, Inc., 881 F.2d 1396, 1404 (7th Cir.1989), cert. denied, 494 U.S. 1019, 110 S. Ct. 1326, 108 L. Ed. 2d 501 (1990), we are confident that it does not require reversal in this case. In Utah Pie, national competitors, using economic muscle from sales in markets other than Salt Lake City, had subsidized below-cost pricing in the Salt Lake City area. The Supreme Court concluded, contrary to the court of appeals, that the plaintiff had adduced sufficient evidence for a jury to infer the requisite possibility of injury to competition, observing that "there was some evidence of predatory intent with respect to each [defendant]" and of a "drastically declining price structure." 386 U.S. at 702-03, 87 S.Ct. at 1336. The Court did not, however, purport to discuss in detail what evidence was sufficient to infer predatory intent, noting only, "Although the evidence in this regard against [one defendant who used industrial espionage against the plaintiff] seems obvious, a jury would be free to ascertain a seller's intent from surrounding economic circumstances, which would include persistent unprofitable sales below cost and drastic price cuts themselves discriminatory." Id. at 696 n. 12, 87 S. Ct. at 1332 n. 12. We understand Utah Pie to have left for later cases a statement of what more precisely constitutes predatory pricing.

Since Utah Pie, the Supreme Court has confirmed that "a firm cannot claim antitrust injury from nonpredatory price competition on the asserted ground that it is 'ruinous.' " Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 338 n. 7, 110 S. Ct. 1884, 1891 n. 7, 109 L. Ed. 2d 333 (1990). "[N]onpredatory price competition for increased market share, as reflected by prices that are below 'market price' or even below the costs of a firm's rivals, 'is not activity forbidden by the antitrust laws.' " Id. at 340, 110 S.Ct. at 1892 (quoting Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 116, 107 S. Ct. 484, 492, 93 L. Ed. 2d 427 (1986)). However, the line between legitimate competitive pricing, encouraged by the policy of free market competition, and predatory pricing, that is destructive of competition, is murky, and agreement on the distinction has not been reached. See Cargill, 479 U.S. at 117 n. 12, 107 S. Ct. at 493 n. 12. While pricing "below some appropriate measure of cost" must be shown to establish it as predatory, see Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 585 n. 8, 106 S. Ct. 1348, 1355 n. 8, 89 L. Ed. 2d 538 (1986), questions remain about the definition of "below-cost" pricing. See Atlantic Richfield, 495 U.S. at 341 n. 10, 110 S. Ct. at 1893 n. 10. Cargill, 479 U.S. at 117 & n. 12, 107 S. Ct. at 493 & n. 12. It is sufficient for this case, however, to observe that predatory pricing must involve, in addition to some level of below-cost pricing that is harmful to competition, the rational expectation of later realizing monopoly profits. The failure to show this additional aspect is fatal. Thus the Court in Matsushita held that a conspiracy, which could not hope to recoup its expenses incurred from alleged below-cost pricing and was therefore economically senseless, did not violate the antitrust laws. See Matsushita, 475 U.S. at 597-98, 106 S.Ct. at 1361-62. The Court explained,

A predatory pricing conspiracy is by nature speculative. Any agreement to price below the competitive level requires the conspirators to forego profits that free competition would offer them. The foregone profits may be considered an investment in the future. For the investment to be rational, the conspirators must have reasonable expectation of recovering, in the form of later monopoly profits, more than the losses suffered.... The success of any predatory scheme depends on maintaining monopoly power for long enough both to recoup the predator's losses and to harvest some additional gain.

* * * * * *

These observations apply even to predatory pricing by a single firm seeking monopoly power.

Id. at 588-89, 590, 106 S. Ct. at 1356-57, 1357-58 (emphasis changed). See also Areeda and Turner, supra note 1, at 698 ("[P]redation in any meaningful sense cannot exist unless there is a temporary sacrifice of net revenues in the expectation of greater future gains.").

Turning to this case, Liggett has not advanced a theory that Brown & Williamson could ever obtain or maintain monopoly power for any period of time, much less a period sufficient to reap the harvest of its alleged below-cost pricing, since Brown & Williamson never had more than 12% of the agreed-upon market. Liggett's theory rests on the notion that following the fight in which Brown & Williamson would effectively contain or destroy the generic segment, Brown & Williamson and the other four manufacturers with successful branded cigarettes would, as an oligopoly, reap profits which Liggett predicts would be restored to a "supracompetitive" level.

On the basis of the testimony of its economic expert, William Burnett, Liggett contends that for years the cigarette market in the United States had been an oligopoly. Pointing to historical price increases, it observes that the six members raised prices in lockstep regardless of the amount of increases or even decreases in the cost of tobacco, and without the entry of any significant new competitor in several decades. When Liggett, in 1980, began offering generic cigarettes at prices 30% below those of branded cigarettes, purchasers switched from branded cigarettes to the generic cigarettes offered by Liggett, thereby denying to members of the oligopoly the assertedly supracompetitive prices they had previously enjoyed. Liggett argues that in response to Liggett's success Brown & Williamson developed and implemented a detailed, two-pronged counterattack to contain expansion of or destroy the generic segment. As outlined by Burnett, on the one hand Brown & Williamson removed any disincentive for distributors of generic cigarettes to switch entirely from Liggett to Brown & Williamson by offering their generic cigarettes in a black and white package which consumers would confuse with Liggett's. On the other hand, to lure these distributors away, Brown & Williamson directed discriminatory rebates at Liggett's highest-volume customers, structured to encourage them to pocket the rebates rather than pass them on as price cuts to consumers. Because consumers would not see two competing generic cigarettes or lower prices, consumer demand for generics would not increase and thereby cut further into sales of branded cigarettes. The higher cost of access to its distributors, however, would force Liggett either to raise its overall income by raising list prices to consumers, or to cease selling generic black and white cigarettes entirely.

In either case, asserted Burnett, Brown & Williamson would recoup its losses and make additional gains from the increased sales of its branded cigarettes at the supracompetitive prices established by the historical parallel pricing in the oligopolistic market. According to Burnett, the other members of the oligopoly would not intercede, because they would similarly benefit from decreased "cannibalization" of the sales of their own branded cigarettes.

Shorn of its details, Liggett's theory depicts Brown & Williamson as an oligopolist attempting to discipline another oligopolist for breaking a pattern of parallel pricing, by driving up the offender's costs using finely-tuned predatory pricing of its own. The theory relies upon an expectation that all the other oligopolists would, out of their self-interests, simply stand by and refrain from also selling generics or other low-cost products which would eat into sales of branded cigarettes. Absent such assurance, Brown & Williamson might well drive Liggett out or its prices up, but continue to lose sales of its more profitable branded cigarettes to cannibalistic sales of generics by the other oligopolists. There is no evidence, however, of any conspiratorial agreement among the oligopolists to stay their hands. Liggett's theory therefore amounts to substituting the conscious parallelism of an oligopoly for conspiratorial agreement or actual monopoly power as the reason Brown & Williamson might rationally expect to be able to recoup its investment in disciplining Liggett.

We are aware of no case in which the predicted economic behavior of an oligopoly was relied on to provide a rational means of recoupment of the losses sustained in a predatory pricing scheme, and economic logic as well as actual experience in this case belie such a holding. Oligopolists might indeed all share an interest in letting one among them discipline another for breaking step and might all be aware that all share this interest. One would conclude, however, that this shared interest would not itself be enough to convince a rational oligopolist facing losses of market share to a competitor's price-cut not to match the cut with its own grab for market share. The oligopolist on the sidelines would need to be certain at least that it could trust the discipliner not to expand the low-price segment itself during the fight or after its success. Of course, all the oligopolists on the sidelines would need to be certain that the others were also confident on this point. Such confidence must be rare, indeed, when the form that the discipline takes is a price-war, which must strike fear in the heart of any oligopolist hoping to protect market share and high prices. More likely, when members of an oligopoly are faced with a competitor's decision to break step, drop prices, and expand market share, they would react competitively.

The facts in this case remove any doubt that no rational oligopolist would be confident that neither Brown & Williamson nor any of the other manufacturers would expand its own sales of low-priced cigarettes at the expense of full-priced branded cigarettes. Brown & Williamson may have intended its disciplinary actions not to affect consumer prices, but outside observers might well not have recognized this plan in the complex and furious rebate war which ensued in the summer of 1984. More importantly, however, Brown & Williamson had already expanded its sales of low-priced cigarettes, when it introduced in 1983 its brand "Richland" at an effective discount of 12.5% to consumers, and R.J. Reynolds, of course, had done the same, first with "Century" and later when it cut the price of its brand "Doral" to the same level as Liggett's generic black and white cigarettes.

Any rational observer would have known that sales of Richland, Century, and Doral would most likely further erode the sales of fullpriced branded cigarettes, regardless of Brown & Williamson's success in disciplining Liggett for introducing generic black and white cigarettes. But we need not merely impute rationality to Brown & Williamson. The very memoranda upon which Liggett relies so heavily as evidence of Brown & Williamson's predatory intent not only predict similar actions as occurred, but conclude after the introduction of Doral that "[t]he earlier concern of expanding the economy segment is no longer tenable...."

The perfect vision of hindsight confirms Brown & Williamson's conclusion and our theoretical suspicions. Soon after the events of 1984, most cigarette manufacturers were offering various types of low-priced cigarettes, including generics, and by trial all were vigorously competing with differing devices and approaches. Sales of lowpriced cigarettes increased from 2.8 billion cigarettes in 1981 to nearly 80 billion in 1989. Their proportional share of the overall cigarette market in the United States grew from .4% to 15%.

In the end, Liggett asked the jury to leap from the fact that the cigarette market is a concentrated oligopoly with a history of parallel pricing to the conclusion that the oligopoly would act uncompetitively when one of its members made a competitive move, suggesting some perniciousness in the oligopoly itself. Yet, in the absence of an agreement among the oligopolists, which nobody contends is the fact in this case, membership alone in an oligopoly provides no basis for proof of illegal conduct. Thus no case suggests that mere participation in an oligopolistic market constitutes conduct illegal under the antitrust laws. As the court stated in E.I. duPont de Nemours & Co. v. Federal Trade Commission, 729 F.2d 128, 139 (2d Cir.1984):

The mere existence of an oligopolistic market structure in which a small group of manufacturers engage in consciously parallel pricing of an identical product does not violate the antitrust laws. Theater Enterprises, Inc. v. Paramount Film Distributing Corp., [346 U.S. 537, 74 S. Ct. 257, 98 L. Ed. 273 (1954) ]. It represents a condition, not a "method;" indeed it could be consistent with intense competition.

While the parallel but not agreed upon conduct of an oligopoly may be a characteristic of a mature market, as we noted above, such parallelism cannot rationally be assured when an act of competition is undertaken by one participating member. To rely on the characteristics of an oligopoly to assure recoupment of losses from a predatory pricing scheme after one oligopolist has made a competitive move is thus economically irrational. As the Supreme Court pointed out six years ago, " 'the predator must make a substantial investment with no assurance that it will pay off.' For this reason, there is consensus among commentators that predatory pricing schemes are rarely tried, and even more rarely successful." Matsushita, 475 U.S. at 589, 106 S.Ct. at 1357 (quoting Easterbrook, Predatory Strategies and Counterstrategies, 48 U.Chi.L.Rev. 263, 268 (1981)).

In short, Brown & Williamson controlled only 12% of the relevant market and could not be assured, when it began its alleged below-cost pricing to suppress competition from Liggett, that the other manufacturers would not also respond competitively. Consequently, the pricing policies undertaken by Brown & Williamson, while perhaps intended to injure Liggett, could not be found to be predatory because they did not provide an economically rational basis" to recoup ... losses and to harvest some additional gain." Matsushita, 475 U.S. at 589, 106 S.Ct. at 1357. Because we conclude that Liggett is unable to demonstrate that it satisfied an essential element of proving a primary-line pricing scheme in violation of the Robinson-Patman Act, we need not reach the other questions decided by the district court. The judgment of the district court is affirmed.

AFFIRMED.